0% found this document useful (0 votes)
2K views385 pages

Davianna Pomaika'i Mcgregor - Na Kua'Aina - Living Hawaiian Culture (2007)

Davianna Pomaika'i Mcgregor - Na Kua'Aina_ Living Hawaiian Culture (2007)

Uploaded by

Salamanca
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
2K views385 pages

Davianna Pomaika'i Mcgregor - Na Kua'Aina - Living Hawaiian Culture (2007)

Davianna Pomaika'i Mcgregor - Na Kua'Aina_ Living Hawaiian Culture (2007)

Uploaded by

Salamanca
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 385

back flap = 3.5 in. .375 in. back = 6.25 in spine = .875 in. front = 6.

back = 6.25 in spine = .875 in. front = 6.25 in front flap = 3.5 in.
.375 in.

hawaiian studies

McGregor
(Continued from front flap)

case study begins by examining the cultural


significance of the area. The ‘òlelo no‘eau
(descriptive proverbs and poetical sayings)
Of related interest

Leaving Paradise
T he word kua‘âina translates literally as
“back land” or “back country.” Davi-
anna Pòmaika‘i McGregor grew up hearing
for which it is famous are interpreted, offer- Indigenous Hawaiians in the Pacific Northwest, it as a reference to an awkward or unsophis-
ing valuable insights into the place and its ticated person from the country. However,
1787–1898
overall role in the cultural prac­tices of Na- in the context of the Native Hawaiian cul-
tive Hawaiians. Discussion of the landscape Jean Barman and Bruce McIntyre Watson tural renaissance of the late twentieth cen-

Nā Kua‘āina
and its settlement, the deities who dwelt tury, kua‘âina came to refer to those who
there, and its rulers is followed by a review

N
actively lived Hawaiian culture and kept the
ative hawaiians arrived in the Pacific Northwest
of the effects of westernization on kua‘âina spirit of the land alive. Kua‘âina are Native
as early as 1787. Some went out of curiosity; many
in the nineteenth century. McGregor then Hawaiians who remained in rural areas; took
pro­vides an overview of the social and eco- others were recruited as seamen or as workers in the fur
care of kûpuna (elders); continued to speak
nomic changes in each area through the trade. By the end of the nineteenth century more than a Hawaiian; toiled in taro patches and sweet
end of the twentieth century and of the ele- thousand men and women had journeyed across the Pacific, potato fields; and took that which is pre-
ments of continuity still evident in the lives but the stories of these extraordinary individuals have cious and sacred in Native Hawaiian culture
9.5 in.

of kua‘âina. The final chapter on Kaho‘olawe gone large­ly unrecorded in Hawaiian or Western sources. into their care. The mo‘olelo (oral traditions)
demonstrates how kua‘âina from the cultural Through painstaking archival work in British Columbia, recounted in this book reveal how kua‘âina
kîpuka under study have been instrumental Oregon, California, and Hawai‘i, Jean Barman and Bruce have enabled Native Hawaiians to endure as a
in restoring the natural and cultural resources unique and dignified people after more than
Watson pieced together what is known about these sail­-
of the island. a century of American subjugation and con-
ors, laborers, and settlers from 1787 to 1898, the year the
e Hawaiian Islands were annexed to the United States. In ad-
trol. The stories are set in rural communities
or cultural kîpuka — oases from which tradi-
Unlike many works of Hawaiian history, dition, the authors include descriptive biographical entries
tional Native Hawaiian culture can be regen-
which focus on the history of change in on some eight hundred Native Hawaiians, a remarkable erated and revitalized.
Ha­waiian society, particularly in O‘ahu and
and invaluable complement to their narrative history.
among the ruling elite, Nâ Kua‘âina tells a w
broader and more inclusive story of the Ha- By focusing in turn on an island (Moloka‘i),
waiian Islands by documenting the continu- moku (the districts of Hana, Maui, and

Nā Kua ‘āina


ity of Native Hawaiian culture as well as the university of hawai‘i press Puna, Hawai‘i), and an ahupua‘a (Waipi‘o,
changes. Honolulu, Hawai‘i 96822-1888 Hawai‘i), McGregor examines kua‘âina life
e ways within distinct trad­itional land use re-
gimes. Kaho‘olawe is also included as a pri-
Davianna Pòmaika‘i McGregor is professor
mary site where the regenerative force of
e living hawaiian culture w
of ethnic studies at the University of Hawai‘i
the kua‘âina from these cultural kîpuka have
and a historian of Hawai‘i and the Pacific.
revived Hawaiian cultural practices. Each
www.uhpress.hawaii.edu
Davianna Pòmaika‘i McGregor (Continued on back flap)

Jacket photo: Franco Salmoiraghi Jacket design: Leslie Fitch

Revised McGregor jacket 2.indd 1


back flap = 3.5 in. .375 in. back = 6.25 in spine = .875 in. front = 6.25 in .375 in. front flap = 3.5 in. 10/25/06 11:17:26 AM
Nä Kua‘äina
The publication of this book was made possible through
sponsorship of the Native Hawaiian Center of Excellence,
John A. Burns School of Medicine, University of Hawai‘i at Mânoa.
Partial funding for this publication was provided by the
Department of Health and Human Services, Health Resources Services
Administration, Bureau of Health Professions, Division of Health Careers,
Diversity and Development.
N Ā K UA ‘Ā I N A
 living hawaiian culture 

Davianna Pòmaika‘i McGregor

University of Hawai‘i Press


honolulu
© 2007 University of Hawai‘i Press
All rights reserved
Printed in the United States of America
02 03 04 05 06 07 6 5 4 3 2 1

Library of Congress Cataloging-in-Publications Data


McGregor, Davianna.
Nä Kua‘äina : living Hawaiian culture /
Davianna Pömaika‘i McGregor.
p. cm.
Includes bibliographical references and index.
isbn-13: 978-0-8248-2946-9 (cloth : alk. paper)
isbn-10: 0-8248-2946-8 (cloth : alk. paper)
1. Hawaiians—History. 2. Hawaiians—Social life and customs.
3. Hawaiians—Interviews. 4. Oral history.
5. Hawaii—Social life and customs.
6. Subsistence economy—Hawaii—History.
7. Natural resources—Social aspects—Hawaii—History.
8. Social change—Hawaii—History. 9. Hawaii—Rural conditions.
10. Hawaii—History, Local. I. Title.
DU624.65. M39 2006
996.9—dc22
2006006901

University of Hawai‘i Press books


are printed on acid-free paper
and meet the guidelines for permanence
and durability of the Council on
Library Resources.

Designed by Leslie Fitch


Printed by Sheridan Press
For My parents
Daniel Pämawaho and
Anita Branco McGregor
. . . my roots

My love,
Noa Emmett Aluli
. . . my inspiration

My daughter,
Rosanna ‘Anolani Alegado
. . . my life . . . my future


 contents 

Acknowledgments
ix

one
Nä Kua‘äina and Cultural Kïpuka
1

two
Waipi‘o Mano Wai:
Waipi‘o, Source of Water and Life
49

three
Häna, Mai Ko‘olau a Kaupö:
Häna, from Ko‘olau to Kaupö
83

four
Puna: A Wahi Pana Sacred to Pelehonuamea
143

five
Moloka‘i Nui a Hina:
Great Moloka‘i, Child of Hina
191

six
Kaho‘olawe: Rebirth of the Sacred
249

seven
Ha‘ina Ia Mai: Tell the Story
286
contents

appendix i
1851 Petition from Puna Native Hawaiians to
Extend the Deadline to File a Land Claim
305

appendix ii
Number of Males Who Paid Taxes in Puna in 1858
306

appendix iii
Moloka‘i, Petition of July 2, 1845
308

notes
319

bibliography
353

index
365

viii
 acknowledgments 

Nä kua‘äina of Häna, Moloka‘i, Puna, Waipi‘o, and Kaho‘olawe are at the


heart of this book—their lives, knowledge, and spirit of resilience.
Our ancestral spirits and deities, ‘aumäkua and akua of the ‘äina—from the
ocean depths and reefs to streams and lush valleys, volcanic rainforests and
sacred mountain peaks, and up into the sky with its many named winds, clouds,
and rains—are the soul of this book.
My family and loved ones lifted me to connect to ancestors, ancestral lands,
and lively times that have passed, by sharing their vivid memories in comfort-
able homes with fine wine and family dinners:
My parents, Daniel Pämawaho and Anita Branco McGregor; their parents,
Daniel Pämawaho and Louise Aoe McGregor, and David William and Anna
Meyer Branco.
My sisters, Danita ‘Imaikalani Aiu and Myrna Anne Pualehua Kai, and
hänai Claire Pruet.
My daughter, Rosanna ‘Anolani Alegado; her husband, Raymond Edward
‘Awa Kong, Jr.; and his parents, Leona and Alvin Abe.
My nieces and nephews, Puaalaokalani, Pi‘imauna, Holly, Mohala, ‘Imai-
kalani, Kapuaonälanii Këhau o Wai‘ale‘ale, Kamana‘opono, and Kanoe Aiu,
and Lehua Kai.
My uncles and aunts, Jackson and Rita Branco, Robert “Skippy” and Verna
Mae Kawai‘ula Branco, and the late Marion LeeLoy.
My cousins, especially Pilialoha, Marylyn, and Samuel Lee Loy; Marion
Louise and Gordon Machado; Wilmar, Lurline, and Momi McGregor; and
Jackie, Billy, Gregory, Marvalee, Robert, Verna, Lola, Anna, Charlotte,
Michael, and James Branco.
My love, Dr. Noa Emmett Aluli, and his ‘ohana, Mokihana Cockett Aluli
and Nick Teves, and Kalai, Pia, Hayden, Webster, and Noa Aluli.
Community leaders shared their vision and insights, experience and knowl-
edge and inspired me to write about the lives of Nä Kua‘äina:
Of the Protect Kaho‘olawe ‘Ohana founders and their families, George
Helm, Aunty Mae, and the Helm ‘ohana; Uncle Harry Künihi Mitchell and
his son Kimo Mitchell; Uncle Leslie Kuloloio and his mother, Aunty Alice,

ix
acknowledgments

and son Manny; Noa Emmett Aluli; and nä kua ‘ohana members, past and
present.
Of the Edith Kanaka‘ole Foundation, Pualani Kanaka‘ole and Edward
Kanahele; Ulunui Kanaka‘ole Garmon; Parley Kanaka‘ole; Nalani Kanaka-
‘ole; Kekuhi, Huihui, Ahi‘ena, Sig, Tangaro, and Kalä.
Of Hui Ala Loa, Judy Napoleon; Colette Machado, Joyce Kainoa, Wren
Wescoatt, and John Sabas.
Of the Pele Defense Fund and Ka ‘Ohana o KaLae, Palikapu and Lori
Dedman, Margaret McGuire, and Mark Lunning.
The attorneys who supported all of these efforts, Alan Murakami, Nahoa
Lucas, Melody Mackenzie, and Mahealani Kamau‘u of the Native Hawaiian
Legal Corporation; Steve Moore of the Native American Rights Fund; Yuk-
lin Aluli; and Tom Luebben.
Over the past 12 years I have worked on many joint projects with Jon
Matsuoka, dean of the University of Hawai‘i at Mänoa School of Social Work,
and Luciano Minerbi, professor of the University of Hawai‘i at Mänoa Urban
and Regional Planning Department. As a team, calling ourselves “3-M”
(Matsuoka, McGregor, and Minerbi), we journeyed to Moloka‘i, Häna, and
Puna. In the course of meeting with community leaders and their attorneys
and organizing focus groups, resource mapping, interviews, and surveys with
‘ohana from these cultural kïpuka, our team has developed a methodology to
systematically gather and document subsistence and cultural customs and
practices, inventory cultural and natural resources, and support the commu-
nity in protecting all of these. The chapters on Moloka‘i, Häna, and Puna are
in part drawn from the larger studies and reports that we conducted as “3-M.”
My patient colleagues of long standing in the Ethnic Studies Department
of the University of Hawai‘i at Mänoa have consistently given me support,
time, space, and resources to pursue my research and community service and
to progress professionally: Dean Alegado, Ibrahim Auode, Sandy Chock, Mar-
ion Kelly, and Noel Kent. Newer faculty are also supportive with their inter-
action and fresh insights: Monisha Das Gupta, Ulla Hasager, Jonathan Oka-
mura, Ty Kawika Tengan, and Elissa Joy White. The dean of the College of
Social Sciences, Richard Dubanoski, has encouraged my ongoing work and
provided me with special recognition. Karl Kim, Kem Lowry, and Karen
Umemoto of the Urban and Regional Planning Department have been special
colleagues, as they also engage in parallel community research and service.
My oldest and dearest mentor and colleague, Franklin Odo, was a member
of my dissertation committee and helped to critique the material in this book

x
acknowledgments

from that document. As the chairperson of the Ethnic Studies Department,


he encouraged and supported my ongoing research and community service. As
a director of the Asian Pacific American Program of the Smithsonian Institu-
tion, he allowed me to extend my research to Washington, D.C., as a scholar-
in-residence. We also worked together with Elizabeth Tatar and Dave Kem-
ble of the Bishop Museum, the Kaho‘olawe Island Reserve Commission,
Barbara Pope, and Rowland Reeve to feature Kaho‘olawe and the Native
Hawaiian movement in a national exhibit in the Smithsonian Arts and Indus-
tries Building called “Kaho‘olawe: Rebirth of a Sacred Hawaiian Island.”
Information gathered for that exhibit is part of the chapter on Kaho‘olawe.
I also want to acknowledge my colleagues in the Asian Pacific American
History Collective and the Association for Asian American Studies, who
impressed upon me the importance of publishing new books for a national
audience that will give voice to a variety of Native Hawaiians and Pacific
Islanders and enrich the appreciation of the Native Hawaiian culture. These
colleagues include Franklin Odo; Henry Yu of the University of California,
Los Angeles; Gayle Nomura and Steve Sumida of the University of Washing-
ton; Dana Takagi of the University of California, Santa Cruz; Gary Okihiro of
Columbia University; Amy Ku‘ulei Stillman and Vince Diaz of the University
of Michigan; John Tsuchida of California State University, Long Beach; John
Rosa of Arizona State University; and Dorothy and Tom Fujita-Rony of the
University of California, Riverside, and California State University, Fullerton.
My students and especially my lab leaders continue to infuse me with new
insights and perspectives. They daily demonstrate to me the importance of
documenting the lives and culture and knowledge of our küpuna and of our
kua‘äina so that Hawai‘i will continue to be Hawaiian.
A special mahalo to my college freshman world civilizations history pro-
fessor, Gavan Daws, whose own exciting work brings historical heroes, hero-
ines, and villains to life, for kindly reading my manuscript, giving me advice,
and speaking truth.
For her belief in me and assistance in preparing this manuscript for press,
I thank Masako Ikeda. Mahalo to Barbara Dunn and Karen Sinn of the Hawai-
ian Historical Society in my search for photos. To Dr. Ben Young and the
Native Hawaiian Center for Excellence at the University of Hawai‘i at Mänoa
Medical School, mahalo nui loa for their contributions to making this book
more accessible.
mahalo . . . mahalo nui loa

xi
 one 

Nä Kua‘äina and Cultural Kïpuka

ain pelted the decks and the howling wind and twenty-foot ocean

R swells madly rocked our boat as we made our way in dawn’s first light
from the port of Lahaina to the island of Kaho‘olawe. We struggled
for a foothold, while grasping for trash bags to relieve ourselves of the queasy
welling up of fluids deep in our guts. Uncle Harry Mitchell called out to us,
“You had enough? And now, are you ready to turn back?” Everyone begged
to turn around. Before the captain could steer the boat around to head back
most of my students boldly jumped into the wild surf off of Olowalu and
swam to shore rather than suffer the pangs of seasickness all the way back
down the coast to Lahaina.
Uncle Harry sat me down. “You are a college professor, eh?” Yes. “And
you saw the storm that has been gathering for the past few days?” No. “It was
windy when you left O‘ahu?” Yes. “And you felt the storm?” No. “You know
that we go across the channel to Kaho‘olawe on a small boat?” Yes. “Did you
know that there were small-craft warnings before you left O‘ahu?” No. “What
were you thinking about?"
I had been totally oblivious to the major elements of a huge storm swirling
together for the past few days. I was the typical single-minded urban Hawai-
ian academic, bent on getting where I wanted to go, but completely out of bal-
ance with the natural forces around me. Uncle Harry explained, “If I had told
you that you couldn’t make it over to Kaho‘olawe this morning you would
have disagreed, argued, and insisted on going. So I took you out in the boat,
not too far off the coast, not even in the channel, until you had had enough
and were begging me to turn around.”
Through my bitter disappointment at not making it to Kaho‘olawe I
learned one of the most important lessons of my life from kupuna Mitchell.
Always be conscious and respectful of the natural elements around me. As

1
chapter one

Uncle Harry would always say, “Watch . . . look at the moon, the stars, the
clouds, they talk to you . . . listen . . . watch!”
“Aloha ‘äina, aloha ke akua, aloha kekähi i kekähi” (love and respect the
land, love and honor God, love and look after one another, these are the three
important things our küpuna always ask us to remember): this was another
mantra of Uncle Harry. From him I learned that one who understands and
lives by these precepts embraces the world of Native Hawaiians. This Native
Hawaiian worldview is called lökähi, or unity, harmony, balance. It refers to
the unity, harmony, and balance in the universe between humans, nature, and
deities or spiritual life forces. For personal well-being, we need to be in bal-
ance with the people around us, and with the natural and spiritual forces of
life.
So, there I was, spring break 1980, out of balance and stuck with twenty col-
lege students, coolers full of food for a week, grounded by a wild late-March
storm that I never saw coming. Uncle Harry took pity on us. He loaded us
into vans, a truck, and a car and took us home with him to Ke‘anae-Wailua-
nui, Maui. The community of taro farmers and fishermen graciously allowed
us to camp in their church hall. For the next few days Uncle Harry threw us
into the taro patches to earn our lodging and he taught us the mo‘olelo of Ke‘a-
nae-Wailuanui, and of the valleys, streams, and gulches from Ke‘anae through
Häna and out to Ohe‘o. We immersed ourselves in the way of life of the kua-
‘äina of Ke‘anae-Wailuanui. I awoke to a worldview and lifestyle that I would
devote my academic endeavors to helping perpetuate. This is the life ways of
the kua‘äina.
I do not write of ruling chiefs, but of those who made the chiefs rulers. I
write of those who first held the lands of Hawai‘i in trust for the Gods of our
nature and whose descendants have a vested responsibility and right to hold
these lands in trust today. I write of the kua‘äina, the keepers of Hawai‘i’s
sacred lands who are living Hawaiian culture. This is a mo‘olelo, a history, or,
in the Hawaiian sense, a succession of knowledge passed on orally from one
generation to the next of kua‘äina, who shared this knowledge with someone,
such as Mary Kawena Pukui in the 1960s or me in the 1980s and 1990s, as
oral history interviews. They are the source of the knowledge of which I write,
and the shortcomings herein are my own.1
I can remember a time when it was demeaning to be called kua‘äina, for it
meant that one was an awkward and rough country person.2 In Hawaiian, kua
means back and ‘äina means land, so kua‘äina is translated literally in the
Hawaiian Dictionary as “back land.” However, in the context of the Native

2
Figure 1 Uncle Harry Kûnihi Mitchell of Wailuanui, Maui, playing his guitar near
Hakioawa, Kaho‘olawe. Uncle, sparkle in his eye, knee-deep in his lo‘i, introduced me
to the lives of the kua‘âina and their role in the cultural regeneration of Kaho‘olawe.
1979. Franco Salmoiraghi.
chapter one

Figure 2 The rural communities where kua‘âina have remained are cultural kîpuka that have
been bypassed by major historic forces of economic, political, and social change in Hawai‘i.
Uncle Harry Mitchell’s Wailuanui is a cultural kîpuka from which Native Hawaiian culture was
regenerated and revitalized on Kaho‘olawe and a new generation of taro farmers and traditional
healers was trained. 1936. Bishop Museum.

Hawaiian cultural renaissance of the late twentieth century, the word kua‘äina
gained a new and fascinating significance. A kua‘äina came to be looked upon
as someone who embodied the backbone of the land.3 Indeed, kua‘äina are the
Native Hawaiians who remained in the rural communities of our islands, took
care of the küpuna or elders, continued to speak Hawaiian, bent their backs and
worked and sweated in the taro patches and sweet potato fields, and held that
which is precious and sacred in the culture in their care.4 The kua‘äina are
those who withdrew from the mainstream of economic, political, and social
change in the Islands. They did not enjoy modern amenities and lived a very
simple life. This mo‘olelo recounts how the life ways of the kua‘äina enabled
the Native Hawaiian people to endure as a unique, distinct, dignified people
even after over a century of American control of the Islands.

4
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

“Ke ha‘awi nei au iä ‘oe. Mälama ‘oe i këia mau mea. ‘A‘ohe Mälama, pau
ka pono o ka Hawai‘i” (I pass on to you. Take care of these things. If you don’t
take care, the well-being of the Hawaiian people will end): 5 these words were
used by küpuna to pass on knowledge and stewardship of their lands to a cho-
sen successor of the next generation. Gifted with this stewardship responsi-
bility, the successors held their ancestral lands and knowledge sacred in their
memories and passed it on in custom and practice from generation to gener-
ation up through the twenty-first century.
Daniel Pahupu was a kua‘äina and a küpuna whom Mary Kawena Pukui
interviewed on the island of Moloka‘i in 1961 as part of a project to gather
ancestral knowledge about the sacred and significant places in the Hawaiian
Islands, referred to in Hawaiian as wahi pana. In 1961 Mary Kawena Pukui
traveled from island to island interviewing kua‘äina, as the keepers of the wahi
pana, in order to document and thereby perpetuate their unique and profound
knowledge for future generations. Conducting the interviews in Hawaiian, the
kua‘äina shared knowledge with Pukui that had been passed on from one gen-
eration to the next about the lands where their ancestors lived, worked, and
sustained a spiritual connection to the life forces of the universe.6 The land and
nature, like members of the ‘ohana or extended family, were loved. The place-
names they were given reflected their particular character and nature and
contain traditional knowledge accumulated by Hawaiian ancestors in utilizing
the natural resources of these areas, providing kua‘äina with information they
need to understand and adapt to the qualities and character of the land in
which they live, such as soil conditions, local flora or fauna, and seasonal fluc-
tuations. Native Hawaiian ancestors also named the various types of rain and
wind of particular districts. The names of places and natural elements not
only provide a profound sense of identity with the ‘äina or land and natural
resources, they also convey a sense of responsibility to provide stewardship of
the area where they live.
In his introduction to Ancient Sites of O‘ahu, Edward Kanahele explained
the significance of wahi pana in the perpetuation of Native Hawaiian cultural
knowledge. He also explained how the understanding of a place, its names,
and the reason for its designation as a wahi pana is essential to understanding
the area’s function and significance in Native Hawaiian society:

As a Native Hawaiian, a place tells me who I am and who my extended


family is. A place gives me my history, the history of my clan, and the his-
tory of my people. I am able to look at a place and tie in human events that

5
chapter one

affect me and my loved ones. A place gives me a feeling of stability and of


belonging to my family, those living and dead. A place gives me a sense of
well-being and of acceptance of all who have experienced that place.
The concept of wahi pana merges the importance of place with that
of the spiritual. My culture accepts the spiritual as a dominant factor in life;
this value links me to my past and to my future, and is physically located at
my wahi pana.
Where once the entire Native Hawaiian society paid homage to
numerous wahi pana, now we may give wahi pana hardly a cursory glance.
Only when a Native Hawaiian gains spiritual wisdom is the ancestral and
spiritual sense of place reactivated. Spiritual knowledge and the wahi pana
are ancestrally related, thus spiritual strength connects to the ancestral
guardians, or ‘aumakua. My ‘aumakua knew that the great gods created
the land and generated life. The gods infused the earth with their spiritual
force or mana. The gravity of this concept was keenly grasped by my ances-
tors: they knew that the earth’s spiritual essence was focused through the
wahi pana. ( James, in E. Kanahele, 1991)

Kua‘äina live in rural communities throughout the Hawaiian islands. In


these areas, Native Hawaiians have maintained a close relationship to and
knowledge of their wahi pana. These rural communities are special strong-
holds for the perpetuation of Hawaiian culture as a whole. An analogy which
conveys a sense of the significance of these areas can be found in the natural
phenomenon of the volcanic rainforest. From the island of Hawai‘i come the
oli or chants of Pele and her creative force. The oli hulihia, in particular, mean-
ing overturned, overthrown, and upheaval, speak of volcanic events, such as
in the following chant.

Kua Loloa Kea‘au I Ka Nähelehele/Kea‘au Is a Long Ridge of Forest

Kua loloa Kea‘au i ka Nähelehele Kea‘au is a long ridge of forest


Hala kua hulu Pana‘ewa i ka lä‘au The hala ridges of Pana‘ewa are the trees
Ino ka maha o ka‘öhi‘a Numerous are the severed ‘öhi‘a
Kü kepakepa kamaha o ka lehua, Zigzag are the severed lehua
Po‘ohina i ka wela a ke Akua The grayish mist is the Goddess’s hot
revenge
Uahi Puna i ka oloka‘a pöhaku Puna is smoky with hot rolling stones
Nä pe‘a ‘ia e ka Wahine Persecuted by the Goddess

6
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

Nänahu ahi ka ka papa o Oluea The plain of Oluea is bitten with fire
Momoku ahi Puna, hala i ‘äpua Puna is cut off by fire, even to ‘Apua
A ihu e, a ihu la, The flow is heading this way and that.
A hulihia la i kai, Turning upside down toward the sea,
A ihu e, a ihu la The flow is heading this way and that,
A hulihi la i uka, An upheaval toward the uplands,
A ua wä‘awa‘a It is so desolate, uninhabitable,
A ua noho ha‘aha‘a Made low by the Goddess
A ua hele helele‘ihelele‘i Falling, falling, nothing but ashes.7

Even as Pele claims and reconstructs the forest landscape, she leaves intact
whole sections of the forest, with tall old-growth ‘öhi‘a trees, tree ferns, creep-
ing vines, and mosses. These oases are called kïpuka. The beauty of these nat-
ural kïpuka is not only their ability to resist and withstand destructive forces

Figure 3 The volcanic rainforest in Puna, Hawai‘i, features numerous beautiful natural kîpuka
of old-growth forest from which fresh fields of lava are eventually revegetated. Thus the Puna
rainforest is a mosaic of old-growth forest and new-growth forest.

7
chapter one

of change, but also their ability to regenerate life on the barren lava that sur-
rounds them. For from these kïpuka come the seeds and spores carried by
birds and blown by the wind to sprout upon and regenerate the forest on the
new lava, sparking a dynamic new cycle of coming into and passing out of life.
The rural communities where kua‘äina have remained are cultural kïpuka
that have been bypassed by major historic forces of economic, political, and
social change in Hawai‘i. Like the dynamic life forces in a natural kïpuka, cul-
tural kïpuka are communities from which Native Hawaiian culture can be
regenerated and revitalized in the setting of contemporary Hawai‘i. More-
over, from the examination of the lives of kua‘äina in Hawaiian cultural kïpuka
emerges a profile of the strongest and most resilient aspects of the Native
Hawaiian culture and way of life. Such an examination provides insight into
how the Native Hawaiian culture survived dynamic forces of political and eco-
nomic change throughout the twentieth century.

Features of Cultural Kîpuka


Originally, cultural kïpuka were traditional centers of spiritual power. In tra-
ditional Hawaiian chants and mythology, major akua or Gods and Hawaiian
deities were associated with these wahi pana. These districts were isolated and
difficult to access over land and by sea. Owing to the lack of good anchorage
and harbors, early traders often bypassed these districts in favor of more acces-
sible areas. The missionaries entered these areas and established permanent
stations during a later period than in other parts of Hawai‘i. Thus, traditional
Native Hawaiian spiritual beliefs and practices persisted there, without com-
petition, for a longer period of time. When Christian influences entered these
areas, they had to coexist with traditional beliefs and practices.
The geography of these districts discouraged the widespread or long-term
development of sugar plantations. In the arid areas, the lack of water resources
made development of sugar plantations unfeasible. In the areas with sufficient
rainfall, the terrain was too steep or rugged for plantation agriculture. Where
plantation agriculture failed, such as in Moloka‘i and the Häna district,
ranches were able to succeed. The ranches employed Native Hawaiian men as
cowboys and allowed them to live with their families in these isolated districts
and pursue traditional fishing, gathering, and hunting activities to supplement
their wages. In some areas small stores provided kua‘äina access to some basic
Western commodities such as kerosene, lanterns, tools, flour, crackers, and
sugar. However, for the most part kua‘äina were not consumer oriented.

8
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

Money to purchase these basic provisions came from selling taro or fish or an
occasional day’s labor for a local entrepreneur or the government road crew.
Where neither plantations nor ranches were established, traditional subsis-
tence activities continued to be pursued, undisturbed by modern economic
development. In the wetland areas taro continued to be farmed, often in con-
junction with rice. In the arid areas, sweet potatoes, dryland taro, and other
traditional and introduced crops suited to the dry soil and climate were culti-
vated. Thus, the natural features and resources of these districts that rendered
them unsuitable for plantation agriculture and ranching played a role in the
survival, and eventual revitalization, of Native Hawaiian cultural, spiritual,
and subsistence customs and practices. Concurrently, the quality and abun-
dance of the natural resources of these rural communities can be attributed
to the persistence of Native Hawaiian cultural and spiritual values and prac-
tices in the conduct of subsistence activities.
Very few haole or Caucasians settled in these districts, and kua‘äina had very
little interaction with the outside community. Chinese who completed their
contracts on the plantation and did not return home or move to the mainland
leased or rented lands from the kua‘äina. Some served as middlemen, market-
ing whatever taro and fish kua‘äina desired to sell in the towns and bringing
back consumer goods for sale or barter in the rural communities. Where there
was a small rural store in these districts, it was invariably owned by a Chinese,
who in some cases was married to a Native Hawaiian woman.
By 1930 there were still seventeen rural districts where Native Hawaiians
were predominant. Andrew Lind wrote of the significance of these areas for
the continuity of Hawaiian culture:

These racial havens—small population islands still relatively secure from


the strong currents which have swept the archipelago as a whole into the
world-complex of trade—are strikingly similar to those which appear in
the census of 1853. The dry and rocky portions of Kau, Puna and the
Kona coast, the deep valley of Waipio, the wild sections of Hana, Maui,
portions of lonely Lanai and Molokai where industrial methods of agricul-
ture have not succeeded, the leper settlement, and Niihau, the island of
mystery—these are the places of refuge for some 4,400 or nearly one-fifth,
of the native Polynesians . . .
The old fish and poi company, with its accompaniment of tutelary
deities, taboos, religion, and magic, still persists in modified form within
many of these isolated communities. A small plot of taro and access to the

9
chapter one

sea and the mountains are apparently all that is required for the satisfaction
of their material wants. The wage from an occasional day’s work on the
government road enables them to purchase the necessary supplies which
the old economy cannot now provide . . . The natives themselves have
found these rural havens where the economy of life to which they are
best adapted can survive.8

The seventeen districts where Native Hawaiians comprised a majority in


1930 were small isolated valleys and districts on the fringes of Hawai‘i’s eco-
nomic and social life. The overall population in these districts averaged 341,
and the number of Native Hawaiians in them averaged 248. The largest dis-
trict, Pala‘au-Ho‘olehua on Moloka‘i, had 1,031 inhabitants, of whom 826
were Hawaiian; and the smallest, Keömuku on Läna‘i, had 54 inhabitants, of
whom 33 were Hawaiian.
On Hawai‘i Island, these districts included Kalapana (88 percent Hawai-
ian); Waipi‘o and Waimanu (66 percent Hawaiian); Keaukaha, an area opened
for Hawaiian homesteading in 1925 (83 percent Hawaiian); the Pu‘uanahulu,
Pu‘uwa‘awa‘a, and Kïholo district (79 percent Hawaiian); the Kohanaiki,
Kalaoa, Hu‘ehu‘e, and Honoköhau district (52 percent Hawaiian); ‘Ala‘ë,
Pähoehoe, Honokua, ‘Opihihale, and ‘Ölelo-Moana district (82 percent
Hawaiian); and Ho‘öpüloa, Papa, Alika, Kaunämano, Kapua, and Miloli‘i dis-
trict (64 percent Hawaiian).
On Maui, the districts with a predominance of Hawaiians included Ke‘anae
to Nähiku (78 percent Hawaiian); Nähiku to Häna (55 percent Hawaiian);
Kïpahulu (80 percent Hawaiian); and Kaupö to Kahikinui (86 percent Hawai-
ian). On Moloka‘i the districts with a majority of Hawaiians included Kawela
to Ualapue (62 percent Hawaiian); Kalawao (66 percent Hawaiian); and the
Hawaiian homestead lands at Pala‘au-Ho‘olehua (80 percent Hawaiian). The
small district of Keömuku on the island of Läna‘i was 61 percent Hawaiian.
The island of Ni‘ihau was 93 percent Hawaiian. On O‘ahu, only the district
that included the Kalihi Receiving Station and the hospital for Hansen’s dis-
ease patients had a majority of Hawaiians; 61 percent of the patients were of
Hawaiian ancestry. The statistics are summarized in table 1.
Except for the homestead districts of Pala‘au-Ho‘olehua, and Keaukaha,
the Hansen’s disease receiving station at Kalihi, and the settlement at Kala-
wao, the ethnic concentrations of Hawaiians were not induced or encouraged
by governmental policy. Among the remaining districts, certain qualities and
patterns of change and continuity can be observed as common to them.9

10
table 1 rural districts with population over 50% hawaiian, 1930
district total hawaiian % hawaiian

Hawai‘i
Kalapana 235 207 88
Waipi‘o, Waimanu 271 178 66
Keaukaha 754 625 83
Pu‘uanahulu, Pu‘uwa‘wä‘a, Kïholo 149 117 79
Kohanaiki, Kalaoa, Hu‘ehu‘e,
Honoköhau 422 221 52
‘Ala‘ë, Pähoehoe, Honokua,
‘Opihihale, ‘Ölelo-Moana 239 197 82
Ho‘öpüloa, Papa, Alika, Kaunämano,
Kapua, Miloli‘i 146 94 64
Maui
Ke‘anae/Häna 337 262 78
Nähiku/Häna 182 101 55
Kïpahulu 147 118 80
Kaupö 185 160 86
Moloka‘i
Kawela, Ualapue 789 487 62
Kalawao 605 400 66
Pala‘au-Ho‘olehua 1,031 826 80
Läna‘i
Kahue to Kamaiki (Keömuku, Läna‘i) 54 33 61
O‘ahu
Kalihi Receiving Station/hospital 114 70 61
Ni‘ihau 136 126 93

Total 5,796 4,222 72


Statistics based on U.S. Bureau of the Census, 1931, pp. 70, 72, table 22.
The district boundaries were found in Governors’ Proclamations, 1926–1930,
pp. 6–21, 128–47; and Map no. 301, O‘ahu, State of Hawai‘i Archives.
chapter one

In the 1930s two anthropologists from the Bernice Pauahi Bishop Museum,
E. S. Craighill Handy and Elizabeth Green Handy, in collaboration with Mary
Kawena Pukui, traveled through all of the major districts of the Hawaiian
Islands to assess the original native horticulture of the islands prior to the
introduction of Euro-American plants. Their findings were published in The
Hawaiian Planter, vol. 1, and in Native Planters in Old Hawai‘i: Their Life, Lore,
and Environment. These volumes provide a snapshot of the lives of the kua-
‘äina in the rural districts during the 1930s. In the foreword to Native Planters
in Old Hawai‘i Craighill Handy wrote:

It was shown that the older generation of country natives still had an
extraordinarily intimate and thorough knowledge of the many varieties of
taro, sweet potato, sugar cane, and banana still cultivated . . . The Hawai-
ians, more than any of the other Polynesians, were a people whose means
of livelihood, whose work and interests, were centered in the cultivation of
the soil. The planter and his life furnish us with the key to his culture.10

Only a handful of cultural kïpuka survived the onslaught of development


after Hawai‘i became a state in 1959. These included the islands of Moloka‘i
and Ni‘ihau; on Maui, the districts of Häna (from Ke‘anae to Kahikinui) and
Kahakuloa; on Hawai‘i, the districts of Ka‘ü, Puna, and Waipi‘o Valley, and
the small fishing communities of Kohala and Kona, excluding Kailua. On
O‘ahu, the Windward Valleys of Kahana, Waiähole, Waikäne, Hau‘ula, and
Lä‘ie and sections of the Wai‘anae Coast, and on Kaua‘i, Waipä, Kekaha, and
Anahola retained features of cultural kïpuka, although the population of
Native Hawaiians has fluctuated.

Cultural Kîpuka and Lòkâhi


Rural Native Hawaiians today descend from kua‘äina who were content to
remain in the isolated districts, though many others moved out during the
twentieth century. For those who stayed behind, life was filled with interest-
ing natural phenomena and forces that challenged them as they sought out
their subsistence needs. The kua‘äina way of life is a model of the Hawaiian
belief, custom, and practice of lökähi. Kua‘äina were intimately conscious of
their ‘äina—the lands and natural resources where they live. They built their
economic activities around the life cycles of the various fish, animals, and
plants they depended upon for food. Thus, from month to month, as the sea-
sons shifted from wet to dry, their food sources changed in accordance with

12
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

the type of fish, fruits, and plants that were in season. This knowledge of the
environment and natural life forces was often passed on and remembered as
Native Hawaiian traditions and beliefs. Native Hawaiians often chose to per-
sonify the forces of nature as spiritual entities or akua and ‘aumakua, Gods and
ancestral spirits. They created legends and myths to describe and remember
the dynamic patterns of change that they observed.
The Kumulipo, genealogy chant of the family of King Kaläkaua and Queen
Lili‘uokalani, exemplifies the Native Hawaiian belief in lökähi. It traces the
origin of humans through a process of evolution in nature, beginning with day
arising from the black primordial night. It continues with coral polyps as the
first life form, which evolves into various forms of marine life and then into
plants and animals on the land, through ancestral deities, and down to several
generations of chiefly ancestors of the Native Hawaiian people. This chant,
therefore, establishes that Native Hawaiians are descended from, and thus
inextricably related to, natural life forms and the spiritual life forces personi-
fied as deities.11
We also learn of lökähi in the Häloa tradition. In this mo‘olelo the first-
born offspring of Wäkea, Sky Father, and his daughter Ho‘ohokukalani,
maker of the stars in the heavens, Häloa Naka, is stillborn. When buried,
Häloa Naka grows into the first kalo plant. Their second-born child, Häloa,
is a progenitor of the Native Hawaiian people. This tradition, again, estab-
lishes that Native Hawaiians are the young siblings of the kalo plant and that
both descended from the deities Wäkea and Ho‘ohokukalani. This relation-
ship is eloquently described in Native Planters in Old Hawaii:

When, therefore, the learned men in early times, all of them taro planters,
compounded this myth as a part of their heritage of ancient lore, which
describes the birth of nature and man as the consequence of the impreg-
nation of Mother Earth by Father Sky, they sealed into their people’s
unwritten literature this idea, that the taro plant, being the first-born,
was genealogically superior to and more kapu (sacred) than man himself,
for man was the descendant of the second-born son of Sky and Earth.
The taro belonged, then, in the native parlance of family status, to the
kai kua‘ana (elder or senior) branch of cosmic lineage, man himself to
the kai kaina or junior.12

In the “Oli Kühohunu o Kaho‘olawe Mai Nä Küpuna Mai” (Deep Chant


of Kaho‘olawe from Our Ancestors), way-finding voyagers are elated at the
sight of the island of Kaho‘olawe. They dedicate the island to Kanaloa, Hawai-

13
chapter one

ian God of the ocean, out of gratitude for his guidance of their double-hulled
canoe across the vast Pacific. In this example of lökähi, the Hawaiian way-
finders who composed this chant bestow upon the island of Kaho‘olawe the
distinction of being honored as a body form of the God Kanaloa. Subse-
quently, the voyagers develop Kaho‘olawe into a center to train navigators in
celestial way-finding, which in its essence involves the acquisition of intimate
knowledge of the natural forces of the ocean, winds, and stars collectively
personified as Kanoloa.13
In tracing unbroken lineal descent from the original Native Hawaiians who
had settled the districts, kua‘äina also claim ancestry not only with the ‘auma-
kua, but also with the ‘uhane or spirits of the land and resources where they
live. Kua‘äina continue to acknowledge the presence of their spiritual ances-
tors in the surrounding land by maintaining respectful practices in the use of
the land, streams, ponds, and ocean. These lands are treated with love and
respect like a küpuna of the ‘ohana. They regularly visit the various areas in
the course of subsistence gathering. While traveling to the various ‘ili or sec-
tions of the traditional cultural practices region, through dirt roads and trails,
along spring-fed streams, and the shoreline, practitioners continuously stay
alert to the condition of the resources. If a resource is declining they will
observe a kapu or restriction on its use until it recovers. They may even replant
sparse areas. They are acutely aware of changes due to seasonal and life cycle
transformations in the plants and animals. Plants and animals in their repro-
ductive stage are not gathered. As kua‘äina gather in their traditional area,
they also renew their understanding of the landscape, the place-names, names
of the winds and the rains, traditional legends, wahi pana, historical cultural
sites, and the location of various native plants and animals. An inherent aspect
of these practices is conservation to ensure availability of natural resources for
present and future generations.
Many kua‘äina have also continued to cultivate fish in ponds and the open
ocean by regularly feeding the fish in conjunction with making offerings at the
kü‘ula shrines that mark their ocean fishing grounds. Taro and other domestic
crops are planted according to the moon’s phases to assure excellent growth.
Kua‘äina take advantage of seasonal fruits and marine life for their regular
diet. Native plants are utilized for healing of illness by traditional methods that
involve both physical and spiritual cleansing and dedication. Cultural knowl-
edge attached to the traditional names of places, winds, and rains of their dis-
tricts inform kua‘äina about the effect of the dynamic forces of nature upon
the ocean and the land in their area, and activities are planned accordingly.

14
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

Legends and chants inform them about how their ancestors coped with such
elements. Thus, in the cultural kïpuka, traditional Native Hawaiian custom,
belief, and practice continue to be a practical part of everyday life, not only
for the old people, but also for the middle-aged and the young.
The undeveloped natural resources in these areas still provide an abun-
dance of foods for the kua‘äina who live in these districts. Forested lands pro-
vide fruits to eat. Vines, plants, and wood are used to make household imple-
ments and tools or as herbs of healing. The forest provides a natural habitat
for animals that are hunted for meat. Aquatic life flourishes in the streams.
The ocean provides an abundance of food. Subsistence activities continue to
be the primary source of sustenance for the kua‘äina. Production in these dis-
tricts is heavily oriented toward home consumption.
Kua‘äina also look after one another through maintaining relationships of
‘ohana or large extended-family networks. Hänai, or the adoptive raising of
children of relatives, continues to be commonly practiced. Ties with family
members who move to another island, especially O‘ahu, are maintained. If
some of the children move away to the city, one or two remain behind to care
for parents and the family kuleana or ancestral lands. Often those who move
away send children home to be raised by the extended family during breaks
from school and holidays. Families often visit each other between islands and
exchange food gathered or raised through subsistence activities.
It should be noted that the methods and techniques of accessing, acquiring,
or utilizing traditional natural resources may have changed over time. How-
ever, this does not detract from the fact that the purpose of the activities is to
provide for Native Hawaiian ‘ohana and their community and that the activi-
ties are guided by traditional Native Hawaiian kapu or restrictions and guide-
lines associated with customary subsistence, cultural, and religious practices.
For example, Hawaiian fishermen may use motorboats rather than canoes to
get to their ancestral fishing grounds. They may use a nylon net rather than
one woven from native plant materials to surround fish or to entangle them
in the overnight fluctuating tides. In most cases they are still utilizing ances-
tral knowledge of ocean tides, currents, and reefs to locate and catch the fish.
Their catch is used to honor family ‘aumakua and to feed their extended fam-
ilies and neighbors. Hawaiian hunters may drive a truck on a dirt road rather
than walk along a trail to reach the area of forest where pigs roam. They may
use a gun rather than a spear or knife. Since agriculture and residential devel-
opment have destroyed the lowland forest areas where the pigs used to be
plentiful and easily reached on foot trails, Hawaiians must go deeper into the

15
chapter one

same forests or higher up the same mountain hunted by their ancestors. The
meat is shared with their large extended families as well as with neighbors who
no longer have the stamina to go out and hunt.
Hawaiian custom and practice are distinguished not only by the honor and
respect for traditional ‘ohana cultural values and customs to guide subsistence
harvesting of natural resources, but also by the uses made of the resources.
Thus, when I speak of subsistence in this mo‘olelo I do not mean that the
kua‘äina acquire all that they need to live from cultivation, gathering, fishing,
and hunting. As the market economy evolved in Hawai‘i, creating a demand
for manufactured goods, and when taxes were imposed by the government,
Native Hawaiians had to earn cash and interact with the market system.
Instead, the definition of subsistence used in this mo‘olelo is that developed by
the Governor’s Task Force on Moloka‘i Fishpond Restoration in 1993: “Sub-
sistence is the customary and traditional uses . . . of wild and cultivated renew-
able resources for direct personal or family consumption as food, shelter, fuel,
clothing, tools, transportation, culture, religion, and medicine for barter, or
sharing, for personal or family consumption and for customary trade.”
In addition, ‘ohana values and customs that guide subsistence activities are
models of the practice of lökähi in modern Hawai‘i. The first rule with regard
to the land, ocean, and natural resources is to only take what is needed. Wast-
ing natural resources is strongly condemned. It is also important to protect the
ability of living resources to reproduce. Thus, kua‘äina gather according to the
life cycle of the resource and fish only during the particular species’ non-
spawning seasons: different fish are caught during different seasons of the year
to allow the animals to reproduce. In addition, kua‘äina alternate the areas
where they gather, fish, and hunt in order to allow the resources to replenish
themselves. If an area is observed to have stressed or declining resources due
to drought, storm damage, or harvesting, a kapu on harvesting in the area is
observed. Resources are replanted if appropriate.
Resources are always abundant and accessible to those who possess knowl-
edge about their location and have the skill to obtain them. There is no need
to overuse a more accessible area. More accessible resources are left for the
küpuna to harvest. Young men and women are expected to venture farther
afield to acquire what they need. The knowledge and skill that has been passed
down intergenerationally is respected and protected. It is kept within the fam-
ily and not carelessly given away to outsiders. This knowledge includes an
understanding of the areas which are kapu or reserved for various members of
the community. Kua‘äina usually fish, hunt, and gather in the areas tradition-

16
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

ally used by their ancestors. If they go into an area outside of their own for
some specific purpose, they usually ask permission and go with people from
that area.
Kua‘äina never speak openly about plans for going out to subsistence hunt,
gather, or fish. When actually venturing out on an expedition, they keep
focused on the purpose and goal for which they set out to fish, hunt, or gather.
If they gather additional resources along the way, they do so when they are
coming out of the area, never when they are headed for their destination. They
are certain to stay aware of the natural elements and alert to natural signs.
They respect the resources and the spirits of the land, forest, and ocean and
do not act loud and boisterous. This enables them to better observe ho‘ailona
or natural signs important for their sense of direction, safety, and well-being.
For example, the sound of falling boulders signals flash flooding in a stream.
Sea birds flying inland before day’s end signal that a storm is moving in from
the ocean.
The resources acquired through subsistence enterprises are shared with
members of the broader ‘ohana, neighbors, and friends. In particular, the
young kua‘äina take care of the küpuna who passed on their knowledge and
experience to them and are now too old to go out on their own. They also take
care of the widows and women who are single heads of households, who don’t
have men to provide for their subsistence needs. Finally, resources sacred to
‘aumakua of their ‘ohana are respected as sacred to them and never gathered.
Thus, kua‘äina living in cultural kïpuka are successful in acquiring the basic
necessities for their families through subsistence activities by employing tra-
ditional cultural and spiritual knowledge and practices passed down to them
from their küpuna.

Benefits of Subsistence
Subsistence activities have added benefits related to family cohesion, health,
and community well-being. A subsistence economy emphasizes sharing and
redistribution of resources, which creates a social environment that cultivates
community and kinship ties, emotional interdependency and support, pre-
scribed roles for youth, and care for the elderly. Emphasis is placed on social
stability rather than on individual efforts aimed at income-generating
activities.14
Through subsistence, families attain essential resources to compensate for
low incomes. They can also obtain food items, especially seafood, that might

17
chapter one

be prohibitively expensive in a strict cash economy. If families on fixed


incomes were required to purchase these items, they would probably opt for
cheaper, less healthy foods that would predispose them to health problems. In
this respect, subsistence not only provides food, but also ensures a healthy
diet.
Subsistence generally requires a great amount of physical exertion (e.g.,
fishing, diving, hunting), which is a valuable form of exercise and stress reduc-
tion and contributes to good physical and mental health. It is also a form of
recreation that the whole family can share in. Family members of all ages con-
tribute to different phases of subsistence, be it active hunting, fishing, gath-
ering, or cleaning and preparing the food for eating. Older family members
teach younger ones how to engage in subsistence and prepare the food, thus
passing on ancestral knowledge, experience, and skill.
Another benefit of subsistence is sharing and gift giving within the com-
munity. Families and neighbors exchange resources when they are abundant
and available, and the elderly are often the beneficiaries of resources shared
by younger, more able-bodied practitioners. Most kua‘äina believe that gen-
erosity is rewarded with better luck in the future.
Resources obtained through subsistence are also used for a variety of spe-
cial life cycle occasions that bond families and communities. Resources such as
fish, limu ‘opihi, wild venison, and so on are foods served at lü‘au for baby birth-
days, graduations, weddings, and funerals. ‘Ohana and community residents
participate in these gatherings, which cultivate and reinforce a sense of family
and community identity. If ‘ohana members had to purchase such resources
rather than acquire them through subsistence, the cost would be prohibitive,
and the number of ‘ohana gatherings would decrease. Subsistence activities
therefore enable ‘ohana to gather frequently and reinforce important relation-
ships and support networks.
The time spent engaged in subsistence in the natural environment also
cultivates a strong sense of environmental kinship that is the foundation of
Hawaiian spirituality. Kua‘äina reinforce their knowledge about the landscape,
place-names and meanings, ancient sites, and areas where rare and endangered
species of flora and fauna exist. This knowledge is critical to the preservation
of natural and cultural landscapes because they provide a critical link between
the past and the present. For example, wahi pana that are referred to in ancient
chants and legends can be lost amidst changes due to modernization. However,
visiting such places and sites while engaged in subsistence provides a continu-

18
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

ity that is critical to the survival and perpetuation of the knowledge of these
cultural places.

Mo‘olelo
I write this mo‘olelo, or succession of oral traditions, to acknowledge the role
of the kua‘äina in the various cultural kïpuka of the Hawaiian Islands in per-
petuating traditional and customary Native Hawaiian belief, custom, and prac-
tice. I hope to show that protection of natural resources and of the subsistence
livelihoods of the kua‘äina in the cultural kïpuka is essential to the perpetua-
tion of Native Hawaiian culture, as a whole, for future generations. What is
at stake in planning for the future of these cultural kïpuka is the perpetuation
not just of a rural lifestyle, but of the Native Hawaiian way of life itself. In
order for those of us who live in Hawai‘i to attain lökähi and live in balance

Figure 4 Representative of two distinct generations and ‘ohana of Moloka‘i subsistence


fishermen, “Mac” Kelsey Poepoe and Kanohowailuku Helm walk the old fishing trail to Mokio
Point near the northwest corner of Moloka‘i, monitoring the subsistence fishing grounds for
Native Hawaiian homesteaders. 2005. Richard A. Cooke III.

19
chapter one

with our fragile island environment, we need to protect our cultural kïpuka.
This may be a way to offset and perhaps begin to reverse the dramatic trans-
formations of the natural and cultural landscapes of places such as Honolulu,
O‘ahu; Kailua-Kona; and Lahaina, Maui.
Intrigued by Andrew Lind’s account of cultural kïpuka in the 1930s, I
decided to select an island, a moku or district, and an ahupua‘a or basic geo-
graphic subdistrict usually coinciding with a valley from his list of remote areas
to research as case studies. This would allow me to study the life ways of the
kua‘äina within the distinct traditional land use regimes from the level of an
ahupua‘a through that of a district to that of an island. I selected Moloka‘i as
the island; Häna on Maui, from Ke‘anae to Kaupö, as the district; and Wai-
pi‘o on Hawai‘i as the ahupua‘a.
As for the island, Moloka‘i was the larger in size and population and had
more diverse and abundant resources than Läna‘i and Ni‘ihau, the other two
islands mentioned by Lind. On Moloka‘i, Hawaiians comprised a majority of
the population through 1930. Among the small islands, Moloka‘i had the
largest number of kuleana holders in 1930, despite the large concentrations of
land under Moloka‘i Ranch, Pu‘u o Hoku Ranch, the Bishop Estate, and the
Territorial Government. Kaho‘olawe, as lands of the Kingdom of Hawai‘i that
were ceded by the Republic of Hawai‘i to the U.S. government, was entirely
under the control of the Territorial Government. It was leased out to Angus
MacPhee for ranching in 1930. Ninety-eight percent of Läna‘i was owned by
the Dole Corporation in 1930. The entire island of Ni‘ihau was owned by the
Ni‘ihau Ranch Company.
In modern Hawai‘i, the people of Moloka‘i continue to proudly proclaim
their island as the “Last Hawaiian Island.” In addition, the kua‘äina of Molo-
ka‘i led the movement to reclaim Kaho‘olawe as sacred Hawaiian land and to
revitalize the Hawaiian cultural practices of aloha ‘äina or love and respect for
the land on every island. This connection between the kua‘äina of Moloka‘i
and the cultural renaissance that developed around the Kaho‘olawe move-
ment clearly illustrated the regenerative quality of the cultural kïpuka.
I selected Häna, on Maui, as the district because of its distinctive landscape
and its pristine and diverse native natural resources, and because the ku‘a‘äina
from that district had also played an important role in the Kaho‘olawe move-
ment and the Hawaiian cultural renaissance. In addition, as discussed at the
beginning of this chapter, I was first introduced into the world of the kua‘äina
through Uncle Harry Mitchell in the Ke‘anae-Wailuanui community of the
Häna district.

20
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

I selected Waipi‘o as an ahupua‘a in order to develop a case study on the


island of Hawai‘i and because it played a critical role in the survival of taro
cultivation in the islands as a whole. My research revealed that the island of
Moloka‘i, the Häna Coast, and Waipi‘o Valley were highly valued by Native
Hawaiian chiefs as areas over which to maintain control. These areas com-
prised lands that were extremely well suited for the cultivation of taro and use-
ful native plants, had streams abundant with edible aquatic resources, and were
close to abundant ocean resources.
These case studies were initially developed for sections of my disserta-
tion,15 which documented the life of Native Hawaiians during the first thirty-
two years of direct American rule, 1898 through 1930. Subsequently, I had
the opportunity to conduct additional oral history interviews and historical
research and studies on the island of Moloka‘i and the Ke‘anae-Wailuanui
ahupua‘a of the Häna district. In addition, I conducted ethnographic research
in the district of Puna, on Hawai‘i, and the ahupu‘a‘a of Waiähole, Waikäne,
and Hakipu‘u, on O‘ahu, in separate studies.
In preparing this manuscript I decided to include the additional material
gathered for Moloka‘i and Ke‘anae-Wailuanui and to include a chapter on the
district of Puna. Puna enriches the case studies because it is the home of Pele
and her family of deities. Its landscape and resources are constantly trans-
formed by seismic episodes and volcanic flows. The Puna families sustain a
dynamic relationship to Pele as ‘aumakua.16
These case studies are not exhaustive histories of the kua‘äina and the cul-
tural kïpuka selected. They are designed to provide an essential history of each
area, to share the insights and perspectives of the kua‘äina who live there, and
to convey the importance of protecting the resources and cultural practices of
these and other cultural kïpuka. They also provide insight into the importance
of continuing to document both the oral and the written histories of Hawai‘i’s
kua‘äina and the cultural kïpuka.
The case studies emphasize the experiences of the kua‘äina in the early
twentieth century up through World War II. The war and the 1946 tidal wave,
which struck right after the end of the war, served as the major turning point
in the social and economic development of the islands, even in the cultural
kïpuka. Moreover, the collections of oral histories of küpuna for Waipi‘o,
Häna, Puna, and Moloka‘i focus on their experiences growing up and living
in these communities in the prewar period.
Through my research of these areas and interaction with the kua‘äina who
live there, I have come to the realization that the mainstream history of

21
chapter one

Hawai‘i focuses too narrowly on the history of change and of cultural impact
upon Hawaiian society, concentrating on O‘ahu and the ruling elite. A broader
and more inclusive history of the Hawaiian islands would document not only
the changes, but also continuity of Native Hawaiian culture. It would develop
a history of the experiences of Native Hawaiian women as well as of men who
raised their extended families by farming and fishing throughout the various
islands of Hawai‘i, including the rural communities I call cultural kïpuka. This
work contributes to a much broader history of the Native Hawaiian people
and the Hawaiian Islands. Because time and space limited my ability to con-
duct ethnographic research on all the cultural kïpuka on all our islands, it is
my hope that this work can inspire and inform new research to be conducted
on the cultural kïpuka not selected and contribute to the development of a
more comprehensive history, one that includes the rural districts on all the
major Hawaiian Islands.
Each case study begins by examining the traditional cultural significance of
the district. The ‘ölelo no‘eau or descriptive proverbs and poetic sayings for
which the area is famous are interpreted, and a descriptive chant for the area
is translated and interpreted. These provide valuable insights into the cultural
resources and features for which the area was known and thus the role of this
area overall in the cultural practices and customs of Native Hawaiians. This is
followed by a discussion of the history of the landscape and its settlement, the
deities who dwelt there, and the ruling chiefs who controlled the area. Next
the case studies review how developments in the nineteenth century affected
the life of the kua‘äina in the area. Each case study then provides elaborate
descriptions of the natural cultural resources available to the kua‘äina for their
subsistence and livelihoods and of the beliefs, customs, and practices which
guided their lives. Finally, an overview of the social and economic changes in
each area through the end of the twentieth century as well as a discussion of
the elements of continuity still evident in the lives of the kua‘äina in these
communities is provided.
There is a final chapter on Kaho‘olawe, which is not a cultural kïpuka.
Instead, it is included to demonstrate how the kua‘äina from the cultural
kïpuka studied were instrumental in restoring the natural and cultural
resources of Kaho‘olawe and reviving Native Hawaiian beliefs, customs, and
practices on the island. Kaho‘olawe demonstrates the regenerative function of
the kua‘äina from the cultural kïpuka examined in the earlier chapters of the
book.

22
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

Establishment of the Indigenous Hawaiian Nation


Research by archaeologists, anthropologists, and ethnographers over the past
thirty years suggests that the history of settlement, before continuous contact
with Europeans in 1778, may be looked at in four distinct wä kahiko or historic
eras: the Colonization Period (1–600 CE), the Developmental Period (600–
1100), the Expansion Period (1100–1650), and the Proto-Historic Period
(1650–1795).17
Ongoing subsurface archaeology continues to uncover evidence that sug-
gests the date of first settlement is close to the time of Christ.18 Archaeologi-
cal evidence suggests that the initial population was small, consisting of a few
canoe loads of families and numbering at most around a hundred. They prob-
ably originated from the Marquesas or Tahiti and came fully prepared to per-
manently settle outside of their home islands, bringing with them food plants
and domestic animals. Hawaiian legends, myths, and chants record, in story
form, the experiences of the original Native Hawaiians. They describe the pri-
mal natural elements encountered by the original Native Hawaiians and how
they adjusted to and coped with them. Often these original ancestors are per-
sonified as akua, küpua, and ‘aumakua. To them is ascribed the bestowal upon
the islands of special features and resources, including springs, streams, fish-
ponds, mountain formations, caves, offshore islets, craters, cinder cones, and
the varied rains and winds.
These deities or original ancestors made it possible for the Native Hawai-
ian people to adjust to the natural environment and resources of the islands
and to live and flourish as a society. For example, in Tales and Traditions of the
People of Old: Nä Mo‘olelo a ka Po‘e Kahiko, Samuel Kamakau describes how
Käne and Kanaloa were honored as Gods who opened springs of fresh water
for the people:

According to the mo‘olelo of Kane and Kanaloa, they were perhaps the
first who kept gods (‘o laua paha Na kahu akua mua) to come to Hawai‘i
nei, and because of their mana they were called gods. Kaho‘olawe was first
named Kanaloa for his having first come there by way of Ke-ala-i-kahiki.
From Kaho‘olawe the two went to Kahikinui, Maui, where they opened up
the fishpond of Kanaloa at Lua-la‘i-lua, and from them came the water of
Kou at Kaupo . . . They broke open rocks so that water would gush forth —
sweet, flowing water— at Wai-hee and at Kahakuloa on Maui, on Läna‘i,
at Waiakane in Punakou on Moloka‘i, and at Kawaihoa on O‘ahu.19

23
chapter one

The landscape was also made livable by the feats of Maui, who, according
to tradition, fished the islands up from the ocean with the magic fishhook the
constellation Manaiakalani. He is also said to have lifted the heavens high
above the earth so that humans could walk upright. He ensnared the sun in
order to lengthen the day. He also forced the ‘alae or mud hen to share the
secret of making fire so that humans could cook their food and have warmth
at night: “There may be seen the things left by Maui-akalana and other famous
things: the tapa-beating cave of Hina, the fishhook called Manai-a-ka-lani,
the snare for catching the sun, and the places where Maui’s adzes were made
and where he did his deeds.” 20
During the Developmental Period (600 to 1100) distinctively indigenous
Hawaiian cultural patterns and implements emerge. Throughout this period,
the inhabitants of Hawai‘i shared ancestry and heritage and developed an
indigenous culture and language uniquely adapted to the islands of Hawai‘i
and distinct from that of other Polynesian peoples. These indigenous Hawai-
ians developed a highly organized, self-sufficient subsistence social system and
extended sovereign control over the Hawaiian archipelago.
The social system was communal and organized around subsistence produc-
tion to sustain ‘ohana, the large extended multigenerational families. Hawaiian
spiritual beliefs, customs, and practices focused on maintaining harmonious
and nurturing relationships with the various life forces, elements, and beings
of nature as ancestral spirits who were honored as deities. Land and natural
resources were not privately owned. Instead, the Hawaiian people maintained
a communal stewardship over the land, the ocean, and all the natural resources
of the islands.
The küpuna provided leadership and guidance to the mäkua or adults who
performed most of the daily productive work of fishing, cultivation, and gath-
ering. Between the islands of Hawai‘i there was some variation of dialect and
names for plants, animals, rains, and winds. There were also variations in phys-
ical structures and cultural and art forms. Origin myths varied according to the
particular migration and genealogical line from which the families descended.
The prominence of akua and küpua also varied by island—for example, Pele
and her family of deities for Hawai‘i, Maui on the island which bears his name,
and various mo‘o or mythical dragonlike lizards on Moloka‘i. However, quali-
tatively, the language, culture, social system, and spiritual beliefs, customs, and
practices were shared among the inhabitants of the islands, and the origin of
the indigenous Hawaiian people’s sovereign nation can be traced to this era.
Dated to this period are basalt adzes and fishhooks that are distinctively

24
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

Hawaiian in form, as well as unique Hawaiian articles such as the ‘ulu maika
or stone bowling disc and the lei niho palaoa or tongue-shaped neck ornament.
By the end of this period, leeward areas of the islands were settled, indicating
an expanding population.
The Expansion period, between 1100 and 1650, includes a period of long
voyages between Hawai‘i and Tahiti up through approximately 1400. These
550 years are distinguished by geometric growth of the population, technolog-
ical innovation, intensification of production, and the emergence of a strati-
fied social system. Remnant structures and artifacts dating to this time sug-
gest that the leeward areas were extensively settled and cultivated during this
period. The chants, myths, and legends record the transpacific voyages of
great Polynesian chiefs and priests, such as the high priest Pa‘ao, the ali‘i nui
or high chiefs Mo‘ikeha and Mo‘ikeha’s sons Kiha and La‘amaikahiki. The
high priest Pa‘ao introduced a new religious system that used human sacrifice,
feathered images, and walled-in heiau or temples. Traditional chants and myths
describe how Pa‘ao introduced a system of ruling chiefs who appropriated
rule over the land through intermarriage, battles, and ritual sacrifice.
The ruling chiefs organized great public works projects that are still evi-
dent today. For example, ‘Umialïloa constructed taro terraces, irrigation net-
works, and heiau throughout Hawai‘i Island, including Ahu a ‘Umi on Huala-
lai. Kihaapi‘ilani oversaw the construction of the Ala Nui or trail around the
entire island of Maui and the Ke Ala a ka Püpü, a whiteshell pathway, on
Moloka‘i after he became ruler over all of the districts on these islands. The
construction of major fishponds, irrigation networks, and field cultivation
systems resulted in surpluses that sustained the stratification of Hawaiian soci-
ety into three basic classes — ali‘i (chiefs), kahuna (priests), and maka‘äinana
(commoners).
Despite these advances and the provision of food, barkcloth, and household
implements by the common people for the households of the chiefs, Hawaiian
society was predominantly a subsistence agricultural economy. There is no evi-
dence of a monetary system or commodity production, although a system of
barter in essential goods between fishermen, mountain dwellers, and taro cul-
tivators existed within the framework of the ‘ohana. Such exchange within the
‘ohana functioned as a sharing of what had been produced upon the ‘ili that
the ‘ohana held and worked upon in common:

Between households within the ohana there was constant sharing and
exchange of foods and of utilitarian articles and also of services, not in

25
chapter one

barter but as voluntary (though decidedly obligatory) giving. ‘Ohana living


inland (ko kula uka), raising taro, bananas, wauke (for tapa, or barkcloth,
making) and olona (for its fibre), and needing gourds, coconuts and marine
foods, would take a gift to some ‘ohana living near the shore (ko kula kai)
and in return would receive fish or whatever was needed. The fisherman
needing poi or awa would take fish, squid or lobster upland to a household
known to have taro, and would return with his kalo (taro) or paiai (hard
poi, the steamed and pounded taro corm) . . . In other words, it was the
‘ohana that constituted the community within which the economic life
moved.21

Under the ruling chiefs, land was not privately owned. The chiefly class
provided stewardship over the land and divided and redivided control over the
districts of the islands among themselves through war and succession. A sin-
gle chief controlled a major section of an island or a whole island on the basis
of his military power. Up until the time of Kamehameha I, however, no one
chief was ever paramount over all of the islands.22
The high chief divided his landholdings among lower-ranked chiefs called
konohiki. They functioned for the chief as supervisors over the people who
lived on the lands and cultivated them. The konohiki's tenure on the land was
dependent upon their benefactor, the chief. Konohiki were often related to the
chief and were allocated land in recognition of loyal or outstanding service to
him. However, unlike elsewhere in Polynesia, the konohiki were rarely related
to the maka‘äinana on the land under his supervision.23 Thus, the konohiki
represented the collective interest of the ali‘i class over the maka‘äinana as
well as the individual interest of his patron chief over the ahupua‘a.
The lands allocated to the konohiki were called ahupua‘a. Ahupua‘a bound-
aries coincided with the geographic features of a valley. They ran from the
mountain to the ocean, were watered by a stream, and included landscape fea-
tures such as mountain ridges or pu‘u and cinder hills.24
The ahupua‘a of the konohiki were further divided into strips of land called
‘ili, allocated by either the chief or the konohiki to the ‘ohana. These ‘ili either
extended continuously from the mountain to the ocean or were made up of
separate plots of land located in each of the distinct resource zones of the
ahupua‘a. The ‘ohana was afforded access to all the resources within the ahu-
pua‘a necessary for survival—vines, timber, thatch, and medicinal plants from
forested mountain areas; sloping land for sweet potatoes and crops that require
higher altitudes; low-lying lands irrigated by stream waters for taro and fresh

26
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

water; and shoreline, reef, and ocean areas for fish, limpids, crustaceans, and
seaweed, the principal sources of protein for Hawaiians.25
Ahupua‘a boundaries reflected the pattern of land use that had evolved as
the most efficient and beneficial to the ‘ohana throughout previous centuries.
The boundaries were adopted and instituted by the ali‘i and konohiki to delin-
eate units for the collection of tribute. These boundaries did not restrict access
by the ‘ohana to those natural resources needed for survival that were unavail-
able within their own ahupua‘a. For example, the adze is an essential tool for
the ‘ohana, yet the basalt used to hew adzes was not available within every
ahupua‘a. ‘Ohana could access the adze quarries even if they were located out-
side the ‘ohana's ahupua‘a. On the island of Moloka‘i, members of ‘ohana liv-
ing in the ahupua‘a of the windward valleys would annually reside for part of
the summer months in the ahupua‘a of Kaluako‘i to make adzes and to gather
and salt fish. The salted fish would sustain them during the winter months
when the ocean off their ahupua‘a was too rough for fishing. Evidence suggests
that the island of Kaho‘olawe was also a place of temporary residence for Maui
‘ohana to gather fish and to acquire the basalt needed for making adzes.
The tenure of the ‘ohana on the land was stable, unlike that of the ali‘i and
the konohiki. Two Hawaiian sayings illustrated this principle. The first was
“Ko luna pöhaku no ke ka‘a i lalo, ‘a‘ole hiki i ko lalo pöhaku ke ka‘a” (A stone
that is high up can roll down, but a stone that is down cannot roll).26 This
means that the chief and his retainers, including the konohiki, who were over
the people could be overthrown and lose their positions of influence. A chief
could be defeated in war and lose his lands. When a chief died and a new chief
succeeded him, the lands were redistributed, and the previous chief ’s kono-
hiki could be displaced. However, the common people who lived on the land
from the days of their ancestors were stable on the land. They were not dis-
placed when the chief or konohiki over them changed. They continued to live
on and cultivate the land of their ‘ili from one chief ’s rule to the next.
The second saying was “I ‘äina no ka ‘äina i ke ali‘i, ai waiwai no ka ‘äina i
ke kanaka” (The land remains the land because of the chiefs, and prosperity
comes to the land because of the common people).27 In other words, the chiefs
held the land, but the common people worked the land and made it valuable.
Though the tenure of the maka‘äinana was stable, they were not tied to the
land and did have the option to move away if they chose to. There is little evi-
dence, however, that moving off of the land of one’s birth was ever a common
practice.
The maka‘äinana produced all the necessities of life for their extended fam-

27
chapter one

ilies from the ‘ili that was allotted to them. In addition to cultivating their own
plots for the subsistence of their ‘ohana, the maka‘äinana were obligated to
cultivate plots of land set aside for the konohiki and chiefs. These were called
haku one and kö‘ele, respectively. The common people were also required to
provide the chiefs and konohiki with an annual ho‘okupu or tribute that
included food and all types of household needs, from tapa cloth and woven
mats to stone and wooden containers and implements, as well as feathers to
make the cloaks and helmets that were symbols of the ali‘i rank. In addition,
the maka‘äinana were obligated to provide labor service and products from the
land upon the request of the chief or konohiki. The ali‘i enjoyed full appropri-
ation rights over all that was produced upon his land grants; however, it was
the labor of the maka‘äinana that supported the entire society.
Maka‘äinana worked cooperatively and shared the fruits of the labor or
laulima. Most of this labor was done within the context of the ‘ohana as the
primary unit of production. The ‘ohana lived in dispersed clusters of house-
holds called kauhale on the ‘ili land granted to them. Within the ‘ohana there
was also cooperative enterprise and reciprocal exchange of labor service called
kökua. This was practiced in the undertaking of major projects such as the
chopping down, hewing out, and hauling of a log for a canoe or the construc-
tion and thatching of a house structure. These types of projects required the
labor of more people than made up one single ‘ohana. In addition, all the
‘ohana within an ahupua‘a could be organized to do massive public works proj-
ects under the supervision of the konohiki. This included construction and
maintenance of the irrigation systems and fishponds.
Although the chiefs and their konohiki had full appropriation rights over
the land and the people, in the main this was a system of mutual obligation
and benefit between the chiefs and the people. The chiefs controlled the
land and distributed it among the maka‘äinana. The chief was required to
manage and oversee the production on the land. He regulated the use of scarce
resources and apportioned these resources among the people according to
principles of fair use. Of these resources, water was the most valued, and the
chief assured that the irrigation system was properly maintained. He con-
served the resources of the land through restriction and replacement policies.
Of great spiritual significance, the chief was responsible for conducting the
proper rituals for the Gods who controlled nature. In return, the maka‘äinana
were obliged to provide labor service and products of the land to the chiefs
and konohiki.
Although Hawaiian tradition records cases of arbitrary, irresponsible, and

28
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

self-serving chiefs who abused the people, they were clearly exceptional cases
and were quickly replaced with responsible chiefs who cared for the well-being
of the people.28 The Hawaiian proverb “I ali‘i no ali‘i no Nä kanaka” (A chief
is a chief because of the people) reflects the Hawaiian attitude that the great-
ness of a chief was judged according to the welfare of the people under him.29
According to the Hawaiian historian David Malo, “In former times, before
Kamehameha, the chiefs took great care of their people. That was their appro-
priate business, to seek the comfort and welfare of the people, for a chief was
called great in proportion to the number of his people.” 30
From 1650 to 1795, the time of the Proto-Historic period, just prior to the
arrival and settlement of Europeans, Hawaiian society was highly stratified
under ruling chiefs who controlled whole islands and groups of islands and
vied for control as a paramount chief. Individual high chiefs continuously com-
peted to extend their control over more and more districts and islands through
marriage alliances, religious ritual, and military conquest. The archaeologist
Patrick V. Kirch provides an incisive description of this period:

With the development of highly sophisticated and intensive agricultural


and aquacultural production, an elaborate political hierarchy and land
tenure system, a religious ideology and ritual practice that included war
and fertility cults performed on massive stone temple platforms, and a
highly stratified social structure, the Proto-Historic Hawaiian culture can
be closely compared with other emergent forms of “state-level” societies
elsewhere in the world (for example, the Olmec culture of Mesoamerica,
the Pre-Dynastic Period of Egypt, or the Mississippian culture of North
America).31

To the extent that Hawaiian society had evolved into a socially and eco-
nomically stratified system by the eighteenth century, the responses of the
Hawaiian people to contact and change after 1778 were divergent and largely
influenced by the social and economic role the individual played in society.
The acceptance or rejection of Western culture was largely the prerogative of
the ruling class of ali‘i. The common people did not play a major role in
determining the political and economic future of Hawai‘i. They let the ali‘i
take the lead, while they struggled to survive the burden of contact—war, dis-
ease, famine, and the tragic widespread loss of beloved family, neighbors and
friends.
As discussed above, the political, economic, and social development that
came with contact, trade, and a plantation system were experienced unequally

29
chapter one

in the various districts of each of the islands. The case studies presented relate
the experience of the Native Hawaiians in the selected districts as these
broader developments unfolded. Below is an overview of the key developments
that affected the lives of the Native Hawaiian people from contact through
the end of the twentieth century.

Contact and Monarchy


In 1778, the year the English explorer Captain James Cook arrived in Hawai‘i,
the Native Hawaiian population was estimated at 400,000 to 800,000 inhabi-
tants.32 Beginning in 1785 Hawai‘i became a regular stopover in the fur trade
between America, Europe, and China. By 1810 Hawai‘i was an integral part
of the China trade route as a source of sandalwood. Gradually Hawai‘i was
pulled into the economic web of the worldwide market economy, causing far-
reaching and irreversible changes that devastated the Native Hawaiian people.
Periodically, the common people suffered famines that gripped the land as the
chiefs gave priority to meeting the needs of the fur and sandalwood traders.
According to Handy and Pukui, “As the desires of the chiefs and the pressure
of the trading captains grew, more and more people were put to the task, fewer
and fewer were left for the normal duties of everyday living; in many areas
planting and fishing virtually ceased, and for a season thereafter there would
be little harvested beyond the needs of the ali‘i and their konohiki (supervi-
sors). It was the people who went hungry.” 33
Exposure to Western continental diseases such as gonorrhea, syphilis, colds,
flu, dysentery, whooping cough, measles, and influenza killed thousands of
Hawaiians. David Malo recorded that in 1804 alone half of the Islands’ pop-
ulation died of ma‘i oku‘u, a disease that was either cholera or bubonic
plague.34
Kamehameha began a series of military campaigns to conquer all of the
Hawaiian Islands upon the death of his uncle, the chief Kalaniopu‘u, in 1782.
In 1790 Kamehameha acquired the Western ship Fair American and the ser-
vices of two Englishmen, John Young and Isaac Davis, to train his warriors in
the use of Western military technology. In 1795, after a four-year period of
peace during which Kamehameha trained his army, built his canoes, and
planted acres of food to feed his army of warriors, he launched his military
campaigns, which led to the conquest of Maui, Moloka‘i, Läna‘i, Kaho‘olawe,
and O‘ahu. In 1810 Kaua‘i also came under Kamehameha’s central authority
when the chief Kaumuali‘i agreed to become a tribute chief of Kamehameha

30
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

rather than try to fight off an invading force of Kamehameha’s war canoes.35
Thus, by 1810 King Kamehameha I, for the first time in the history of the
Hawaiian Islands, established a central absolute monarchy with sovereign rule
over all the islands.
Upon the death of King Kamehameha I in 1819, those chiefs who were
closely allied to him feared a rebellion from rival traditional chiefs. As a means
of undermining their rivals, the Council of Chiefs, under the leadership of
Mö‘ï Kamehameha II, Kuhina Nui Ka‘ahumanu, and High Chief Kalani-
moku, instituted the ‘Ai Noa or abolition of the state religion.36 By abolishing
the traditional chiefly religion under which rivals could claim rank, prestige,
and position, the Kamehameha chiefs consolidated political power under the
control of their monarchy.
Although Native Hawaiian religion ceased to have the official sanction of
the royal government, Hawaiian spiritual beliefs and customs continued to be
honored and practiced in most of the rural communities and settlements of
the kingdom. Families continued to honor their ‘aumakua. Traditional kahuna
lä‘au lapa‘au or herbal healers continued their healing practices using native
Hawaiian plants and spiritual healing arts. Family burial caves and lava tubes
continued to be cared for. The hula and chants continued to be taught, in dis-
tinctly private ways. Among the deities who continued to be actively honored,
worshipped, thought of, and respected, even to the present, were Pele and her
family of deities. Every eruption reinforced and validated her existence to her
descendants and new generations of followers.37
In 1820, the year following the ‘Ai Noa, American missionaries began to
settle Hawai‘i and convert Hawaiians to Christianity. In the same year com-
mercial whaling began to attract increasing numbers of foreign settlers, who
demanded rights of citizenship and private ownership of land.38 In 1839, nine-
teen years after King Kamehameha had established absolute rule over all of
the islands, his son, Kauikeaouli, King Kamehameha III, initiated a serious of
steps to set up a constitutional monarchy wherein the rights of the maka‘äi-
nana, distinct from those of the chiefs and of the king, were recognized. The
rights of foreigners who became naturalized citizens were also distinguished.

Native Hawaiian Responsibilities and Rights Are Vested in the Land


The first step in this process was the signing of the 1839 Bill of Rights. Up to
this point, foreigners were unable to become naturalized citizens. Thus, when
the law refers to “Nä Kanaka a pau,” or all of the people, it refers only to the

31
chapter one

native people of the Hawaiian Islands, the Native Hawaiians. It does not refer
to foreigners residing in Hawai‘i. The Bill of Rights recognized a division of
rights, the king being sovereign and distinct from the chiefs and the common
people. It guaranteed the protection of the rights of the people, the Native
Hawaiians, together with their lands, their building lots, and all their prop-
erty. The 1839 Bill of Rights states in part:

5. Ua hoomalu ia ke kino o Na Kanaka a pau, a me ko lakou aina, a me


ko lakou mau pa hale, a me ko lakou waiwai a pau; ke Malama lakou i
Na kanawai o ke aupuni, aole hoi e lawe ia kekahi mea, ke olelo ole ia kela
mea ma ke kanawai. O ke ali‘i e Hana i kekahi mea kue i keia Kumukana-
wai, e pau kona noho alii ana ma keia pae aina o Hawaii nei, ke hoomau
ia ma laila, pela Na kiaaina, a me Na luna a me Na konohiki a pau
[emphasis added].

5. Protection is hereby secured to the persons of all the people, together


with their lands, their building lots and all their property, while they con-
form to the laws of the kingdom, and nothing whatever shall be taken from
any individual except by express provision of the laws. Whatever chief shall
act perseveringly in violation of this Constitution, shall no longer remain a
chief of the Hawaiian archipelago, and the same shall be true of the gov-
ernors, officers and all land agents [emphasis added].39

The second step was the enactment of the 1840 constitution and the com-
pilation of laws for the Hawaiian kingdom. Under the constitution, executive,
legislative and judiciary branches of government were set up. The constitu-
tion included the same statement regarding protection of the people, their
lands, their building lots, and all of their property. In addition, the constitu-
tion clearly stated that although the lands from Hawai‘i to Ni‘ihau belonged
to the king, he did not own them as private property. Instead, the constitution
states that the king held the lands of the islands of Hawai‘i in common with
the chiefs and the people. Under this constitution, the responsibilities and
rights of the king, the chiefs, and the people were vested together, in com-
mon, in the land, at a time when Native Hawaiians were the only citizens of
the islands. Foreigners were not allowed to own land in Hawai‘i until a spe-
cial law was passed in 1850. The 1840 constitution states in part:

Eia ke ano o ka noho ana o Na alii a me ka hooponopono ana i ka aina.


O Kamehameha I, o ia ke poo o keia aupuni, a nona no Na aina a pau mai
Hawaii a Niihau, aole Na e nona ponoi, no Na kanaka no, a ma Na alii, a o

32
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

Kamehameha no ko lakou poo nana e olelo i ka aina. No laila, aohe mea


pono ma mua, aohe hoi mea pono i keia manawa ke hoolilo aku i kekahi
lihi iki o keia mau aina me ka ae ole o ka mea ia ia ka olelo o ke aupuni.

The origin of the present government, and system of polity, is as follows.


Kamehameha I, was the founder of the kingdom, and to him belonged the
land from Hawai‘i to Ni‘ihau, though it was not his own private property.
It belonged to the people, and the chiefs in common, of whom Kameha-
meha I was the head, and had the management of the landed property.
Wherefore, there was not formerly, and is not now any person who could
or can convey away the smallest portion of land without the consent of
the one who had, or has the direction of the kingdom.40

In 1846 Kamehameha III, the heir of Kamehameha I and ruling monarch


of the Hawaiian Islands, initiated a process to establish private property in the
Hawaiian Islands in response to the irrepressible demands of European and
American settlers and their respective governments. The king and the legis-
lature adopted “An Act to Organize the Executive Departments of the Hawai-
ian Islands,” which established a Board of Commissioners to Quiet Land
Titles. This act also included the “Principles Adopted by the Board of Com-
missioners to Quiet Land Titles in Their Adjudication of Claims Presented
to Them.” These principles served to guide the establishment of a system of
private property in Hawai‘i.
The introduction to the principles reaffirmed the joint responsibilities,
rights, and interests of the king, the chiefs, and the ‘ohana in the lands of the
Islands. From the time of High Chief Kamehameha Paiea and up until the
creation of a system of private property, the king and the chiefs held all
land as a sacred trust, and the indigenous ‘ohana continued their stewardship
responsibility and tenure over the lands of their ancestors. Under the new law,
members of the ‘ohana were now called hoa‘äina, or tenants of the land (lit-
erally translated, the term means “friend of the land”). The principles stated
in part:

O Na pono a pau i pili i ke Alii maluna o Na konohiki nui, a me Na mea


malalo o lakou, oia Na pono o Na konohiki nui maluna o Na hoaaina o
lakou, a me Na lopa a pau i noho i ko lakou aina. Nolaila, me he poe hui
la lakou, a ua pili ka aina ia lakou a pau . . .
Nolaila, he mea kupono maoli, a he mea pololei no hoi i ka haawi
ana o ke Alii i ke kuleana alodio, ke haawi i ke konohiki maluna, oia hoi ka

33
chapter one

mea i loaa mua ka aina Na ke Alii mai, no ka mea, i ka Hana Na pela, aole
i Hana ino ia Na konohiki, a me Na hoaaina malalo ona; ua hoomaluia
lakou e ke kanawai, e like ma ka wa mamua. He mea akaka loa hoi ka hiki
ole i ke Alii ka haawi aku i ke kuleana alodio ia hai, no ka mea, ina pela, ua
nele ke konohiki mua. Aka, ina loaa i ke konohiki mua kona aina ma ke ano
alodio, ma ke kuai, a ma ka haawi wale o ke Alii, ua mau no ke kuleana o
Na hoaaina, a me Na lopa, no ka mea aole nele kekahi mea e ae no ka
hoolilo ana o ka Moi i kona iho. Nolaila, o ke konohik i kuai me ke Alii a
loaa kona aina ma ke ano alodio, ua hiki ole ia ia ke pai i ka poe malalo
ona, e like ma ka hiki ole i ke Alii i keia manawa ke pai i ke konohiki.

The same rights which the King possessed over the superior landlords
and all under them the several grades of landlords possessed over their
inferiors, so that there was a joint ownership of the land; the King really
owning the allodium, and the person in whose hands he placed the land,
holding it in trust . . .
It seems natural then, and obviously just, that the King, in disposing
of the allodium, should offer it first to the superior lord, that is to the per-
son who originally received the land in trust from the King; since by doing
so, no injury is inflicted on any of the inferior lords or tenants, they being
protected by law in their rights as before; and most obviously the King
could not dispose of the allodium to any other person without infringing
on the rights of the superior lord. But even when such lord shall have
received an allodial title from the King by purchase or otherwise, the rights
of the tenants and sub-tenants must still remain unaffected, for no purchase,
even from the sovereign himself, can vitiate the rights of third parties. The
lord, therefore, who purchases the allodium, can no more seize upon the
rights of the tenants and dispossess them.41

In a later section, the principles clearly state that there are three classes of
persons who have vested rights in the lands of Hawai‘i — the government, the
landlord, and the tenant:

Ua akaka loa hoi, ekolu wale no mea kuleana ma ka aina hookahi. 1. O ke


Aupuni. 2. O Na konohiki. 3. O Na hoaaiana, a nolaila he mea nui ka
hoakaka i ka nui o ko kekahi kuleana, a me ko kekahi.
It being therefore fully established, that there are but three classes
of person sharing vested rights in the land,—1st, the government, 2nd, the
landlord, and 3rd, the tenant, it next becomes necessary to ascertain the
proportional rights of each.42

34
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

These principles, looked at together with the declaration in the constitu-


tion of 1840, actually describe how any one section of land in the Hawaiian
Islands is vested with multiple layers of responsibilities and rights. Native
Hawaiian ‘ohana, referred to as hoa‘äina in the principles, who had cultivated
their gardens and taro pond fields for generations and had gathered resources
from mauka to makai, from the mountain to the ocean, in their resident ahu-
pua‘a had one layer of vested interest, responsibilities, and rights in the lands
of the ahupua‘a. Over them, the landlord chief or konohiki responsible for the
overall management of the ahupua‘a and the well-being of the ‘ohana and
hoa‘äina who resided there also had a layer of vested interest, responsibilities,
and rights in each of gardens and taro pond fields that made up the ahupua‘a
and in the ahupua‘a as a whole.
Finally, King Kamehameha III was descended from King Kamehameha I,
who had conquered all of the chiefs and wrested control over each island.
Therefore, King Kamehameha III, ultimately, had inherited a vested layer of
interest, responsibilities, and rights over the individual gardens and taro pond

Figure 5 Kua‘âina have inherited the rights of the hoa‘âina. A Hawaiian taro farmer in Waipi‘o
Valley embodies the image of the kua‘âina featured in this book, who, like the ho‘âina, bent their
backs and worked and sweated in the taro patches and sweet potato fields and held that which
is precious and sacred in the culture in their care. 1974. Franco Salmoiraghi.

35
chapter one

fields in each of the ahupua‘a, as well as over each of the ahupua‘a in each of
the districts on each of the islands. The principles provided the following
example of how the multiple interests in any one tract of land might be
divided out:

Ina hookoia kela manao, e hiki no, ina he aina i ka lima o ke konohiki, a e
noho ana Na hoaaina, a ina like wale no ka aina a pau, hiki no ke mahele
maoli, i ekolu Apana like, a e haawi i ke konohiki i palapala alodio no kona
Apana, a pela no ko ka hoaaina, a koe hoi kekahi hapakolu i ke Alii i waiwai
no ke Aupuni.
According to this principle, a tract of land now in the hands of a
landlord and occupied by tenants, if all parts of it were equally valuable,
might be divided into three equal parts, and an allodial title to one then be
given to the lord, and the same title be given to the tenants of one-third,
and the other one-third would remain in the hands of the Kings, as his
proportional right.43

Therefore, the establishment of a private property system in Hawai‘i was


a process of dividing out the multiple layers of interest in each piece of land,
each ahupua‘a, and each island. The first step in this process of dividing out
multiple interests in the land was for the king and the landlords, or the chiefs
and konohiki, to distinguish their respective claims. The second step was for
the king and the chiefs to commute a portion of their respective claim to the
government of the Kingdom of Hawai‘i. The third step was for the common-
ers who lived on the lands to file for their portion of the lands claimed by the
king and the landlords or chiefs and konohiki.
Each of the mähele or divisions was in essence a series of quitclaim arrange-
ments between the king, on one hand, and a particular chief or konohiki, on
the other, relating to lands in which both had previously claimed an interest.
The summary is as follows:

1 That the King should retain all of his private lands as his personal and
individual property, subject only to the rights of tenants.
2 That one-third of the remaining lands be allocated to the Hawaiian
government; one-third to the chiefs or konohikis; and the remaining
one-third to the tenants or common people.
3 That the division between the chiefs or konohikis and the tenants might
be effected whenever either party required such a division, subject to
confirmation by the King and Privy Council.

36
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

4 That the tenants on the King's private lands were entitled to one-third
of the lands actually possessed and cultivated by them, and that such
division should be made whenever either the King or the tenant
required it.
5 That the divisions provided for in rules 2, 3, and 4 should be made
without any prejudice to any fee simple grant theretofore made by any
of the Hawaiian Kings.
6 That the chiefs or konohikis might satisfy the commutation due by
them, by the payment to the government of a sum equal to one-third
of the unimproved value of the lands awarded to them, or by conveying
to the government a one-third part of such lands.
7 That the lands allocated to Kamehameha III were to be recorded in the
same place and manner as all other allodial titles but that all lands allo-
cated to the Hawaiian government were to be recorded in a separate
book.44

The results of this Mähele were as follows:

Crown lands reserved for the monarchy 984,000 acres (23.8%)


Lands granted to 245 chiefs 1,619,000 acres (39.2%)
Government lands, distinct from Crown 1,523,000 acres (37%)

4,126,000 acres (100%) 45

All of these lands granted by the Board of Commissioners to the Crown,


the government, and the chiefs continued to be subject to the rights of the
hoa‘äina. The phrase “koe wale no ke kuleana o Na kanaka e noho ana ma ua
mau aina la,” which the government translated as “subject or reserved only to
the rights of the tenants,” is at the end of the declaration by the board estab-
lishing the Crown and government lands and appears on the grants of land
issued by the board.46
The establishment of a private property system in Hawai‘i transformed the
relationships and mutual responsibilities between the ali‘i and the ‘ohana, who
remained as hoa‘äina or tenants under the ali‘i. The rights and claims of the
ali‘i were addressed through Ka Mähele, under which 245 ali‘i were granted
a combined total of 1.6 million acres.47
The rights of the hoa‘äina were twofold. First, through February 14, 1848,
they had the right to file a claim against the lands apportioned to the chiefs
and konohiki for those lands which they cultivated and upon which they lived.
When the final land grants were made under the Kuleana Act of 1850, 8,205

37
chapter one

hoa‘äina received 28,600 acres, or 0.8 percent of all of the lands of Hawai‘i.
All of the land granted to the hoa‘äina could have fit into the island of Kaho-
‘olawe, which has 28,800 acres. Although all of the 29,221 adult males in
Hawai‘i in 1850 were eligible to make land claims, only 29 percent received
land; 71 percent remained landless.48
Several factors may have contributed to the low number of applications and
awards. Overall, the concept of private ownership of land was a totally foreign
notion. The Hawaiian language does not even have a word for private prop-
erty ownership of land. The word kuleana, which was used to translate the law,
refers to personal possessions such as clothing. Thus, many Hawaiians did not
appreciate or understand the importance of filing a land claim within the given
two-year period in order to continue living upon their ‘ili. And although the
law was published and posted in key locations, it was vaguely worded, using
foreign concepts that were not understood by the common people. Another
reason may have been that those who lived in out-of-the-way places did not
hear about the law or heard of it too late to file a claim. Furthermore, some of
the maka‘äinana were intimidated by the chiefs not to make land claims against
them. And finally, most of the maka‘äinana lived as farm tenants of the chiefs
and functioned outside the nexus of a cash economy. Therefore, the fee for
surveying the land, between $6 and $12, was beyond the reach of a majority
of the maka‘äinana.
In the campaign to set aside the Crown lands for Native Hawaiians to
homestead in 1921, Prince Jonah Kühiö Kalaniana‘ole focused on these lands
as the principal trust held by the Hawaiian monarchy for the Native Hawaiian
people. According to Kühiö, King Kamehameha III and the Council of Chiefs
had recognized that the common people had one-third interest in the lands of
Hawai‘i at the time of the Mähele. When the common people only received
0.8 percent of the land on an individual fee simple basis, the remaining portion
of the one-third interest of the common people in the of the lands were held
in trust by the monarchy as the Crown lands. Prince Kühiö explained this
point in an article he wrote for Mid-Pacific Magazine in February 1921:

The act creating the executive department contained a statute establishing


a board of royal commissioners to quiet land titles . . . This board decided
that there were but three classes of vested or original rights in land, which
were in the King or Government, the chiefs, and the common people, and
these three classes of interest were about equal in extent . . . The common
people, being left out in the division after being recognized as owners of a
third interest in the kingdom, believing that new methods had to be

38
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

adopted to place them in possession, assumed that these lands were being
held in trust by the crown for their benefit. However, the lands were not
reconveyed to the common people, and it [sic] was so held by each monarch
from the time of the division in 1848 to the time of the dethronement of
Queen Lili‘uokalani in 1893.49

What remains clear is that the king and government of the Kingdom of
Hawai‘i recognized that the Native Hawaiian landlords and common people
had vested interests, responsibilities, and rights in the land. The vested rights
of 245 chiefs and konohiki or landlords were transformed into fee-simple
ownership of a combined total of 1.6 million acres through the process of the
Mähele. The vested rights of more than three-fourths of the common people
were never transformed into fee-simple ownership. Kühiö, who was also
Hawai‘i’s delegate to the U.S. Congress, presented a compelling argument
that the people believed the land in which they held a vested interest contin-
ued to be held in trust by the monarchy for their benefit.
The second right of the hoa‘äina was provided by the king and the legisla-
ture in section 7 of the Kuleana Act, which granted to them their traditional
gathering rights, rights to drinking water and running water, and the right of
way, provided that permission was obtained from the landlords. Thereafter,
in 1851, the legislature amended section 7 of the Kuleana Act and deleted the
requirement that the hoa‘äina obtain the permission of the landlords in order
to exercise their traditional rights. Since 1851, the law has read as it now does
in Chapter 7, section 1, of the Hawai‘i Revised Statutes (HRS):

Where the landlords have obtained, or may hereafter obtain, allodial titles
to their lands, the people on each of their lands shall not be deprived of
the right to take firewood, house-timber, aho cord, thatch, or ki leaf, from
the land on which they live, for their own private use, but they shall not
have a right to take such articles to sell for profit. The people shall also
have a right to drinking water, and running water, and the right of way.
The springs of water, running water, and roads shall be free to all, on
all lands granted in fee simple; provided that this shall not be applicable
to wells and watercourses, which individuals have made for their own
use.50

In 1850, over the protests of Native Hawaiians, foreigners were given the
right to own land. From that point on foreigners, primarily Americans, con-
tinued to expand their interests, eventually controlling most of the land, sugar
plantations, banks, shipping, and commerce of the islands.51

39
chapter one

In the same year, new taxes were imposed upon the common Hawaiians:
a kuleana land tax, a $2 school tax for males, a 50-cent horse tax, a 25-cent
mule tax, and a $1 dog tax. Changes in the traditional land system and newly
imposed taxes forced greater numbers of Hawaiians to enter the work force
as wage laborers. They labored in the plantations as well as on ranches and in
small enterprises such as the gathering of pulu (tree fern fiber used to stuff
pillows and mattresses) and pepeiao akua (tree fungus), coffee growing, and
production of salt for export.52
Though the foundation for wage labor to develop into the dominant form
of labor was laid by the 1850s, it was the emergence of sugar as the primary
commodity around which the Hawaiian economy would be organized that
provided the impetus for the complete transformation of the Hawaiian social
system. The ‘ohana began to gradually change from the primary unit of work
and the context within which to make a livelihood to having no direct relation-
ship to the organization of work and production. Instead, the ‘ohana began to
serve as a source of refuge, comfort, and support to Hawaiian laborers who felt
overworked and socially alienated from their ‘ohana and family homesteads
when they labored on the plantations and in port towns. An 1873 article in the
Ka Nühou newspaper described the ‘ohana in just these terms:

The kanaka [Native Hawaiian] has no need to be very constant, and does
not suffer if he has neglected accumulation and aprovision [sic] for old
age. The bounty of the whole race affords a sure refuge to any bankrupt,
cripple, or pauper among their number. A kanaka can never become dead
broke and dread the poor house, because he will always be welcome to fish
and poi in any native hut that he enters. And so it is hard to get plantation
hands out of such easy going, spending, mutually helping people.53

Though coffee, rice, tobacco, cotton, livestock, and silk were experimen-
tally developed for large-scale commodity production and export, ultimately
sugar proved to be the most viable and profitable to produce on a large-scale
plantation basis.

Reciprocity, Overthrow, and Annexation


The critical turning point in the establishment of sugar as Hawai‘i’s principal
trade commodity was the U.S. Civil War. However, when the Civil War ended
and the United States imposed tariffs upon sugar imported from Hawai‘i, a
Reciprocity Treaty between Hawai‘i and the United States was negotiated,

40
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

which became effective in 1876. The reciprocity treaty stimulated the unprec-
edented growth of the sugar industry and Hawai‘i’s economy. Immense
amounts of capital were invested in land, labor, and technological develop-
ments. The profits derived from it were reinvested in further expansion of
sugar production. The phenomenal expansion of the sugar industry was under
the direction and for the benefit of the American and European factor-planter-
missionary elite. Native Hawaiian elite lacked the capital to invest and benefit
from the sugar industry, and common Native Hawaiians were displaced from
their traditional lands as the cultivation of sugar expanded. In 1893 U.S. Com-
missioner James Blount described the treaty in his report on the conditions
that led up to the illegal overthrow of the Hawaiian monarchy: “From it there
came to the islands an intoxicating increase of wealth, a new labor system, an
Asiatic population, and alienation between the native and white race, and
impoverishment of the former and enrichment of the latter, and the many
so-called revolutions, which are the foundation for the opinion that stable
government cannot be maintained.” 54
The Reciprocity Treaty effectively resulted in Hawai‘i’s becoming an eco-
nomic colony of the United States. When it expired in 1886, King Kaläkaua

Figure 6 In the 1890s Hawaiians still fished for subsistence off of Waikîkî, but this changed
at the turn of the century when tourism developed along its shores. 1890s. J. A. Gonsalves,
Hawaiian Historical Society.

41
chapter one

was reluctant to renew it. Not to be deprived of their economic wealth, Amer-
ican planter interests organized a coup d’état against King David Kaläkaua,
forcing him to sign the Bayonet Constitution, which took away his sovereign
powers as king and restricted the civil rights of Native Hawaiians. The cabinet
installed by the coup renewed the Reciprocity Treaty, and the king was com-
pelled to approve it. In 1889 eight men were killed, twelve wounded, and sev-
enty arrested in the Wilcox Rebellion, which attempted to restore the Hawai-
ian constitution. By 1890 non-Hawaiians controlled 96 percent of the sugar
industry, and Hawaiians were reduced to only 45 percent of the population
owing to the importation of Chinese, Japanese, and Portuguese immigrant
laborers by the sugar planters.
In 1893 the United States Minister assigned to the Kingdom of Hawai‘i,
John L. Stevens, conspired with a small group of non-Hawaiian residents of
the kingdom, including citizens of the United States, to overthrow the indig-
enous and lawful government of Hawai‘i.55 On January 16, 1893, U.S. mili-
tary forces invaded Hawai‘i, and the next day a provisional government was
declared. It was immediately recognized by the U.S. minister plenipotentiary
to Hawai‘i.
In 1898 the United States annexed Hawai‘i through the Newlands Joint
Resolution of Annexation without the consent of or any compensation to the
indigenous Hawaiian people or their sovereign government. Hawaiians were
thereby denied the mechanism for expression of their inherent sovereignty
through self-government and self-determination. They also lost control over
their national lands and ocean resources.56
Through the Newlands Joint Resolution of Annexation and the 1900
Organic Act, the Republic of Hawai‘i ceded to the United States government
1.8 million acres of land owned by the Crown and government of the origi-
nal Kingdom of Hawai‘i. The U.S. Congress exempted these lands from the
existing public land laws of the United States by mandating that the revenue
and proceeds from these lands be “used solely for the benefit of the inhabi-
tants of the Hawaiian Islands for education and other public purposes.” This
established a special trust relationship between the United States and the
inhabitants of Hawai‘i.57

Territorial Period
From 1900 through 1959 Hawai‘i was governed as a territory of the United
States. The official U.S. policy was to Americanize the multiethnic society of

42
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

the Hawaiian Islands, beginning with educating Hawaiian children through


the American public school system. Hawaiian and other non-English lan-
guages were banned as a medium of instruction; English was made the only
official language. An elite group of Americans who were the owners and man-
agers of what was called the Big Five factors had monopoly control over every
facet of Hawai‘i’s economy.58 They controlled the sugar plantations, shipping,
banking, and commerce.
In 1900 plantations harvested 289,544 tons of sugar from 66,773 acres of
Hawaiian land. By 1920 the plantations harvested 556,871 tons of sugar from
114,100 acres of Hawaiian land. This increased in 1930 to 930,627 tons of
sugar from 136,136 acres of land. The security of a stable American market
for Hawaiian sugar after annexation led the sugar planters to expand the num-
ber of acres planted in sugar and to invest in an infrastructure to accomplish
that. Of critical importance to the expansion of the industry was the develop-
ment of vast irrigation systems that carried millions of gallons of fresh water
from the wet windward sides of the islands to the dry leeward plains. On
O‘ahu, the planters constructed the Waiähole tunnel and ditch system from
1913 to 1916; ultimately, stream waters from Waihe‘e to Kahana on wind-
ward O‘ahu were diverted for the production of sugar on the dry ‘Ewa plains.
On Maui, additional ditch systems were constructed from 1903 to 1920 to
carry the waters of the Ko‘olau streams from Nähiku through Ha‘ikü over
into Pu‘unene. On Hawai‘i, the upper and lower Hämäkua ditch systems were
constructed in 1906 and 1910, respectively, and the Kohala ditch from 1905
to 1906.
The impact of these irrigation systems upon rural Hawaiian taro farmers
reverberated throughout the twentieth century. Cut off from the free flow of
stream waters into their lo‘i kalo or taro pond fields, many kua‘äina gave up
taro farming and moved into the city to find new livelihoods. Some of these
families stopped paying taxes on their rural lands when they moved into the
city and as a result eventually lost ownership of their ancestral lands through
adverse possession by plantations and ranches. In other areas, the long-term
impacts led to lowering of the water table, reduction of aquatic stream life and
nearshore marine life dependent on the infusion of fresh water into nearby
bays, and neglect of traditional irrigation networks.
The military was another force in the Americanization of the islands. In
1908 the United States began to develop Pu‘uloa into Pearl Harbor — dredg-
ing the channel and constructing a dry dock, barracks, warehouses, an ammu-
nition depot, a submarine base, a radio center, and a hospital. By 1930 the

43
chapter one

harbor was a major industrial base for the servicing of the U.S. Pacific Fleet.
At the same time, the army established bases on Hawaiian national lands
under its control at Lë‘ahi (Diamond Head) for Fort Ruger; at Waikiki for
Fort DeRussy; at Kalihi for Fort Shafter; and in Wahiawa and the Wai‘anae
mountains for Schofield Barracks. By 1941 the American naval presence at
Pearl Harbor was so massive that the Japanese attacked Hawai‘i, convinced
that this would cripple the American fleet in the Pacific. The military had
become the largest single source of income and employment in the Islands,
thereby guaranteeing the support of a major part of Hawai‘i’s local popu-
lation.
By 1900 the pure Native Hawaiian population had declined to 29,800, with
another 7,800 Hawaiians of mixed ancestry. Immigrant plantation workers
and their descendants made up the majority of the population, but under U.S.
law first-generation Asians were excluded from becoming naturalized citizens.
Thus, in the realm of politics, Native Hawaiians held the plurality of votes and
controlled the legislature and the delegate to U.S. Congress up through World
War II. It was not until after World War II that second-generation Asian
descendants matured to voting age and became a major political force in the
islands. Hawaiian leaders allied with the Big Five under the banner of the
Republican Party during the Territorial years. Thanks to political patronage,
Hawaiians held a majority of the government jobs and dominated certain pri-
vate-sector jobs such as cowboys on ranches, longshoremen on the docks, and
in the electric and telephone companies. In 1927 Hawaiians held 46 percent
of executive-appointed government positions, 55 percent of clerical and other
government jobs, and over half of the judgeships and elective offices. Through
1935 Hawaiians held almost one-third of the public service jobs and domi-
nated law enforcement, although they made up only 15 percent of the popu-
lation of the islands.59
Despite these obvious advantages, close to half the Hawaiian population
failed or refused to assimilate and mainstream into the developing economy.
Instead they remained in remote valleys and isolated rural pockets, providing
for their large extended families through subsistence farming and fishing.
During this period a major distinction internal to the Hawaiian community
evolved between the urban Hawaiians who assimilated and accommodated to
the socioeconomic system dominated by the American elite and the rural
Hawaiians or kua‘äina who remained in the backcountry areas and maintained
a traditional Hawaiian way of life.
During the Territorial period a “local” culture combining Native Hawai-

44
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

ian culture with the cultures of the various immigrant groups who settled in
Hawai‘i began to evolve. Most of the immigrants who were imported to work
on Hawai‘i’s plantations had been peasant farmers in their countries of ori-
gin. They shared with the majority of Hawaiians, who were planters and
fishermen, a reliance upon the land and its resources and a strong respect for
extended family relationships. Loyalty, respect, and caring for family elders
and the overall well-being of all family members were important values that
came to characterize “local” people. In rural plantation communities, the
immigrant workers shared the common experience of oppressive working con-
ditions, living in plantation camp housing, and being in constant debt to the
plantation store. Children of immigrant workers and Native Hawaiians alike
attended Hawai‘i’s public schools. There they were socialized by the American
school system. The children learned together, ate and shared meals together,
and communicated across cultural barriers in pidgin. They learned to hunt for
pigs and gather fruits in the forest. They caught fish or gathered marine or
aquatic life from common fishing grounds. The rate of intermarriage between
Hawaiians and immigrant groups, particularly the second and third genera-
tions, was very high.
World War II ushered in major changes in the social, economic, and polit-
ical life of the islands. Many Hawaiians left their rural enclaves to join the
service or to work in high-paying military jobs in Honolulu. The military
were also stationed in rural areas throughout the islands. The war experience
broadened the social horizons and raised the expectations and aspirations of
all Hawai‘i’s people for a higher standard of living. Raising the age for com-
pulsory education to eighteen also forced rural families out of the most
remote areas in order to comply with the law and send their children to inter-
mediate and high school. There was also a large exodus of people in search of
better job opportunities from Hawai‘i to the U.S. mainland.
The tidal wave of April 1, 1946, hit many rural coastal communities with a
force they were never able to recover from. The tidal wave took lives, smashed
houses, tore up roads, inundated taro fields and farms, destroyed fishpond
walls and breakwater walls, and scared many families into permanently mov-
ing out of their isolated low-lying rural peninsulas and valleys to live on
higher ground. Many coastal communities never rebuilt. A few coastal com-
munities became sparsely repopulated over the long course of the twentieth
century.
Labor unions successfully organized workers and gained collective bargain-
ing contracts on the docks and plantations, at utility companies, and in trans-

45
chapter one

Figure 7 The tsunami of April 1, 1946, hit coastal cultural kîpuka with a force from which
few recovered and contributed to the exodus of ‘ohana to urban centers. Downtown Hilo,
April 2, 1946. U.S. Army Signal Corps, Bishop Museum.

portation, hotels, restaurants, and the public sector. Leaders of the Japanese
community joined ranks with labor to reorganize the Democratic Party. The
Democratic Party defeated the Republican Party in 1954. Gradually Hawai-
ians were replaced in government jobs by Japanese. The Democratic Party led
the movement to gain statehood for Hawai‘i.

Statehood
Statehood stimulated unprecedented economic expansion in Hawai‘i. The
number of hotel rooms more than tripled, and the number of tourists
increased fivefold within the first ten years. Pineapple and sugar agribusiness
operations were phased out and moved to cheaper labor markets in Southeast
Asia. The prime agricultural lands that remained were developed into profit-
able subdivision, condominium, and resort developments. Left jobless, former
plantation and cannery workers had few employment options. They obtained
lower-paying and less stable jobs in the expanding tourist industry. An excerpt
from a social impact statement concerning the effects of a proposed freeway

46
NĀ KUA ‘Ā I NA and cultural K ĪPUK A

connecting rural O‘ahu to urban Honolulu offered an insight into the frustra-
tions and social pressures that Hawaiian and local people began to associate
with development:

Some long-time residents have the feeling that they are being dispossessed
of their traditional access to the beauties and bounties of nature around
them. Anxieties arise as open space is filled up by newcomers and the taxes
on land keep going up. Frustration is felt as the future character of their
shrinking world is being decided by landowners and developers, govern-
ment planners and elected officials in offices and meeting rooms far away.
And there is a problem of the carry over of these insecurities to the
younger generation. There are indications of social breakdown as reflected
in the rate of unemployment, the growing incidence of family separations,
the heavier welfare loads and the increase in juvenile delinquency and adult
crimes.60

Changes to the rural and agricultural areas concerned all of Hawai‘i's local
people, but especially the Hawaiian community, because of its traditional con-

Figure 8 This photo poignantly shows the tides of change that swept through the islands
when Hawai‘i became a territory of the United States and tourists began to visit the islands,
a trend that exploded after statehood. 1930s. Hawaiian Historical Society.

47
chapter one

centration in the rural pockets. American progress seemed to be overdevelop-


ing the islands and replacing the Native Hawaiian and local way of life. How-
ever, beginning in the 1970s, through an extraordinary convergence of events,
the island of Kaho‘olawe became the focal point of a major political movement
challenging American control of Hawai‘i. This movement became a catalyst
for a widespread Native Hawaiian cultural renaissance, which ultimately gal-
vanized into a movement for Native Hawaiian recognition and sovereignty.
Kaho‘olawe and the cultural renaissance that it spawned will be discussed in
a later chapter of this work, as an example of the role of the kua‘äina from the
cultural kïpuka in the regeneration of Native Hawaiian culture in the late
twentieth century. Moreover, the cultural renaissance highlighted the impor-
tance of the cultural kïpuka and helped reinforce the efforts of the kua‘äina to
protect their way of life from the assaults of proposed tourist and industrial-
ization projects in their communities.

Ha‘ina Ia Mai Ana Kapuana: Tell the Story


Here is my mo‘olelo of the Native Hawaiian kua‘äina. May their simple lives
in cultural kïpuka of the Hawaiian islands live on—not just in memory but in
determined efforts to protect and perpetuate their way of life and to have the
people of Hawai‘i attain and live lökähi.

48
 two 

Waipi‘o Mano Wai:


Waipi‘o, Source of Water and Life

During a spell of great drought, when a great famine was experienced all
over the lands from Hawaii to Kauai all the wet lands were parched and the
crops dried up on account of the drought, so nothing remained even in the
mountains. Waipio was the only land where the water had not dried up, and
it was the only land where food was in abundance; and the people from all
parts of Hawaii and as far as Maui came to this place for food.
—abraham fornander, Fornander Collection of Hawaiian Antiquities, vol. 4

AIPI ‘ O MANO WAI (Waipi‘o, source of water and life) is a popular

W saying about Waipi‘o because of its ability to sustain the people


of Hawai‘i and Maui during an early thirteenth-century drought
and enable them to survive. Located on the Hämäkua Coast of the island of
Hawai‘i, the remote, lush, and peaceful Hawaiian valley of Waipi‘o is rich in
natural resources, of which water is the most significant and abundant. For 650
years, from the time of ‘Umialïloa, around 1450 ce, through to the twenty-
first century, Waipi‘o has been renowned as one of the premier wetland taro
valleys of the Hawaiian Islands.
Picturesque Waipi‘o Valley lies across nine square miles, or 6,100 acres. At
the river’s mouth, along the ocean shore, Waipi‘o is a little more than three-
fourths of a mile wide, and its floor continues broad and nearly flat for about
three and a half miles inland. There the valley narrows to 600 feet wide and
then extends into a deep gorge for another five miles.1 At the mouth of the
valley, the walls of Waipi‘o are 1,000 to 1,300 feet high. Two miles inland
they are 2,000 to 2,300 feet high, and six miles inland the valley walls rise to
3,000 feet. Five tributary streams form ‘ili or smaller valleys that make up the
greater Waipi‘o Valley— Hi‘ilawe, Waimä, Kuiawa, Alakahi, and Kawainui.
These streams cascade down the sheer Waipi‘o cliffs in long, silvery, majestic
waterfalls. Hi‘ilawe, the largest and most prominent of these waterfalls, is

49
chapter two

Figure 9 Waipi‘o mano wai, source of water and life—past, present, and future. 1974.
Franco Salmoiraghi.

widely celebrated in two famous songs about Waipi‘o—“Hi‘ilawe” and “Wai-


pi‘o.” 2 Two large streams and several small ones traverse the valley floor,
depositing the fertile alluvial soil that makes the land ideal for the planting of
taro. The streams converge into one large river about half a mile inland before
emptying into the ocean. The primary natural hazards in the valley are high
winds, floods, and tsunamis.
Historically, water for all of the taro land was plentiful. It ran through
‘auwai or irrigation ditches and flumes into the taro pond fields. Frequent rains
kept the streams and ‘auwai running steadily. Only wetland taro was raised in
Waipi‘o. Numerous great old trees grow along the stream and at the base of
the hillsides. Overall, the valley is wet and cool.

A Traditional Center of Spiritual Power


The natural beauty and abundant natural resources of Waipi‘o are celebrated
in legend and oral tradition in accounts of the many spiritual deities who are
believed to have lived in the valley during their time on the earth. These spir-
itual forces are honored as ancestors of great, powerful, and prominent Hawai-
ian chiefs. The productive kua‘äina of Waipi‘o cultivated and sustained the
50
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

valley’s rich resources and enabled their chiefs to prevail against and ultimately
dominate the chiefs and people of the entire island of Hawai‘i.
Wäkea, God of the sky, mated with Papa, Goddess of the earth, and she
gave birth to the islands of Hawai‘i. In his old age, Wäkea is said to have
retired to Waipi‘o to live out the rest of his days.3 The godly Milu succeeded
Wäkea as chief in Waipi‘o. The brother Gods Käne and Kanaloa, who trav-
eled throughout the islands opening up freshwater springs to mix with their
‘awa, dwelt at Alakahi in Waipi‘o in company with lesser Gods.4 Maui, the
demigod associated with fishing throughout the islands, lifting the sky, slow-
ing the pace of the sun across the sky, and stealing the secret of fire making
from the Gods, obtained the Ipumakaniakamaumau (gourd of constant winds)
from the kahuna Kaleiiolu in Waipi‘o Valley in order to fly his kite. Maui also
fought with Käne and Kanaloa at Waipi‘o. According to legend, they killed
Maui in Waipi‘o when he tried to steal their bananas.5
Puapualenalena, a kupua or spirit in the form of a dog, lived in Waipi‘o. He
was a great thief who was caught stealing the sacred and restricted ‘awa of the
chief. In order to earn his pardon, Puapualenalena had to steal the magic conch
shell, Kihapü, from the spirits who lived above Waipi‘o and who constantly
disturbed the people of the valley by blowing on it at all hours of the night.6
The spirit dog returned Kihapü to the Waipi‘o chiefs. In the twentieth cen-
tury, the major landowner in Waipi‘o, the Bernice Pauahi Bishop Museum,
exhibited the Kihapü as part of its permanent Hawaiian artifact display.
Nanaue was a shark-man born in Waipi‘o Valley. He frequented the pool
at the base of Waipi‘o Falls and preyed upon his neighbors when they went
swimming or fishing in the ocean. When discovered as the shark who attacked
and ate the people of Waipi‘o, he escaped to Häna and then went on to Molo-
ka‘i, where he was finally killed at a place thenceforth named Pu‘u Mano or
Shark Hill.7
Lonoikamakahiki first met his beautiful wife, Kaikilani, in Waipi‘o Valley
beside the falls of Hi‘ilawe, where she dwelled in a breadfruit grove. Waipi‘o
became the resting place of the ka‘ai or woven sennit casket that held his
sacred remains.8

Migratory Chiefs
In the mo‘olelo of Waipi‘o, the migratory chief Olopana settled in the valley
and married Lu‘ukia. There they lived with Olopana's brother, Mo‘ikeha, and
their sister, Hainakolo. When the valley was devastated by a flood, Olopana
and Lu‘ukia set out for Kahiki. In some accounts, Olopana was accompanied
51
chapter two

by his brother, the chief Mo‘ikeha; in others, Mo‘ikeha was already living in
Kahiki.
Mo‘ikeha later returned to Hawai‘i but settled at Kapa‘a, Kaua‘i, where he
became the ruling chief of that island. Upon his death, his son Kila became
the ruling chief of Kaua‘i. His brothers, jealous of Kila's status, enticed him
away to Waipi‘o and abandoned him there. In Waipi‘o, Kila lived as a com-
moner until his rank was discovered by a priest of the Paka‘alana heiau. He
was adopted by Kunaka, then chief of Waipi‘o, who made him a konohiki or
land agent. As land agent he is credited with introducing the system under
which the common people were obliged to work a set number of days for the
chief. Later the evil deed of his brothers was exposed and punished. Kila,
however, remained in Waipi‘o, leaving only to journey to Tahiti with La‘a-
maikahiki in order to deposit the bones of his father, Mo‘ikeha.9
Pilikaeaea, the chief who, according to oral tradition, was brought by Pa‘ao
from Tahiti to rule Hawai‘i in the year 1090, first established his reign in Wai-
pi‘o Valley.10 Through intermarriage with descendants of the Nanaulu or Ulu
line of indigenous rulers he established the Pili line of rulers of Waipi‘o, from
whom Kamehameha I ultimately descended.

The Ruling Chiefs


In 1240 the ruling chief of Waipi‘o, Kaha‘imoelea, built up the royal resi-
dence in Waipi‘o.11 He was succeeded in 1270 by his son Kalaunuiohua, who
is famous for having waged successful wars of conquest against the chiefs of
Maui, Moloka‘i, and O‘ahu. He might have been the first ruling chief to unite
all of the islands under his control if he had not been defeated by the high
chief of Kaua‘i.12 About 1390 Kihanuilulumoku, grandfather of ‘Umialïloa,
ruled Waipi‘o.
‘Umialïloa is estimated to have ruled over Waipi‘o around 1450.13 At that
time, with the extensive taro lands of Waipi‘o as a resource base, ‘Umialïloa
conquered the entire island of Hawai‘i and ruled it as one chiefdom. He then
moved the capital of the island chiefdom away from Waipi‘o to Kailua-Kona.
‘Umialïloa constructed a system of heiau associated with astronomical obser-
vation and the maintenance of a lunar and solar calendar. ‘Umialïloa was one
of the greatest ruling chiefs of Hawai‘i, noted not only for his conquest of
Hawai‘i but also for his contributions to the social and economic development
of the island, especially in farming and fishing. He is credited with develop-
ing the systematic cultivation of taro in Waipi‘o, using terraces and irrigation

52
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

ditches and the expansion of paved roads linking Kona and Ka‘ü up across
the mountains.14 The Hawaiian historian Samuel Kamakau described his
contributions:

During ‘Umi-a-Lïloa’s reign he selected workers and set them in various


positions in the kingdom. He separated those of the chiefly class (papa
ali‘i), of the priestly class, of the readers of omens (papakilo), those skilled
in the affairs of the land (po‘e akamai o ka ‘äina), farmers, fishermen, canoe
builders, warriors, and other skilled artisans (po‘e pale ‘ike) in the work
they were best suited for; and each one applied himself to his own task.
‘Umi-a-Liloa did two things with his own hands, farming and fishing.
He built some large wet taro patches in Waipi‘o, and farming was done on
all the lands. Much of this was done in Kona. He was noted for his skill in
fishing and was called Pu‘ipu‘i a ka lawai‘a (a stalwart fisherman).15

In more recent history, Kamehameha I was raised in the districts of Kohala


and Waipi‘o from infancy to boyhood.
Associated with these high-ranking and sacred chiefs were the most sacred
heiau on the Hawai‘i island —Paka‘alana, Hale o Lïloa, Honua‘ula, Hokuwelo-
welo, Moa‘ula, and Kuahailo (Kuwahailo). Of these, the most revered was
Paka‘alana, and it was built before the time of Kila. At the time of Kila and
Kunaka it functioned as both a luakini heiau or sacrificial temple and a pu‘u-
honua or place of refuge. Lïloa built the luakini heiau Honua‘ula and dedi-
cated it to Ka‘ili. When Lïloa died, his son Hakau built the Hale o Lïloa or
mausoleum for his father’s ka‘ai, which held the ‘iwi or sacred remains of
Lïloa.16 Moa‘ula Heiau was built by Hakau but dedicated by his half brother
‘Umialïloa, who sacrificed Hakau as the first offering. Hokuwelowelo Heiau
is in the Läläkea ‘ili of Waipi‘o. It is said to have been built by the Gods and
was where the magic conch shell, Kihapü, was guarded until it was stolen by
the mischievous spirits and later returned by Puapualenalena. In the Ka‘au ‘ili
of Waipi‘o is the Kuahailo Heiau or Kuwahailo Heiau, which, according to
oral tradition, was built by Kuwahailo, one of the ancient Gods of Hawai‘i,
who named it after himself. He is believed to have lived in a cave in a small
side valley near the top of the south cliff.17

Wars of Kamehameha
Indicative of the continued prominence of Waipi‘o at the time of High Chief
Kamehameha are two fierce attacks upon Waipi‘o, as a central base of support

53
chapter two

for Kamehameha. In 1790 High Chief Kamehameha launched a war against


High Chief Kahekili of Maui with the assistance of canoes provided by High
Chief Keawema‘uhili of Hilo. While Kamehameha was at war on Maui, High
Chief Keouaküahu‘ula of Ka‘u decided to make war on Keawema‘uhili and to
ravage the lands of Kamehameha, including Waipi‘o. According to Samuel
Kamakau, Keouaküahu‘ula killed Keawema‘uhili at Alae in Hilopalikü and
then carried the war to Hämäkua. Kamakau provides an account of the ruth-
less attack on Waipi‘o by Keouaküahu‘ula: “He descended into Waipi‘o and
broke down the fishponds, drying up Läläkea, Muliwai, and all the other
ponds. He pulled up the taro of Waipi‘o, broke down the banks of the taro
patches, and robbed the people from Waipi‘o to Waimea.” 18
Learning of the attacks upon his people, High Chief Kamehameha sus-
pended his war against High Chief Kahekili and returned to Hawai‘i to retal-
iate against High Chief Keouaküahu‘ula and protect his people. High Chief
Kamehameha was able to rout Keouaküahu‘ula from Kohala, Waimea, and
Hämäkua. He then ordered his people from these districts to farm the land
and restore it to prosperity. Kamehameha was able to retain his rule over
Hämäkua, Kohala, and Kona, while Keouaküahu‘ula succeeded in expanding
his control beyond Ka‘u to the districts of Hilo, and Puna, lands previously
ruled over by High Chief Keawemauhili, whom Keouaküahu‘ula had slain.19
The second major attack upon Waipi‘o was a year later, in 1791. High
Chief Ka‘eokulani of Kaua‘i, in league with High Chief Kahekili of Maui,
invaded Waipi‘o from Hana. According to Samuel Kamakau, High Chief
Ka‘eokulani rallied his men from Kaua‘i, saying, “O you of Kauai! chiefs, sol-
diers, warriors, and dear little ones, be strong, be brave! Drink the water of
Waipi‘o and eat the taro of Kunaka!” Upon landing at Waipi‘o, Ka‘eokulani
viciously plundered the valley:

They landed at Waipi‘o. There Ka‘eokulani carried out his vow. He


wantonly destroyed everything in Waipi‘o. He overthrew the sacred
places and the tabu threshold of Lïloa; he set fire to Kahoukapu’s sacred
threshold of nioi wood and utterly destroyed all the places held sacred for
years by the people of Hawai‘i. No one before him, not even Keoua who
had passed through there the year before and destroyed the land and the
food had made such wanton destruction.20

In retaliation, Kamehameha engaged the forces of Ka‘eokulani and Kahe-


kili in the ocean off of the cliffs of Waimanu in what is called the Battle of
Kepüwaha‘ula the Red-Mouthed Cannon. The battle was named for Kameha-

54
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

meha’s Western cannons, which were manned by his British advisors Isaac
Davis and John Young.

It was because of the sound of the cannons (pü kuni ahi), the firing of the
muskets, and the flame flashing from these weapons that this battle was
called the Battle of the Red-Mouthed Cannon . . . After a hot battle
between the two sides, it was seen that most of the damage was done to
Ka‘eokulani and his companion, Kahekili. The greater part of his fleet was
sunk, and some of the men of those canoes swam to other canoes so that
Ka‘eokulani and Kahekili began to seek means of escaping being taken
prisoner by Kamehameha . . . Kamehameha’s forces had also received
damage. Some of his canoes had sunk, having been struck by cannon balls
from the other side.21

Warriors and canoes on both sides were lost. The battle ended indecisively,
and Ka‘eokulani and Kahekili managed to escape back to Häna.

Passing of the Sacred


Upon the death of King Kamehameha I, successor chiefs abolished the sacred
kapu or religious distinctions of the chiefs and the sanctity of their places of
worship. However, the sacred sites of Waipi‘o continued to function as cen-
ters of religious worship. William Ellis’s 1823 description of the valley esti-
mated that there were 1,325 kua‘äina who lived in 265 houses in the valley and
farmed and fished for their livelihoods. There was no developed trail into the
valley. Entry required sliding down the sides of the cliff and clinging to trees
and bushes on the way down. Ellis’s journal reveals a valley heavily cultivated
by the kua‘äina of Waipi‘o, who also continued to honor the sacred bones of
their ancestral chiefs.

The bottom of the valley was one continuous garden, cultivated with taro,
bananas, sugar cane, and other productions of the islands, all growing
luxuriantly. Several large ponds were also seen in different directions,
well stocked with excellent fish. A number of small villages, containing
from twenty to fifty houses each stood along the foot of the mountains,
at unequal distances on each side, and extended up the valley till project-
ing cliffs obstructed the view.22

According to Ellis, the Hale o Lïloa, where the ‘iwi of Lïloa were enshrined,
stood intact within an enclosure of the Paka‘alana heiau under a wide-spread-

55
chapter two

ing hala tree. In light of its status as an active and sacred place of worship, the
man in charge of the heiau did not allow Ellis to enter the Hale o Lïloa:

We tried, but could not gain admittance to the pahu tabu, or sacred
enclosure. We also endeavoured to obtain a sight of the bones of Riroa,
but the man who had charge of the house told us we must offer a hog
before we could be admitted; that Tamehameha, whenever he entered,
had always sent offerings; that Rihoriho, since he had become king, had
done the same, and that no one could be admitted on other conditions.23

Six years later, in 1829, High Chiefess Ka‘ahumanu, who had converted to
Christianity and was the regent for King Kamehameha III, made a special trip
to this heiau on a personal mission to end the persistent idolatrous worship of
the sacred chiefs of Waipi‘o.24 She removed 6 chiefly ka‘ai of deified Waipi‘o
chiefs that Native Hawaiians had continued to actively honor and worship.
Among these were ka‘ai that contained the iwi of the high chiefs Lïloa, Lono-
ikamakahiki, Kauhola, and Lole.25 Subsequently, High Chiefess Ka‘ahumanu
also journeyed to Honaunau to remove the ka‘ai of twenty-three deified chiefs
from the Hale o Keawe. The ka‘ai from Waipi‘o and Honaunau were taken
to the cave of Ho‘aiku in the great cliff at Ka‘awaloa and concealed.26
On the eve of the Mähele in 1847 the missionary Hiram Bingham wrote
of Waipi‘o’s “numerous garden-like plantations of bananas, sugar cane, pota-
toes, the cloth plant and the kalo, in different stages of advancement.” He
estimated the population at 1,200 to 1,500 Hawaiians.27
Despite the numerous kua‘äina living in Waipi‘o, only 102 land awards
totaling 374 acres were made under the 1848 Mähele and 1850 Kuleana Act in
Waipi‘o Valley. The bulk of the land, 5,800 acres, was claimed by and awarded
to Queen Hakaleleponi Kapakuhaili (Hazaleleponi) Kalama, wife of Kaui-
keaouli, King Kamehameha III.28 Most of the area of Waikoloa in Waipi‘o,
listed as an ahupua‘a with unspecified acreage, was awarded to William Pitt
Leleiohoku. Until her death on December 30, 1836, Leleiohoku was married
to High Chiefess Nahi‘ena‘ena, daughter of Kamehameha I with Ke‘öpüo-
lani, and sister of Kamehameha II and Kamehameha III. At the time of the
mähele, Leleiohoku was married to Princess Ruth Ke‘elikolani, a half sister of
Alexander Liholiho, Kamehameha IV, and Lot Kamehameha, Kamehameha
V. Princess Ruth inherited Leleiohoku’s lands upon his death in 1850. Ulti-
mately these lands became part of the Bernice Pauahi Bishop Estate. An
unspecified number of acres were granted to Mary Kaoanaeha as an ‘ili‘äina in
the area within Waipi‘o called Kalaokui. Kaoanaeha was the daughter of the

56
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

English officer John Young, who had played a critical role in the victory of
Kamehameha in the Battle of Kepüwaha‘ula. Her mother, also named Kao-
anaeha, was the daughter of Keli‘imaika‘i, a brother of King Kamehameha I.29
The 374 acres awarded to the kua‘äina of Waipi‘o ranged in size from half an
acre to fifteen acres, with an average size of three and half acres per award.
The forces of change that transformed the Hawaiian Islands as a whole also
penetrated the rural isolation that otherwise buffered Waipi‘o. The flu epi-
demic of 1850 had a devastating impact on the Hawaiian families in the val-

Map 1 The Coulter map of the population of Hawai‘i Island in 1853 indicates a population of
750 in Waipi‘o Valley. The Puna section indicates a population of 2,850 persons in 1853.
o = 50 persons. Source: Coulter, Population and Utilization of Land and Sea, p. 28.

57
chapter two

ley. Though the number of persons who succumbed to the flu is not recorded,
the impact can be measured by the food shortages that resulted from the
debilitating effects of the epidemic upon the kua‘äina of Waipi‘o. The Wai-
mea district, which depended upon Waipi‘o for food, experienced shortages
because regular deliveries of taro from Waipi‘o were interrupted.30 As shown
in map 1, John W. Coulter estimated that the population in Waipi‘o had
declined to 750 in 1853.
Six years later, in 1859, the congregational church was completed in Wai-
pi‘o. However, no mission station was ever established in the valley itself. The
ministers who served the people of the valley were based in Waimea and in
Kohala and made periodic visits to Waipi‘o. The lack of a permanent mission
station in Waipi‘o meant Christianity presented less of a direct threat to the
traditional Native Hawaiian spiritual beliefs and practices that persisted
among the kua‘äina who were born and raised in the valley.31
The kua‘äina of Waipi‘o continued to actively raise taro for markets
through Hawai‘i Island. The Pacific Commercial Advertiser of May 5, 1866,
reported that pa‘i‘ai or hard, pounded, undiluted taro would be shipped rather
than transported overland to Kawaihae, Kona, Ka‘u, and Hilo by the farmers
of Waipi‘o. According to the article, the people of Waipi‘o had formed an
association and declared that they would not take their poi to the market but
would instead have the market come to them via supply ships. The writer
observed that Waipi‘o “probably furnishes more of the Hawaiian staff of life
than any equal area of land on the islands.” 32
In 1867 Father Bond visited Waipi‘o and commented on the trade that had
developed between Waipi‘o and the surrounding districts. He was singularly
impressed with the enterprising qualities of the 640 Waipi‘o Hawaiians whom
he estimated to reside in the valley.

The rich bottom of Waipio Valley affords inexhaustible quantities of food


for Hilo and South Kohala, as well as for Hämäkua markets; and the people,
accordingly, have a thrifty, well-to-do appearance; and what was particularly
gratifying to my own mind, was the open wide-awake countenances which
met our gaze in the congregation; and not less the spirit and bearing of
independence, which made the people seem to me more like our own than
had any previous gathering with whom we have met.33

In 1880 George Bowser observed only thirty to forty houses in Waipi‘o Val-
ley and estimated the population to be reduced to 150. As a result, he noted,

58
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

only half of the valley floor was under cultivation in both taro and rice. The
following year Charles Reed Bishop bought land in the ahupua‘a of Waipi‘o
from Hazaleleponi Kalama at a public auction. Fifteen years later, in 1896,
Bishop donated these lands to the Bernice Pauahi Bishop Museum.34
Despite the social changes affecting the island of Hawai‘i, when Isabella
Bird stood at the overlook at the top of the pali (cliff ) and looked down into
Waipi‘o in 1886, she observed a productive and thriving Native Hawaiian
community of farmers and fishermen.

I should think the valley is not more than three miles long, and it is walled
in by high inaccessible mountains. It is in fact, a gulch on a vastly enlarged
scale. The prospect below us was very charming, a fertile region perfectly
level, protected from the sea by sandhills, watered by a winding stream,
and bright with fishponds, meadow lands, kalo patches, orange and coffee
groves, figs, breadfruit and [coconut] palms. There were a number of grass-
houses, and a native church with a spire, and another up the valley.35

Chinese Influence and the Turn of the Twentieth Century


Between 1880 and 1920 the valley’s Native Hawaiian population was aug-
mented by the influx of Chinese whose contracts with the neighboring plan-
tations at Honoka‘a and Kukuihaele had expired. Many intermarried with the
daughters of the kua‘äina living in the valley. They subleased Bishop Museum
lands in order to cultivate rice on a commercial basis — first from Samuel
Parker and later from the Hawaii Ditch Company (later called the Hawaiian
Irrigation Company). Through 1901 the Bishop Museum lands were leased
to Parker for $7,500 a year. He grazed cattle in the upper valley and subleased
portions of the lower valley to rice and taro farmers.36 The population is esti-
mated to have grown from 150 to over 1,000 during this period. Though the
increase was largely due to new Chinese residents, because of intermarriage
there were always more full and part Hawaiians than full-blooded Chinese in
Waipi‘o Valley.37
On January 1, 1901, a new twenty-one-year lease was negotiated with
Parker that contained a clause requiring that outside offers be made for sur-
plus water and water power from the lands. Parker was given sixty days to meet
the offer or otherwise lose the lease. In 1904 the Hawaii Ditch Company made
an offer to the Bishop Museum for Waipi‘o’s water that Parker did not care
to meet. By mutual agreement, the lease was transferred to the Hawaii Ditch

59
chapter two

Company for the remainder of its term at an annual rent of $3,000. The
Hawaii Ditch Company was primarily interested in the upper valley, where it
developed an extensive irrigation system.
In 1907 the Hämäkua Ditch Company constructed the Upper Hämäkua
Ditch, which diverted water from the Kawainui, Alakahi, and Ko‘iawe streams
above Waipi‘o. Upon its completion, the upper ditch was able to deliver 15
million gallons per day to various sugar mills along the Hämäkua coast. Soon
thereafter, the Hawaiian Irrigation Company began to construct the twenty-
five-mile Lower Hämäkua Ditch to supply water for cane fluming, mill oper-
ations, and domestic water systems along the Hämäkua Coast outside of Wai-
pi‘o Valley. When it was completed in 1910 it was able to carry 30 million
gallons of water a day from three of the five streams that feed Waipi‘o itself
—Kawainui, Alakahi, Ko‘iawe.
The company continued to lease out the lower valley for taro and rice cul-
tivation and the grazing of cattle. In 1915, for example, it had 151 acres sub-
leased for taro and rice at $25 per acre per year, from which it grossed
$3,775.38
By 1906 there was one store in Waipi‘o Valley, and it was owned by a
Chinese family. There were two churches, a Congregational and a Catholic
church, in the valley.39 Seven Chinese rice growers grew large quantities of
rice in the valley with twelve to thirty helpers apiece. At harvest time, tin cans
were strung out over all the paddies on long cords attached to a central tower.
When the cords were pulled, the tin cans banged throughout the valley and
scared the birds away.40
Hawaiians never grew rice. They continued to cultivate taro for both home
consumption and outside markets throughout this period in the upper part of
the valley, where the streams flowed swift and cold. In the lower valley, where
the streams meandered and slowed in their flow, the water was warmer. These
lands were marginal for growing taro and were therefore leased to Chinese for
rice cultivation.41
David Makaoi, who was born in 1904 and lived in Waipi‘o Valley from the
age of two until he was fifteen, provides a description of how the valley looked
in the early twentieth century:

Well, it was beautiful in my time. We had rice patches near the ocean. And
then, taro patches further up in the valley. Hardly any trees on the floor.
Course there were guava groves along the edges of the stream, you know.
Especially rice patches at a certain time of the year when they were just

60
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

young, nice and green. You see green patches down there along near the
beach. And then, before harvest time they were all yellow. So, it was one
color one time, and another the next time . . . further up were the taro
patches . . . taro farms . . . most were operated by the Chinese—for the
Chinese poi factories. They had their own taro farms, to grind their poi.
But once in awhile, if they were short of taro, they came to buy some
from Hawaiians.42

In 1917, near the end of World War I, the price of rice soared, and Wai-
pi‘o rice farmers prospered. However, disaster struck in the form of floods in
April and December 1918. The April flood damaged 25 percent of the rice
crop. The December flood destroyed 50 percent of the rice crop.43 After 1918
the cost of labor increased, and the prices for taro and rice began to decline.
One by one the Chinese gave up their rice fields, and the growers and their
helpers moved out of the cottages they had built upon their leased lands. As
more and more of the land was abandoned, the Hawaiian Irrigation Company
steadily lost its income from the subleases.44
The last grass houses built in Waipi‘o were dedicated by Native Hawaiian
families in 1920 in accordance with traditional Native Hawaiian rituals. Once
the framework was completed, the ‘ohana baked a pig together with taro and
lauloa in an imu or underground oven in the center of the house. When the
food was cooked and the imu opened, the owner and everyone who was to live
in the house sat down and ate the pig and vegetables. They saved the bones
and threw them into the stream for their ‘aumakua. After this ceremony, the
house would be thatched, and the door opening would be made by the
kahuna.45
In 1922 the Bishop Museum signed a new thirty-two-year lease with the
Hawaiian Irrigation Company for an annual rent of $3,000 or, at the museum’s
discretion, at the rate of 5 percent of the market value of products obtained
from the land. The president and manager of Hawaiian Irrigation claimed to
be losing money on the lease because of low lease rents that, combined, totaled
only $150 per annum. The cost of re-leasing and collecting rents from their
Chinese tenants, paying the land tax, and keeping up roads, fences, and build-
ings was higher that the amount of money collected from the lease rents.46
The last rice crop raised in Waipi‘o Valley was harvested in 1927. Rice pro-
duced in Waipi‘o was simply too expensive compared to that imported from
California. In 1928 it cost $4.50 to produce a 100-pound bag of rice in Wai-
pi‘o, while a 100-pound bag of rice grown in California could be purchased

61
chapter two

in Hilo for $2.98.47 With the phasing out of rice production in Waipi‘o, fam-
ilies moved out of the valley, and the population decreased to 271 in 1930. Of
this number, 178 people, or 66 percent, were Hawaiian. Of the remaining 34
percent, 63 persons were Chinese, 9 were Japanese, 2 were Korean, and 19
were Filipino, according to the 1930 census.48

Waipi‘o Valley, circa 1931


Circa 1931 Waipi‘o typified the remote and secluded cultural kïpuka
described by Andrew Lind and studied by E. S. Craighill Handy, Elizabeth
Handy, and Mary Kawena Pukui. The small community was predominantly
Native Hawaiian, and the pace of life was slower than that of people exposed
to the social and economic changes that were transforming the rest of the
island.
In 1931 the anthropologist Stella Jones conducted fieldwork in Waipi‘o
Valley. The unedited and unpublished transcription of her interviews with the
Native Hawaiian residents provide a unique snapshot image of the life of the
kua‘äina in Waipi‘o Valley in 1931. This account is augmented by oral history
interviews conducted in 1978 by the Ethnic Studies Oral History Project of
people who grew up in Waipi‘o Valley around 1931. Together, these shared
memories provide insight into the lives of the kua‘äina of Waipi‘o and are rep-
resentative of the lives of kua‘äina in other similarly situated cultural kïpuka
throughout the islands of Hawai‘i at that point in time. For this reason, I will
explore the life of the kua‘äina in Waipi‘o in 1931 through the shared mo‘o-
lelo of those who directly experienced it.

Community
According to Jones, only 200 people lived in Waipi‘o Valley in 1931. Of this
number, there were twenty children who were under school age, fifty-four
children who were enrolled in one of the six grades of the one-room elemen-
tary school in the valley, and ten children who walked up the pali to the inter-
mediate school at Kukuihaele. A number of children had been hänai or infor-
mally adopted into Hawaiian families upon the death of one of their parents.
Most of the young adults went to intermediate school in Kukuihaele. Those
who pursued high school had to live in Hilo, returning home only during
school breaks. However, once they completed their schooling, the majority of
high school graduates sought jobs outside the valley and did not return to live

62
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

and to raise their families. Usually, one of the children remained behind to
care for the kuleana and to look after the old folks. In one Hawaiian family
with three boys, the oldest boy worked as a draftsman in Honolulu; the sec-
ond boy worked as a machinist in Honolulu; and the third boy remained in
the valley and farmed with his parents.
The Waipi‘o school had six grades in two rooms, three grades per room.
David Makaoi described the classes: “The teacher cannot talk to all at once.
So, what he would do is to talk to one class in one subject and then go on
down the line. Each time he takes the next class, the previous class would have
to do their written assignment in class . . . Of course if you finish your home-
work before time, your classwork, you can tune in on what they’re saying.” 49
The teachers were very strict. For any little offense the teacher would whack
the students with a yardstick. Some teachers would hit the students’ heads
against the blackboard for making a mistake.
School started at eight o’clock in the morning. The children would get up
at seven o’clock and eat a simple breakfast of taro, or maybe nothing at all.
Recess was short, and lunch, was just a half hour. Lunches were not served, so
the children were expected to bring their meal to school. Some of the Hawai-
ian children would take fish and poi; others just took a couple of crackers;
many simply skipped the meal. If they were fortunate to earn some money by
picking lü‘au leaves for the poi factory, serenading at Christmas or New Year’s
time, or selling frogs or fish, the children would buy something from the store,
usually bread with butter and jelly. Half a loaf of bread with jelly sold for only
five cents. Sometimes wild beans similar to lima beans were boiled, pounded
into a paste, and wrapped up for lunch. The Chinese children usually had
money to buy their meal. Often the children would exchange lunches.50
Subjects taught in the school in Waipi‘o ranged from English, history, and
geography to math and hygiene. They did not teach physical education. The
children did work in the school garden, however, once a week. Each child
could choose his or her own little patch to raise something such as peanuts or
beans.
After school, the children had chores to perform. They would have to haul
fresh water in buckets for the family and also gather and chop wood for cook-
ing. Sometimes they could earn anywhere from two to five cents by hauling
water and carrying wood for the Chinese. The children also fed any animals
the family raised, such as pigs and chickens. Some of the children also per-
formed chores in the morning before school started. They might have to mix

63
chapter two

the poi for the family’s breakfast or lunch and make the coffee. Those with
cows would milk them in the morning and cut grass for them in the after-
noon.51
When their chores were done on weekends and during the summer vaca-
tion, the children swam and fished in the streams and mountain pools and
played marbles, hide and seek, bean bag, and other such games. Sometimes
they challenged each other to see who could catch the most fish. They also
constructed their own toys: they would make a ball by stuffing rags inside a
larger bag and sewing it up or make cars and trucks out of wood, spools, and
soda-water corks.52
David Makaoi provided a vivid description of his childhood days in
Waipi‘o:

Once in awhile they had maybe a baseball game between some Waipio boys
and somebody from Kukuihaele. Not every time though, but just once in
a rare while. So, the rest of the time, you just had to find your own recre-
ation. Fishing, swimming, things like that. Just on your own . . . I went
fishing. I had so many places to fish, too . . . We had to work most of the
time so if we had a free time—of course, we went to swim at Nenewe Falls,
or even in the streams. And of course when holidays came I enjoyed play-
ing music.53

With the exception of the ten kuleana, most homes were built upon land
leased from the Bishop Museum. Most of the Hawaiians’ homes were small
and very old; the larger homes were owned by Chinese. The Hawaiian dwell-
ings usually included a lanai or extended porch that was thatched with braided
coconut leaves. For a typical family there was one bedroom for the parents; a
living room, where the children slept; and a small lanai. The parents had a koa
wood bed, but the children simply put a sheet or mat on the floor and slept
with a blanket and a pillow.54
Only one thatched house remained in the valley in 1931. It was thatched
with sugarcane and occupied only intermittently by a Japanese fisherman who
lived there while watching over the mullet pond.
Most of the Hawaiian households raised a few pigs, which were penned
near the homesites. They also raised a few chickens on taro peelings from the
poi factory. The chickens were usually killed for a feast at Christmas, and a
pig was usually cooked in an imu for a New Year’s lü‘au. Pigs were also cooked
for lü‘au on other special occasions, especially for a baby lü‘au, held on a child’s
first birthday.

64
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

The few Chinese had garden plots around their homes where they grew
peanuts, lettuce, cabbage, onions, parsley, bitter melons, and tomatoes. They
also raised Muscovy ducks and pigeons for home consumption.

Land Tenure
In 1931 there were ten Native Hawaiian–owned kuleana, estimated to total
100 acres. The rest of the taro land in the valley was primarily owned by one
major landowner, the Bishop Museum and leased to the Hämäkua Ditch Com-
pany. The Honoka‘a Sugar Company managed the land for the ditch com-
pany. The annual rental for taro land ran from $20 a patch to $50 an acre per
year, which the Hawaiians considered to be inconsistent. For example, a Mr.
Kanekoa rented two acres for $52 a year, while a Mr. Kaohemoku paid $40
for only three-quarters of an acre.55 About 100 acres of land in the valley were
not planted in taro. Waipi‘o residents felt that this deliberate practice enabled
the landowner to maintain a high price for the cultivated taro lands.56
At one time the Hämäkua Ditch Company let the uncultivated land be
used for pasturage at a cost of 50 cents a head per month. However, by 1931
the company had subleased all of the pasture land to one Chinese immigrant,
Mock Chew, who charged $1 a head per month for all cattle, horses, mules,
and donkeys over six months old that were not penned up.57 Most of the
horses and mules were owned by Chinese and kept fenced in upon pasture
land that they leased directly from the Hämäkua Ditch Company. Several
Native Hawaiian families owned approximately 200 head of free-roaming cat-
tle, which sometimes wandered outside of the pasture area and damaged the
taro. They did not have enough grass to produce good milk. Most of the
Hawaiians in the valley felt that Mock Chew should not have been allowed
the privilege of charging a fee for pasturage on these cattle when he made no
effort to cultivate good pasture land or to provide fences.

Livelihoods
Most of the kua‘äina worked for themselves. They lived primarily on poi and
freshwater fish, occasionally eating meat or seafood. They cultivated and har-
vested their own taro, which was cooked and pounded for home consumption.
They also made their own fishing nets and caught the fish themselves.
Twice a week, the Hawaiian taro farmers would take boiled taro to the
Chinese poi factory owner to sell. When the Hawaiians wanted to earn cash,
they would work for the Chinese in the taro patches, at the factory, or by
hauling taro out of the valley. A few of the residents in 1931, such as Sam

65
chapter two

Kaaekuahiwi, principal of Waipi‘o School, John Thomas, and Mr. Wilson,


were educators in the Waipi‘o school and at Kukuihaele. There were no pro-
fessional fishermen in the valley. Sometimes, when the Chinese wanted to earn
extra money, they would fish out in the ocean and sell their catch outside the
valley. Native Hawaiians who fished the deep sea did so only for home
consumption.

Taro Production: The Major Livelihood in Waipi‘o


From cleaning the land to scraping freshly steamed corms, taro was the main-
stay of the Hawaiians of Waipi‘o. Most of the Hawaiians did not plant taro
for sale. Instead, they usually planted only enough taro for home consump-
tion.58 The men maintained the lo‘i kalo — cleaning the banks, planting, weed-
ing, and harvesting. The families would cook enough taro for the week. The
women and children in the family peeled the taro with coconut shells. Then
the men would pound the poi on a board, using a stone to mash the cooked
taro with water. Some families used a big board about five feet long, and the
men would sit at either end and pound the poi. The taro was usually mashed
into pa‘i‘ai, wrapped in ti leaves, and placed in a wooden barrel. At each meal,

Figure 10 The Mock Chew family, descendants of Hawaiian and Chinese farmers in Waipi‘o
Valley, sit on the front steps that, before the 1946 tsunami, led into their family home. 1974.
Franco Salmoiraghi.

66
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

the amount of pa‘i‘ai needed would be taken out of the barrel and mixed with
water to eat as poi.59
Sam Kaaekuahiwi described the steps involved in cultivating taro.60 First,
the land was cleared of weeds and grass. Then the soil was plowed and left dry
for one week. After one week, water was let in and the lo‘i lay flooded for
three weeks. At that point, the farmers harrowed the land using a horse, a
mule, or an ox. When the land had been properly prepared, the taro shoots
were planted in twos in alternate rows, two feet apart. The lo‘i was kept wet
the entire time for the first three to four months, the water regulated so that
it just covered the roots and so the correct amount of cool running water
flowed through the patch. The patch was also weeded periodically. Fertilizer
was not used at this time.61 One man could take care of about 6 acres of taro
alone. With one worker he could usually farm ten acres.
Different varieties of taro matured at different rates. The api‘i taro, called
the short taro, matured in 8 or 9 months. It included the ‘apu wai and lehua
varieties of taros. The uaua taro, called long taro, took eighteen months to
mature. It included the uaua elele, uaua piko, and uaua molino varieties. These
taros could even be left in the field up to two or two and a half years until a
market for it could be found, provided the water was properly regulated at the
right temperature. The land for these types of long taro did not have to lie
fallow between crops as it did for the short taro.
When pulling taro for sale, the farmer usually broke off the tops with his
hands, threw the tubers into gunnysacks, and loaded the sacks onto a mule.
Some farmers, however, tied the plat ends together and threw them across the
mule’s back. The best tops were retained for seeding. A bag of taro averaged
100 pounds. The average yield was 200 to 300 bags to the acre. Good taro lost
little weight in being made into poi. The water added usually made up for the
weight lost by peeling the tuber. Poor taro, however, did lose weight.
The mules carried the taro to one of two poi factories in Waipi‘o, both
owned by Chinese —Akioka and Chang. At the factory the taro was washed
by flumed water and placed in the steamer, where it was steamed for three
hours. After it was cooked, the taro was placed in a tub and hosed down. Then
the tubers were scraped by women, one by one. They sat around the tub on
wooden boxes and used coconut shells to scrape the taro clean. One woman
could clean six tubers in a minute. The women earned 50 cents a day and
worked from 6:00 a.m. to 10:30 or 11:00 a.m.
The troughs used to receive poi at the two poi factories were immense old
poi boards formerly used by two men sitting on either end. Before the intro-

67
Figure 11 Uncle Joe Kala working in his lo‘i in Waipi‘o, where taro cultivation has
been a way of life for generations. 1974. Franco Salmoiraghi.
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

duction of engines to grind the poi, around 1911, the Chinese employed old
Hawaiians to pound poi for them. At that time there were three factories, and
the elderly Hawaiians were paid $1.50 a day for pounding.
In 1931 the poi factory that had the big engine made $80 worth of poi on
Tuesdays and $120 worth of poi on Thursdays. The little engine made about
$80 worth of poi once a week.
Farmers also sent 100-pound bags of taro out of the valley to other parts
of the island. One farmer sent a thousand 100-pound bags out of the valley,
every month. Another farmer sent eighty 100-pound bags to Kona every
month. Others sent around two hundred 100-pound bags. When the bags left
Waipi‘o they usually weighed 105 pounds, to allow for 5 pounds’ water loss
by evaporation during transport before it reached its final destination.
Hilo controlled the market for taro and poi on the island. During World
War I, there was a shortage of taro. The price of taro soared, and the poi fac-
tories bought up all the taro produced in Waipi‘o and sent it to Hilo and
Kamuela. The Hawaiians even sold the taro that they normally consumed at
home because they could make so much money, eating breadfruit and buying
flour to make up for the lack of taro. During this time, the Hawaiians had a
lot of money. One year after the war ended, though, prices went back down
to their normal level.62
In 1925 taro sold for $2.50 a bag in Hilo. At one point it had sold for as
much as $3.75 a bag. However, after 1925 Maui farmers started to ship their
taro to Hilo via steamship. They undersold the Waipi‘o farmers by selling at
$1.25 a bag. To compete, Waipi‘o taro sold for as low as 50 cents a bag in the
field or 75 cents a bag pulled.

Subsistence Resources
Waipi‘o Valley provided its people with an abundance of natural resources for
their day-to-day sustenance. They did not have to seek high-wage jobs or ven-
ture outside of the valley for their subsistence. Taro was the staple food of the
Hawaiians. It was boiled or fried and eaten whole or else pounded into poi.
Poi was pounded on an ongoing basis in the valley with pounders that had been
fashioned by their fathers and grandfathers out of beach stones. Breadfruit,
which was available on a seasonal basis, was usually cooked in the imu and
pounded into poi. It was eaten by itself or mixed together with taro poi.
The leaves of the taro, especially the young lü‘au or taro shoots, were cut
up and boiled for greens with pork, chicken, or jerked beef.63 They were also
gathered along the streams and in common land areas. The aquatic life of the

69
chapter two

mountain streams was the main protein source for the Hawaiians of Waipi‘o
in 1931. ‘O‘opu fish and ‘öpae or shrimp were caught in the streams and taken
home, cleaned, and either salted for later consumption or cooked right away
by boiling or frying with salt or shoyu. Sometimes the ‘o‘opu were baked in ti
leaves. Freshwater shrimp were sometimes eaten raw. David Makaoi provided
a graphic description of the preparation of fish for eating: “I didn’t tire of eat-
ing ‘o‘opu. That’s one thing I found out. Could cook it and maybe roast it
sometimes, or boil it. The gravy tastes nice, it's fat. It’s tasty. And fry. And
sometimes cook in the ti leaf too over the fire. Gives a different flavor. So, you
can cook it in many ways. So, you never get tired of eating fish. So we had fish
most of the time.” 64
The people of Waipi‘o made fish traps with the ‘ie‘ie vine to catch o‘opu
and shrimp. The trap would be placed in the stream, facing a rock, and the
‘o‘opu and shrimp would be scared into the trap. During floods, the ‘o‘opu
washed downstream and could be easily caught with traps placed at strategic
points in the stream. In the ocean, the ‘o‘opu spawned and hatched their
young, called hinana. The Hawaiians would go to the mouth of the river and
scoop hinana up with the nets. Many escaped and swam upstream, where they
lived in the pools deep in the back of the valley and in the stream along the
muddy banks. In dry weather, the ‘o‘opu could be caught in the upper pools
of the valley. Slapping the water would scare them into holes in the sides of
the pool. Then one could stick a hand into the hole and gently grab the fish
by the head.65
Mullet found in the lower valley were usually cooked in ti leaves and fried
or boiled. Püpü and escargot-like shellfish were raised in the taro patches.
They were usually left to stand for three days in a kerosene tin and then
cleaned, after which they could be heated on a stone until they cracked and
were either removed with a needle or sucked out. Sometimes they were boiled
with garlic, black beans, and salt and then served.
Freshwater fish were plentiful in Waipi‘o. George Farm described how he
once took the father of a friend to Waipi‘o Valley during the depression of
1929–30. He drove by car to the top of the valley and then rode by horse
down into the valley. He was very impressed with the abundance of fish:

Lot of fish, Waipio. You don’t have to go hunt for it. The fish in the taro
patches, fish in the ditches, fish in the streams, all over the place. And then
one night when we slept there, the first night, they had big storm down
there. Lot of rain. And the streams got flooded over, eh? and in the morn-

70
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

ing, about 7 o’clock, I see young children with the bucket. They running
in the bushes, they pick up fish. Fill up the buckets, going home, empty
and bring it back again. Filling up fish whole morning, you know. So I
stopped them one time. I say, “say, where you get that fish from? Where
you buy that fish?” “No, no. No buy. Plenty in the bushes.” You see, that
much fish in Waipio Valley.66

Limu or seaweed was a very important source of vitamins and minerals to


the Hawaiians. They would gather it from the ocean, clean it, salt it, and eat
it raw with poi. It was never cooked. The ocean provided limu kohu, limu hulu-
huluwaina, ‘ele‘ele, lipu‘upu‘u, mane‘one‘o, and lipahe‘e varieties of seaweed. Limu
kohu, the hardest to get, also lasted the longest, two to three months, while the
others usually lasted a week. Salt was not made in Waipi‘o but was obtained
from Kawaihae, Hawai‘i, or directly from Honolulu by the Chinese store-
keeper.
The valley was also rich in fruits, trees, and plants for eating and healing.
It had ‘öhi‘a‘ai or mountain apple, papaya, banana, avocado, breadfruit, guava,
and mango trees. Coffee also grew wild in the valley. It was gathered, dried,
and roasted in a skillet for home use. There were also lemon trees, orange
trees, coconuts, hau trees, noni fruit, ‘awa, and chili peppers. Kukui trees were
plentiful, and its nuts were used for garnishing food, for medicine, and to
make small torches. Wauke and olonä, traditionally important for making tapa
and cordage, also grew wild in the valley, although they were not actively used
in 1931. Pandanus trees grew in the valley; its leaves were gathered for weav-
ing. Even as late as 1931 women wove mats, hats, purses, handbags, and head-
bands from the pandanus leaves.67
Wild goats in the mountains were sometimes shot for food. Both Hawai-
ians and Chinese owned 200 head of cattle which grazed in the valley. They
were periodically slaughtered for home consumption by families in the valley.
Every two months or more someone would plan to slaughter a cow and go
throughout the valley to all the families to let them know. He would butcher
it in the pasture, and the meat would be placed in piles on fern or water lily
leaves. Everyone who wanted to buy meat would come to the site and select
the piles, which sold for $1, $2, or $5, that they wanted to buy. The bones,
which were used for soup, were given away free. They did not use a scale.
People brought their own bags to take home their meat. The meat was usu-
ally salted, dried, and kept in crocks or barrels until it was ready to be eaten,
because there were no iceboxes or freezers.68

71
chapter two

Aside from these occasions, meat was rarely eaten in the valley, and the
nearest butcher was at the town of Honoka‘a. One would have to make an
order ahead of time, and he would arrange for the meat to be delivered to
Kukuihaele on the specified day. David Makaoi explained why meat was not
often eaten:

Hardly any meat. ’Cause only once in awhile, when somebody kills a cow
for the whole valley. Then they get to buy beef, so many pounds to take
home. And salt it most of the time. And that’s pipikaula [smoked beef ].
That’s the only way to preserve it and still be nice for eating. So that’s
why I enjoy pipikaula, nowdays, here, because it has a good flavor.69

Interaction with the Market Economy


Despite the abundance of natural resources for food, certain items — sugar,
flour, salt, shoyu, rice, canned salmon, canned sardines, salt salmon, cod fish,
dried shrimp, corned beef, cooking oil, matches, kerosene oil, soap, beer, wine,
sake, and clothing—were purchased in stores in Waipi‘o, Kukuihaele, and
Honoka‘a.70 The store in Waipi‘o also sold baked goods, such as bread with
butter and jelly, doughnuts, and cakes, which were popular with the school
children.
In general, the people of Waipi‘o only purchased the bare necessities.
According to David Makaoi:

We didn’t buy too much, though, because for the rest of the things we were
self-supporting. We just say, “Why spend money? Get up and make our
own.” We were independent. We didn’t need much cash for things in the
store. So, that’s why they said, “Why live in town? You have to buy every-
thing with cash.” (Laughs) In the country, you don’t need much cash.71

Those Hawaiians who sold taro to the store could purchase goods on credit.
Those who sold the taro to another broker or who did not sell taro had to
purchase goods on a cash basis.72

Transportation
At one time, there was an entrance to Waipi‘o from Waimea and Kawaihae by
means of a trail in the rear of the valley. There was also a lower road beneath
the pali. The narrow trail connecting the valley with the outside world was
eventually widened into a horse path to facilitate the transport of taro and rice
by land.

72
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

In 1904 residents from Waipi‘o and Waimanu wrote a letter to Governor


George R. Carter asking the territorial government to spend money appro-
priated by the 1903 legislature for a new road down the Waipi‘o pali. They
also asked for government lands between Waipi‘o and Waimanu to be turned
into homesteads for poor American citizens of the valley.73 The road was never
constructed, and in 1931 there was still only a horse-and-mule trail leading
into and out of the valley.
To send taro to Hilo, the farmers and poi factories paid 25 cents per 250-
pound bag for it to be transported by mule to Kukuihaele. They then paid 30
cents a bag to transport the taro by truck to Pa‘auilo, where it was loaded onto
a train for Hilo. In 1931 the Chinese owned ninety-five mules, and the
Hawaiians owned about ten mules or horses.
By 1930 public opinion in Waipi‘o had turned against the construction of
a road. They feared that it would bring others into the valley to plant and
would lower the price of taro and increase the rent on the land. They also did
not like the prospect of having prison labor in the district while the road was
being constructed.
Everything that went in and out of the valley by land traveled on a horse,
a mule, or humans. Sometimes two men would haul produce on a pole slung
between their shoulders.74 About half a dozen outrigger canoes were owned
by Hawaiians in the lower valley. However, they were primarily used for fish-
ing, not transportation.

Lifestyle, Beliefs, Customs, and Practices


Fannie Hauanio Duldulao, who was born in Waipi‘o in 1911 into a Hawaiian
family that raised taro in Waipi‘o, probably expressed the feelings of every-
one who was born and raised there:

Well, I love the place because I was born and raised there until I grew up
— a great-grandmother today. And then the feelings of the place is actually
really warm feelings . . . it’s a valley of aloha and then full of love. And
when I was born and raised there, I had everything that I can think of
without spending money. Everything was really from the land, what we
raised. The valley. Like taro and everything.75

There were no major crimes reported in Waipi‘o prior to World War II.
It was a valley that modern developments had bypassed. Occasionally a politi-
cian would go into the valley at election time and make a speech on the

73
chapter two

veranda of the store. But for the most part, it was too difficult to access the
valley. And the residents preferred it that way. A road would mean the end of
the abundance of resources that made life in a valley as isolated as Waipi‘o not
only possible, but also desirable.76
In 1931 there were three churches with active memberships—a Protestant
church, a Mormon church, and a Chinese temple. There were two cemeter-
ies—a Hawaiian one and a Chinese one. A Catholic church had been built and
established at one time but was no longer in use by 1931.77
The Christians held one joint service in a church built by the missionary
Titus Coan. In general, twenty-five adults who were Mormon, Christian Sci-
entist, and Catholic participated in the services. However, there were only
seven members who contributed to the support of the church, including the
salary of a part-time minister.
The Mormon church had the strongest following. According to Sam Kaae-
kuahiwi, many of the members had joined the church because the Mormons
were reputed to be good at healing. Several had joined after apparently being
cured of a sickness by prayers offered by the Mormon congregation for their
recovery. The ex-sheriff was Mormon. He had two wives and two families.

Figure 12 Fannie Duldulao and Romualdo Duldulao at their Waipi‘o kuleana. 1978.
Franco Salmoiraghi.

74
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

Although the Waipi‘o Hawaiians participated in Christian services, they


still maintained their traditional beliefs and practices. In particular, most of the
families in Waipi‘o acknowledged and retained a respect for their ‘aumakua,
their family guardians. One of the residents in 1931 shared an experience she
had with her family’s ‘aumakua, the shark:

One of my ‘aumakuas was a God-damn fool. He caused the sickness in my


ankles. I went to kahuna after kahuna, but always the sickness came back.
For two years I was unable to stand on my feet. Then a kahuna told me
what to do. He made me go at 12 o’clock at night to the water, and he
dipped me 5 times in the water to please one ‘aumakua so that he would
help me (against the malevolent ‘aumakua), and when I came back, dogs
mustn’t pass, chickens mustn’t pass, man mustn’t pass or get in my way.
After that I went back and dipped again 5 times and took awa root as a gift
to the ‘aumakuas in the sea, the shark ‘aumakuas.78

Many of the families in Waipi‘o traced their ancestry to a shark ‘aumakua


who frequented the offshore waters in the form of a big shark. They were told
not to look at that shark if they saw it in the ocean, or they would be blinded.
They were not to treat the shark badly. The ‘aumakua of others was black
with a red mouth. If they saw that shark when they went out fishing, it was
considered a good sign, and they would proceed with confidence that help was
available if needed.79
Other families traced their ancestry to turtles. If a family member related
to the turtle was in trouble out on the ocean, a huge turtle, large enough for
a human to ride on, would appear to rescue that person. Fannie Hauanio Dul-
dulao related how her mother believed in the turtle ‘aumakua and told the
children, “They know if the family need help. It’s a surprise, you can see the
turtle float.” 80 A few of the families were related to the mo‘o or lizard ‘auma-
kua. They were careful not to treat lizards roughly.81
The residents who served as informants to Stella Jones in 1931 remem-
bered when each house had an ‘aumakua shrine. They could recall how peo-
ple used to pray to the ‘aumakua. Their parents had shared stories with them
about how kahuna ‘anä‘anä or sorcery could be used to break someone’s back,
and about how a boy who was in love with a girl who did not reciprocate could
go to a kahuna or sorcerer who could employ manulele or love sorcery to make
the girl care for the boy.
With regard to fishing, Sam Kaaekuahiwi remembered how the old people
had practiced the kü‘ula rituals. He explained the kü‘ula ritual as he under-

75
chapter two

stood it: “Maybe it is an idol placed on the shore at a certain spot. They would
bury the image there . . . Sometimes they would have a wooden god of kau-
wila wood, it was in the form of a tapa beater. They would tie a cord to it and
drop it in the water, praying to the akua and leaving it overnight. When fish-
ing the next morning they would be successful.” 82
In addition, one man was designated as the kilo, or observer of fish. He
would go along the path and look for the fish, then tell the fisherman what
kind of fish he saw and where they were. The fisherman, following his direc-
tions, would take his net, cast it, and surround the fish.
Hawaiian taro farmers in Waipi‘o planted according to the moon phase.
They carefully observed the moon and could predict when the full moon or
mahealani phase would occur. Taro and most other Hawaiian plants grew best
if planted on the night of the full moon.83
Hawaiian rituals and practices were also important in healing. Honohono
grass was pounded and used for healing cuts and sores. Laukahi leaves were
also pounded and used to cure sores. The pistil of the laukahi was boiled into
a tea and drunk to heal cancer. Pöpolo berries and leaves were used to cure
colds. Fannie Hauanio Duldulao described the use of pöpolo:

Well, they call that popolo. You know those leafy things, eh? They have
that small little purple seed. And then, they just pound that and then
squeeze it. And then you drink the juice. Even the shoots, that’s how they
pick up. Like this, you just pick ’em up, you know, so much, one handful.
And then you go home, put in the cheesecloth or whatever, as long it’s
clean. Then you pound that. But some, they put in the ti leaf then they
heat ’em up. But mom said it’s better to have fresh from the plant. And
that is good for cold, too; especially when babies start to cough.84

The bark of the mountain lehua could also be boiled into a tea and drunk
to heal colds. Burns and boils could be healed with the application of certain
leafy native Hawaiian plants. Hawaiian herbs could even be used to cure bro-
ken, sprained, or dislocated bones.85

Waipi‘o: “‘Âina Aloha”


Waipi‘o, secluded abode of akua and legendary chiefs, persisted into the mid-
dle of the twentieth century as a heartland of traditional Hawaiian culture and
lifestyle. While living in distinct households and cultivating individual patches
of taro for their own families, Waipi‘o Hawaiians shared in the abundance of

76
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

natural resources in the valley. Together, they respected and cared for the
streams and irrigation system that fed their individual lo‘i kalo. Together,
they respected the balance of natural resources, taking only what was neces-
sary for their family’s daily sustenance and respecting the reproductive cycles
of the aquatic and plant life in the valley. Their children played together,
attended school together, and matured to adulthood together. The people in
the valley occasionally joined together to celebrate life’s great events — a baby
lü‘au, a birthday, a wedding. They also celebrated Christmas and New Year’s
together.
David Makaoi described a memorable Chinese New Year’s when he was in
the eighth grade:

On New Year’s Eve and on Chinese New Year’s Eve, several musicians
would serenade the Hawaiian and Chinese homes, respectively. I was one
of them, beginning with the sixth grade, until my high school years. I
welcomed that opportunity to earn some good money. We usually sere-
naded in groups of two or more musicians. On one occasion, however, on
Chinese New Year’s, I performed all alone with my adopted uncle, Kamaka,

Figure 13 Waipi‘o Hawaiians shared in the abundance of natural resources in the valley.
1978. Franco Salmoiraghi.

77
chapter two

as chaperone and lamp holder. This happened during my eighth grade year.
I netted $5, though I had to split 50 /50 with my adopted Hawaiian uncle.
In those days, $1 a day was the average laborer’s daily wage. The next day
was a school day. Without any sleep I went to Kukuihaele School the next
morning. With nickels and dimes jingling in my pockets, I did not fall
asleep in school . . . I was walking on “cloud nine.” 86

One New Year’s Eve, David Makaoi joined Sam Li‘a, the famous song-
writer and musician from Kukuihaele, in serenading the Hawaiian families in
Waipi‘o Valley. Sam played the violin, David played the ukulele, a third man
played the banjo, and the last played the guitar. At the end of the night, the
group split the proceeds for the night evenly.87 They composed songs that
told of their love for Waipi‘o and committed to memory their exploits and
experiences.
During Prohibition, a number of families made okolehao from ti root
cooked in an imu. They would make about ten gallons at a time, which would
last about three months. It was not sold, but made solely for home consump-
tion and shared with visitors and friends.88
The Hawaiian way of life continued to thrive among the households and
taro patches nestled at the foot of the towering cliffs of Waipi‘o Valley. All
along the rest of the Hämäkua Coast, though, the sugar plantation economy
dynamically transformed the landscape and social life of Hawaiian communi-
ties and villages. Throughout 1931 Waipi‘o remained an enclave of Hawaiian
people and a Hawaiian way of life, despite the influx of Chinese rice planters
and their wage workers. The lifestyle of the Waipi‘o Hawaiians starkly con-
trasted with the lifestyle of urban Hawaiians. The persistence of Hawaiians in
Waipi‘o Valley provided an important continuity for Native Hawaiians to
their heritage. The Waipi‘o Hawaiians demonstrated well into the twentieth
century the knowledge, skills, and resourcefulness of their people, who main-
tained harmonious and respectful relations to the land and with each other.

Forces of Change
Waipi‘o Valley continued to be a Native Hawaiian enclave through the end of
World War II. In 1941 a flood ravaged the kua‘äina of the valley. Through-
out World War II, young men were attracted out of Waipi‘o into the military
and the more lucrative military jobs in Honolulu, especially at Pearl Harbor.
In 1945 the Waipi‘o Valley Grammar School closed. Finally, a second natu-

78
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

ral disaster dramatically altered the lives of the Waipi‘o kua‘äina. The 1946
tidal wave, which reached a height of fifty-five feet, inundated the valley and
destroyed most of the homes and taro patches. Many of the dislocated fami-
lies left Waipi‘o, never to return. Only fifteen to twenty kua‘äina continued to
live in the valley.89
The 1946 tidal wave devastated the valley. As a result, many kua‘äina moved
to nearby outside rural towns along the Hämäkua coast. Nevertheless, two
years later Paradise of the Pacific magazine characterized Waipi‘o as a remote
valley where old Hawaiians spent their days catching fish, growing taro, and
pounding it into poi, and their nights retelling stories of the past.90
By 1954 the resident population in the valley increased to between thirty
and forty people. There were three Hawaiians living in the valley, and the
rest were Filipinos living in makeshift shanties. At the same time, there were
170 kuleana owners, whose land holdings amounted to 309.4 acres. Bishop
Museum owned another 534.6 acres, and the government controlled 66 acres.91
Not all of the kua‘äina who had land or who farmed in Waipi‘o lived in the
valley. Most lived nearby in surrounding rural communities above the valley,
especially in Kukuihaele. Combined, the kua‘äina still cultivated 300 acres of
taro, 2–3 acres of lotus, less than 2 acres of water chestnuts, 11 acres of maca-
damia nuts, and 15 acres of coffee in the valley.
A 1958 flood again destroyed the taro crops in the valley. A 1960 Land
Study Bureau report completed only two years later estimated that 100 acres
were still cultivated in taro, along with 11 acres of macadamia nut trees, 5
acres of lotus root, and 2 acres of coffee. According to the report, the perma-
nent resident population was virtually nonexistent; most of the farmers lived
outside the valley and commuted in order to work on their taro patches.92
The launching of a full-scale tourist industry on Hawai‘i included widen-
ing, realigning, and improving the Hawai‘i Island belt highway. As part of this
construction, the Honoka‘a-Kukuihaele extension was completed in 1962,
opening the Waipi‘o Valley Lookout to thousands of visitors. Meanwhile, the
county constructed a mile-long road accessible by four-wheel-drive vehicles
into the valley. It descends 800 feet to the valley floor at a 20 percent grade,
with as much as a 45 percent grade at some points. The road dramatically
increased the number of visitors to the valley and created the potential for
tourist businesses to operate there.93 The farmers replaced their mule teams
with four-wheel-drive trucks to market their crops outside the valley.
These developments coincided with the stationing of Peace Corps trainees
in the back of Waipi‘o Valley, near Hi‘ilawe Falls. They built their own thatch

79
chapter two

huts, farmed, and fished in preparation for volunteer work in the Pacific and
Asia. They also practiced how to construct a bridge.94
The first actual threat of tourist development in the valley itself came in
1966 in the form of a proposal to build a restaurant and rest stop in the val-
ley, to be called the Waipi‘o Ti House. Because the rest stop was to be built
on land zoned for conservation, the developers needed a conditional use per-
mit from the State Board of Land and Natural Resources. The board approved
construction of the rest stop but not the restaurant.95 The Waipi‘o Ti House
was built and dedicated in February 1971 as a rest stop for hikers, with a rest-
room and facilities where lü‘au or parties could be held. In April 1972 a pub-
lic outcry was raised over the installation of an unauthorized power line from
the pali down to the Ti House because it was considered to be a precursor to
development of a restaurant. By November 1972 1,900 people had signed a
petition opposing a restaurant in Waipi‘o Valley. The State Board of Land
and Natural Resources denied the permit for a restaurant license in January
1973. At that point, the developer gave up and donated the 5.5-acre property
on which the Ti House sat to the Bernice Pauahi Bishop Museum.96

The Continuing Significance of Waipi‘o


Water and taro production remain the key determinants in the future of the
Waipi‘o and the kua‘äina with ancestral ties to the valley. Flooding continues
to limit the capacity of the kua‘äina of Waipi‘o to live there and the viability
of cultivating commercial crops. Flooding has also limited tourist devel-
opment.
In the 1960s a fourth Waipi‘o stream, Waimä, was diverted into the Lower
Hämäkua Ditch system. Combined, the water ditches diverted half of the
water that would naturally flow through Waipi‘o Valley.97 In 1963 heavy flood-
ing in Waipi‘o wiped out nearly half of the taro crop and farmers lost between
$200 and $1,500 worth of crops. The upper valley had six and a half inches of
rain in 24 hours, washing out the roads as the water rose up to five feet in
some areas.98 The last major flood, in 1979, destroyed many taro farms.
The maintenance of the Lower Hämäkua Ditch is a factor in sustaining the
healthy flow of the Waipi‘o streams. In 1989 the Lower Hämäkua Ditch tun-
nel, which runs behind Hakalaoa Falls, collapsed. A temporary flume was con-
structed around the collapsed tunnel, which ultimately involved the diversion
of Hakalaoa Stream. This reduced the twin falls of Hi‘ilawe to a single water-
fall. In 1993 the Hämäkua Sugar Company shut down and abandoned the

80
WA I PI ‘ O M ANO WA I : waipi ‘ o, source of water and life

Lower Hämäkua Ditch System. In 1995 the State Department of Agriculture


stepped in to operate and maintain the ditch system. By that point, one of the
four intakes was blocked by rubble, and twenty wooden flumes showed signs
of saturation, rot, and leaking. It was estimated that the system lost 4 million
gallons of water a day.99 In 1999 the U.S. Department of Agriculture’s Nat-
ural Resources Conservation Service completed a Watershed Plan to repair
the Lower Hämäkua Ditch to provide water to individual farmers along the
Hämäkua Coast and allocating water for Waipi‘o Valley itself. The Hämäkua
Ditch system remains a major factor in the availability of water and the con-
trol of flooding in the valley. The ability of the farmers of Waipi‘o to work in
coordination with various agencies and the farmers of the Hämäkua Coast to
harness and balance the flow of the waters of Waipi‘o is of critical significance.
Taro continues to be the primary crop cultivated in Waipi‘o. Waipi‘o is
upheld as a model for traditional ahupua‘a management organized around taro
production. Commitment to taro production and farming by the kuleana own-
ers and Bishop Museum has been crucial to protection of this unique cultural

Figure 14 Taro cultivation is pursued as both a commercial enterprise and a cultural and
educational learning opportunity by contemporary generations in Waipi‘o Valley. 1974.
Franco Salmoiraghi.

81
chapter two

and natural resource from the ravages of tourist development. Conservation


zoning has provided protection to the valley’s natural and cultural resources.
Taro production within the framework of traditional Native Hawaiian ahu-
pua‘a management also makes Waipi‘o an ideal place for ecotourist operations
as well as Native Hawaiian cultural and educational programs. Taro is pur-
sued as both a commercial enterprise and a cultural and educational learning
opportunity. The development of cultural learning activities centered on the
historical cultural sites and on Waipi‘o’s taro irrigation and cultivation net-
work complexes will help sustain Waipi‘o as a traditional center of taro farm-
ing. These educational centers also train a new generation of taro farmers
steeped in the traditions of Waipi‘o and in protocol related to the cultivation
of taro and the cultural resources of the valley. The Edith Kanaka‘ole Foun-
dation and the Kanu O Ka ‘äina Charter School have been pioneers in initi-
ating such activities. Students learn about the complexity of cultivating and
irrigating taro and about the cultural protocols and chants traditional Hawai-
ian farmers used to invoke their Gods to make their crop healthy and abun-
dant. Students also practice the values of shared and cooperative enterprise
and of respect and care for the land as they work together to grow and har-
vest taro.
As Waipi‘o moves into the twenty-first century, it remains a cultural kïpuka,
a center of traditional and customary Hawaiian cultural, subsistence, and reli-
gious activities. For the most part, the “kua‘äina” who participate in these
activities in the valley do not reside in the valley itself, as did the kua‘äina
prior to World War II, but in neighboring rural communities that bridge into
the modern society and the market economy. Just over 100 people actually
reside in the valley itself. Waipi‘o continues to be a center for training new
generations in taro cultivation, related cultural practices, and the kua‘äina
values and practices of lökähi. Recognized also as a traditional cultural and
spiritual center, Waipi‘o Valley and its new generation of kua‘äina are con-
tributing to the regeneration and perpetuation of Native Hawaiian cultural
beliefs, customs, and practices as a whole. Waipi‘o mano wai — source of water
and life.

82
 three 

Häna, mai Ko‘olau a Kaupö:


Häna, from Ko‘olau to Kaupö

Nä Ko‘olau

‘Ike ‘ia i ka nani o nä pali uliuli I see the beauty of the deep green cliffs
Aloha ku‘u home o nä Ko‘olau love my home in Ko‘olau
E huli mäua i Wai‘änapanapa Two of us travel to Wai‘änapanapa
He wai lukini ‘änapanapa mai nei The fragrant glistening waters ripple in
the wind
Kü mai ka pu‘u o Ka‘uiki Ka‘uiki Hill stands upright
‘O Häna ‘iu‘iu pöhai ke aloha And Häna lives in peace with my love
Kau mai ke ‘änuenue i Kaupö A rainbow is placed over Kaupö
Pä mai ka makani kä‘ili aloha The gentle wind caresses me and I take
the feeling of love.
—harry künihi mitchell

äna, one of the largest districts of Maui, is celebrated in the song

H printed here as the epigraph as a place of natural beauty and romance.


The traditional ‘ölelo noe‘au or saying Häna, mai Ko‘olau a Kaupö (from
Ko‘olau to Kaupö) provides us with the traditional boundaries of Häna, start-
ing in the moku or district (also called kalana or ‘okana) of Ko‘olau and extend-
ing through the moku of Häna and Kïpahulu to the kona or leeward moku of
Kaupö.1 The Häna district consists of almost one-third of the island of Maui.
On Maui, the ahupua‘a are marked from stream to stream, rather than from
ridge to ridge. On the northwest, O‘opuloa Gulch marks the Häna district’s
boundary with the Hämäkua district. On the southeast side, Wai‘öpai Gulch
marks the boundary of the Häna district with Kahikinui.
The land surveyor Curtis J. Lyons, writing in 1875 about the principles of
Hawaiian land divisions, described the system used in East Maui:

83
chapter three

On East Maui, the division [of land] in its general principles was much the
same as on Hawaii, save that the radial system was better adhered to. In fact
there is pointed out, to this day, on the sharp spur projecting into the east
side of Haleakalä crater, a rock called the “Pohaku oki aina”— land-dividing
rock, to which the larger lands came as a centre. How many lands actually
came up to this is not yet known.2

Within the districts of Häna are located smaller land divisions, or ahu-
pua‘a. These sub-district land divisions in general extend from the sea to the
uplands. Some extend inland only as far as the forest, while others sweep up
to the top of the mountain. A few go into the crater to meet ahupua‘a from
other districts at the piko (umbilical) stone, Pöhaku Pälaha, on the northern
rim of Haleakalä crater.
Before 1927, settlements along the Häna Coast were accessible only by
ocean or along rugged horse and mule trails. The Häna district is a prime
example of an isolated rural district where the pace of economic, social, and
cultural change proceeded more slowly than on other parts of the island. Häna

Figure 15 The road connecting “Hâna, from Ko‘olau to Kaupò,” opened in 1927 and
afforded intermittent work for Hâna Hawaiians on road maintenance crews. 1927. Tai Sing Loo,
Hawaiian Historical Society.

84
häna: from ko ‘ olau to k aupö

remained an area of continuity of Hawaiian culture and lifestyle through the


end of the twentieth century. The Hawaiian families and communities that
remained in Häna from generation to generation, pursuing subsistence liveli-
hoods and employing traditional methods of fishing and planting, provided all
Hawaiians with a connection to a unique Hawaiian way of life.3

Carving the Landscape


Wind and rain carved the dramatic landscape of the Häna district and shaped
the lives of the residents. Set patterns of the wind and the rain as they moved
across the land and ocean were named for their qualities and effect upon
nature and the kua‘äina. In 1922, when Thomas Maunupau and Kenneth
Emory journeyed to Kaupö, they met sixty-year-old Joshua Ahulii, who had
been born and raised in the district. He shared with them the names of the
famous winds of Kaupö and the neighboring ahupua‘a of Häna:

Kualau or Kuakualau—is the strong wind and the rain out in the ocean. In
Kona this wind brings in the ohua like fish along the beaches. It is custom-
ary for it to blow in the evening and in the morning but sometimes blow at
other times—. Where are you, O Kualau, Your rain goes about at sea.
Moa‘e—this is a customary wind. It blows strongly but pleasantly from
the sea and sometimes from the land. It is sung about, thus: “Where are
you, O moa‘e wind/ you’re taking my love with you.”
Moa‘e-ku—this is a customary wind like the Moa‘e but much stronger.
This wind was said to have been born in Häna, grew up in Kïpahulu,
attained maturity in Kaupö, became aged in Kahikinui, grew feeble at
Kanaio, rested and let its burden down at Honua‘ula. Here is a song for
this wind: “Where are you. O Moa‘e-ku / You make much work on a
stormy day.”
Malualua—This is the companion of the white, misty rain of Häna.
This is a famous wind of this land. It blows strongly and pleasantly from
the ocean and blows the rain back to the mountain. This is the song of
this wind, The misty white rain of Häna, Companion of the Malualua.
Kulepe—This wind comes with rain. It is strong and blows out to sea
from the land.
Kaomi—It was a strong, blustering wind whose strength does not last
long but blew like a gentle pressure. It is sung of thus: “The wind blows
in a gale, Then it gently presses.”

85
chapter three

Naulu—This wind goes with the Naulu clouds. The Naulu is the wind,
It bears the Naulu clouds along.
Kiu—It is a wind that flies along and seems to sneak by to the mountain
of Haleakalä. It is called the Kiu of Haleakalä. Here is a song of this wind:
“The Kiu is the wind that lives on the mountain.”
Hoolua—It is a strong wind from the sea, making the billows rise and
the white caps come on the ocean. This is a song sung by fishermen: If the
Hoolua is the wind, The skill of a canoesman is seen.
Mumuku—It is a strong wind blowing from the land and it is as though
it was being blown out by another wind.
Koholalele—It is a strong wind blown over the dry area of Kohala to
this place and on up to the mountain.
Koholaleleku—Is a wind like the Koholalele but stronger.
Makani ka‘ili aloha o Kïpahulu. The love-snatching-wind-of-Kïpahulu
is the usual Kïpahulu wind. It blows down from the mountain and goes
out to sea.4

Upon being asked, Ahulii also shared the names of the rains of the district:

Noenoe uakea o Häna—is a misty rain and white. It comes in the morning
and ends as the morning waxes . . . the song of the rain, Misty and white is
the rain of Häna, Companion of the Malualua wind.
Häna ua lani haahaa (Häna of the low rains from heaven) — is described
thus: A low hanging cloud comes from the ocean and then the rain falls.
That is why it was so named. The names of the rains of this very famous
place are pretty and I do not think there are rains anywhere else to com-
pare with these. No wonder it is said that Maui is the best.
Awa rain —It is a dark cloudy rain that falls all day in the mountains.
Koko rain —It spreads over the surface of the sea with a rainbow. It is a
sign of trouble of some importance when seen.
Naulu rain —It is a rain that moves over the mountain on a clear day
with a Naulu cloud.
Noe rain —It is a light shower and mist that remain in kula lands.
Haleuole rain —This is a naughty rain. When one wants to relieve
nature that is the time it comes with such suddenness and clears up just as
suddenly and then falls again. That is why it was so called (wipeless).
[Maunupau noted that he thinks this was not an ancient name.]
Lilinoe rain of Haleakalä —A famous rain belonging to that mountain.

86
häna: from ko ‘ olau to k aupö

Ukiu rain of Makawao — It is a fine rain with wind that blows down
from the mountains.
Peepapohaku (Hide-behind-rocks) rain—is an annoying rain something
like the Haeuole in falling and clearing away suddenly, thus sending people
to hide behind the rock walls. That's how it got its name. [Maunupau
noted that he does not believe that the ancients named it.]

The winds and rains identified by the küpuna of Kaupö grace the entire
Ko‘olau district as they blow in from the east across the ‘Alenuihähä Channel
and encounter the lush, steep, magnificent slopes of Haleakalä.

Godly Inhabitants of Hâna


Maui, the Hawaiian demigod credited with feats that made the earth livable
for humans, performed many of his famous deeds in the Häna district. He
snared the sun at the Ko‘olau Gap on the rim of Haleakalä and forced it to
slow down during the summer months.5 He hooked the islands of Hawai‘i
and pulled them up from the ocean when he went fishing with his brothers in
the fishing ground called Po‘o, directly off Kïpahulu and in line with the hill
Kaiwiopele.6 He also lifted the sky so that humans could walk upright while
standing on Ka‘uiki in Häna. Thus, while the clouds may hang around Hale-
akalä, they still do not touch Ka‘uiki.7
The gods Käne and Kanaloa are credited with going about all the islands to
establish springs of fresh water, including east Maui. It is said that they landed
at Pu‘uokanaloa (hill of Kanaloa), a small hill just north of Keone‘ö‘io when
they first came from Kahiki. They dug a water hole by the beach and found
the water brackish. So they went about 200 yards inland, dug another hole, and
created the spring called Kawaiakala‘o. These gods also opened the Kanaloa
fishpond at Luala‘iluakai, providing the brackish water needed for fish to
spawn.8 From here, they went on to Nu‘u, where they dug another spring.9
They also opened the Käne and Kanaloa springs at Ke‘anae near the ‘öhi‘a
gulch.10
A hairy type of wauke plant, useful for the beating out of bark cloth, is said
to have first sprouted in Kaupö out of the body of Maikoha, the youngest son
of Konikonia and Hina‘aikamälama.11
Mo‘o and sharks are the two prominent ‘aumakua associated with Hawai-
ian families from Häna. Stories about men being held hostage by female mo‘o
in legendary times abound. For example, the chief Puna‘aikoa‘e was kidnapped

87
chapter three

from Kaua‘i by the mo‘o woman Kalamainu‘u and held captive as her husband
in a cave in Häna.12
The Häna people believed that mo‘o lived in certain ponds and springs.
Fish that lived in mo‘o ponds were forms of the mo‘o and not to be eaten. If
one were caught and opened up, they were usually found to be soft and bit-
ter. If these ponds were used by someone for washing dirty clothing or were
polluted in some other way, the person responsible was usually punished with
an illness or some other form of misfortune.13
Even as late as the 1960s küpuna informants spoke of an old blind woman,
Tutu Pale, who claimed kinship to a mo‘o who lived in a nearby pond. She
often walked to the pond in the moonlight to fish and talked as she went along.
She explained that she was talking with her cousin, the mo‘o, who frequently
accompanied her to the pond. After ten years, she regained her eyesight. Her
son eventually put her in a nursing home, but his legs swelled up and would
not heal until he asked forgiveness for placing his mother in the home rather
than taking care of her.14
Sharks were also honored as ‘aumakua in Häna. The shark ‘aumakua lived
and bred in the waters off Kïpahulu. According to Josephine Marciel, a resi-
dent of the district:

This land was famous for sharks. Because it was full of sharks. People with
bad mouths, if [they] went swimming would be taken by sharks. Those who
defy sharks would be taken. This land is famous with tales of sharks before.
When the wiliwili blooms watch out for the sharks. The sharks bite. It’s
mating season. That’s the time when the sharks chase the females. When
the wiliwili blooms, the sharks become fierce. At Nu‘u, the wiliwili is
plentiful.15

The shark-man, Nanaue, a being with a human body and a great shark
mouth on his back, was driven out of Waipi‘o when he was exposed as the
predator who killed the people of the valley. He settled for a while at Kïpahulu
until he was forced out to flee to Moloka‘i, where he was finally killed.16
According to küpuna informants, Häna was also famous for the apparition
of Känehunamoku, the floating island of Käne. Mary Waiwaiole claimed to
see it floating on the water by Ka‘uiki Hill on certain mornings when she rode
by truck to work on the plantation. Once, members of the Waikaloa family saw
the island and then actually packed their things and went down to the shore
to wait for the island to return so that they could go away on it. However, the
island never floated back for them.17

88
häna: from ko ‘ olau to k aupö

Laka, the male god of forest growth and patron of the hula dance, is named
as the son of Kumuhonua (Earth firmament) and Lalohonua (Earth under-
ground) thirty-six generations before Papa and Wäkea, the first parents of
human progenitors.18 Laka was born at Kïpahulu and brought up at ‘Alae Nui
(“the great mudhen”), a land division in the Kïpahulu valley.19

Kü‘ula, Hina, and A‘ia‘i


The first fishpond in Hawai‘i is said to have been built by the fish god himself,
Kü‘ulakai, where he lived with his wife, Hina, and their son, A‘ia‘i, at Leho-
‘ula, ‘Aleamai, in Häna, near Kaiwiopele. He was the head fisherman for the
ruling chief of Häna. His marvelous fishpond attracted a lot of attention espe-
cially because the pond, through his power, was always full of fish. At Wailau
on Moloka‘i was a chief who had the power, as a kupua or demigod, to turn
into a gigantic eel, 300 feet long. In his eel form he was attracted to the fish-
pond and slipped into the inlet. However, after he had fed well he could not
get out without breaking down the wall. He hid in a deep hole beyond ‘Älau
island called “hole of the ulua,” and Kü‘ulakai baited the famous hook, Manai-
akalani, with roasted coconut meat to lure him out of hiding. When hooked,
the eel was dragged ashore by two ropes held by men who stood on opposite
sides of the bay while Kü‘ulakai stoned the eel to death. The body of the eel
turned to stone and can still be seen today. A follower of the dead chief /eel
moved from Wailau to Häna determined to avenge the death. He managed to
be appointed as a messenger of the chief to the fishpond. One day, upon
returning from the fishpond, he conveyed instructions that he said Kü‘ulakai
had given him about how to prepare the fish for cooking as instructions for
killing the chief. He said that Kü‘ulakai told him that the chief's head should
be cut off and the body sliced, salted and baked in an imu. The angered chief
ordered the death of Kü‘ulakai, Hina, and A‘ia‘i by burning them in their
house at night while they were asleep. Kü‘ulakai knew of the order and decided
that he and Hina should return to the ocean. They left A‘ia‘i behind with
objects to use in attracting fish and instructions on how to construct ko‘a or
altars to mark fishing grounds and to honor Kü‘ulakai and Hina as patrons of
fishing. Beckwith provided a description of the fishing ko‘a that A‘ia‘i estab-
lished in the Häna district:

The first fishing ground marked out by Aiai is that of the Hole-of-the-
ulua where the great eel hid. A second lies between Hamoa and Haneoo
in Hana, where fish are caught by letting down baskets into the sea. The

89
chapter three

third is Koa-uli in the deep sea. A fourth is the famous akule fishing
ground at Wana-ula . . . At Honomaele he places three pebbles and they
form a ridge where aweoweo fish gather. At Waiohue he sets up on a rocky
islet the stone Paka to attract fish. From the cliff of Puhi-ai he directs the
luring of the great octopus from its hole off Wailua-nui by means of the
magic cowry shell and the monster is still to be seen turned to stone with
one arm missing, broken off in the struggle. Leaving Hana, he establishes
fishing stations and altars along the coast all around the island as far as
Kipahulu. At the famous fishing ground (Ko‘a-nui) in the sea of Maulili
he meets the fisherman Kane-makua and presents him with the fish he
has just caught and gives him charge of the grounds, bidding him establish
the custom of giving the first fish caught to any stranger passing by canoe.
Another famous station and altar is at Kahiki-ula.20

A‘ia‘i taught people how to make nets and lines and showed them how to
lure octopus with the cowry shell. A‘ia‘i eventually left Maui and established
fishing ko‘a from Kaho‘olawe to Läna‘i and Moloka‘i, O‘ahu, Kaua‘i, Ni‘ihau,
and finally Hawai‘i.21

Kaiwiopele
Pelehonuamea also shaped the landscape of the Häna Coast. The goddess
dwelt at Haleakalä and built it up to its present size until her mortal enemy—
her sister, Nämakaokaha‘i, an ocean deity who could assume the form of a
dragon — discovered where she lived. Nämakaokaha‘i arrived at Haleakalä with
another sea dragon, Haui, and together they viciously attacked Pelehonuamea
and dismembered her body. Parts of the body landed in Häna near Kau‘iki and
formed the hill called Kaiwiopele (the bones of Pele). Nämakaokaha‘i believed
that she had finally destroyed her enemy. However, the spirit of Pelehonu-
amea lived beyond her body form and transformed into a more powerful force
as a goddess. The goddess Pelehonuamea fled to the island of Hawai‘i, where
she found a permanent home at Halema‘uma‘u, Kïlauea, Hawai‘i.22

Settlement and Development of Ko‘olau, Hâna, Maui


Menehune are believed to be among the original settlers of the earliest sites
in the Ko‘olau district, such as the Hale o Käne, Lonaea, and Lo‘alo‘a heiau
of Kaupö.23
Throughout the Ko‘olau district, Native Hawaiians settled in valleys
watered by streams with proximity to the ocean. Here they planted along the

90
häna: from ko ‘ olau to k aupö

streams and up the sides of gulches. Gradually, they terraced the gulches in a
network of interconnected lo‘i kalo or taro pond fields. Native Hawaiians also
settled on the slopes and flatlands above and between the valleys, regions that
were favored with lots of rain and also had access to the ocean. Here, they cul-
tivated dryland taro as well as sweet potatoes, yams, and bananas for food,
wauke for bark cloth, olonä for cordage, ‘awa for a relaxant drink, and other
edible and useful native plants. Native Hawaiians also developed settlements
along coastal areas of the drier sections where springs percolated up through
the rocky shore or in shallow waters of the bays. From the streams and near-
shore ocean they harvested fish, shellfish, and seaweed.
Moving along the coast, north to south, from ‘O‘opuloa Gulch and east-
ward to Nähiku, Native Hawaiian families settled and cultivated gardens in
the narrow valleys fed by small streams —‘O‘opuloa, Waikamoi, Puohokamoa,
and Haipua‘ena.24 Next, in the broad, deep valley of Honomanü, which has
a large stream and a broad beach for fishing canoes and net fishing, a large
Native Hawaiian population constructed terraces deep into the valley for taro
cultivation. The Nua‘ailua Valley and the adjacent slopes and flatlands were
also settled and cultivated in taro terraces. At Kea‘ane, the early Hawaiians
first settled in the uplands above the peninsula and cultivated dryland taro up
into the forested areas. Under the direction of a Ke‘anae chief, the people also
carried soil from the uplands onto the lava peninsula and, in a unique feat of
engineering, diverted water using an old canoe as a wooden flume in order to
develop an extensive network of taro pond fields.25 A large population of
Native Hawaiians also settled in and developed irrigated taro terraces in the
adjacent valley and broad lands of Wailuanui. Going east from Wailuanui,
Native Hawaiian families lived in the gulches of East Wailuaiki, West Wai-
luaiki, Kapili‘ula, Waohue, Pa‘akea, Kapa‘ula, and Makapipi and cultivated
taro and other useful crops.26
There are no large streams or gulches from east of Nähiku out to Hämoa,
but the area receives a lot of rain. The soil is made up of decomposed lava with
humus. At ‘Ula‘ino , Honomaele, Honokalani, Helani, Olopawa, Häna, and
Hämoa, Native Hawaiian families built homes and cultivated dryland taro on
the slopes and flatlands. They also cultivated yams, sweet potatoes, bananas,
wauke, olonä, and ‘awa. Upland from Hämoa, the Native Hawaiians cultivated
their food crops in a valley named ‘Öpaeku‘i during the dry season. The val-
ley is fifteen miles long and an offshoot of the marshy upland called Wai-
ho‘i. While working in their upland gardens, Hawaiian planters subsisted on
mashed shrimps (‘öpae ku‘i), which they took with them in calabashes, and poi

91
chapter three

pounded from the starchy core of the hapu‘u fern, which was cooked in imu
ovens of stone in the earth. The planters also gathered wild uhi and hoi (types
of yams) in the forest of this area.
South of Hämoa, several streams carved out gulches into the landscape,
and Native Hawaiians settled along the streams and developed taro terraces
at Maka‘alae, Waiohonu, Pu‘uiki, Pöhue, Pukuilua, Hä‘ö‘ü, Hulihana, Mü‘o-
lea, and Koali. Beyond Koali, the deep little valley of Wailua was extensively
settled and cultivated at four distinct levels, moving in from the sea up the val-
ley toward the mountain. Past Wailua, the land is steep and high and unsuited
to fishing or planting.
In Kïpahulu, Native Hawaiians settled and cultivated wet taro at Kukui-
‘ula, Lolokea, Hanawï, ‘Älele, Kalepa, and Nuanualoa. Dryland taro was cul-
tivated in the low forest that fringed the area.
Kaupö is a dry area. Dryland taro was planted in the lower forest belt.
However, sweet potato was the staple food of the families who settled here.
Even their poi was made from sweet potatoes. There were some settlements
in Waiha and Punalu‘u where a few taro terraces were developed. Water came
from springs such as Punahoe and Waiü and in the Manawainui Valley.

Figure 16 Waikauikalâ‘au, or “the suspended water,” a wooden flume made of an old canoe,
carried water from the Palauhulu Stream into the Ke‘anae Peninsula in the nineteenth century.
1883. C. J. Hedemann, W. T. Brigham Collection, Bishop Museum.

92
häna: from ko ‘ olau to k aupö

Hawaiian mauka-makai (mountain-ocean) use of the ahupua‘a in southeast


Maui was linked to the planting cycle, which was dependent upon the varia-
tions in rainfall according to elevation and seasons. In the uplands, where it
usually rained daily, planting could be done year round. In the lowlands,
planting was usually done in conjunction with the rainy season. When the
rains moved on to the lowlands, each family lived at temporary habitation sites
along the coast where they cultivated small plots of sweet potatoes and gourds.
This important seasonal habitation cycle is documented in the interviews with
Sam Po, a native of Kanaio. According to him, even up through the latter half
of the nineteenth century the kua‘äina in the district continued to live season-
ally mauka and makai and plant in accordance with the annual rains.27
The ocean along the entire Ko‘olau District provided Hawaiians with var-
ious sources of food, including numerous varieties of fish, crab, shellfish, and
seaweed. Fishing, such as diving, was an individual enterprise. Deep-sea fish-
ing in canoes was conducted by the men in the ‘ohana, while hukilau or large-
net fishing was a community enterprise.
Fishing and ocean gathering were carried out according to the moon
phases and the stars. When the stars were numerous and bright, that was the
time to go and look for the shellfish such as küpe‘e ( Nerita polita), which usu-
ally hide during the day. This gathering was done in the utmost silence, lest
the shellfish drop and burrow to hide themselves.28

Figure 17 All members of the Hâmoa community of Hâna would gather to pull in the
great hukilau fishing nets. 1936. Harold T. Stearns, Bishop Museum.

93
chapter three

Salt for the Häna district was gathered at Nu‘u, where there were kaheka
(natural hollows) in the rocks in which salt (pa‘akai) accumulated when the
shallow ponds that formed during rough seas dried up in the sun. People from
throughout the Häna district would travel to Nu‘u in the summertime to
gather salt. Nu‘ualo‘a, in Kaupö, had several veins of ‘alae, a red earthy min-
eral that gets its color from its high iron content. The Hawaiians commonly
ground their salt together with ‘alae to enrich it with iron.29 Families would
gather an entire year’s supply during the summer, dry it, and store it in caves.
Summer was also the spawning time for the manini, the humuhumunukunuku-
apua‘a, the mullet, and the äholehole fishes. These would be caught and salted
to provide food throughout the year.
At the end of the development period on Maui, during the time of Kaka-
alaneo, the division of lands on Maui was carried out under the priest Kahuna
Kalaiha‘ohi‘a (“hew the bark of the ‘öhi‘a tree”). This marked the beginning
of the rule of chiefs over islands, with landlords over large divisions under
them and agents appointed by the landlords overseeing districts and small
divisions. At about this time on O‘ahu, Mailikükahi marked out the bound-
aries for the land divisions and initiated measures to strengthen the rule of a
chief over an expanding population.30

The Migratory and the Ruling Chiefs of Hâna


Nu‘u and his wife Nu‘umea or Nu‘umealani, or “the female who propagates
from heaven,” arrived at Nu‘u on Maui with their canoe, called the canoe of
Käne. Nu‘u was a great kahuna, associated with the era of overturning and
the time of the great flood. In the genealogy of Kumuhonua, Luanu‘u, son of
Nu‘u, also called Känehoalani, was the ancestor of the Mü and the Menehune
people. Nu‘u came after the first Hawaiian from a foreign place, and after him
came Hawai‘i Nui or Hawai‘iloa.31
The mo‘olelo of Häna provides at least two different accounts of the ori-
gin of the ruling chiefs of Maui. In both versions, Häna was ruled separately
from West Maui. According to one source, the ancestors of the mö‘ï (kings)
of Maui were Paumakua, a southerner voyager, possibly a Tahitian (975 or
1200), and Haho (1000 or 1225). However, the districts of Ko‘olau, Häna,
Kïpahulu, and Kaupö were often under different mö‘ï not closely connected
with the rulers of western Maui.32
A second source states that Maui was divided into two separate kingdoms,
where the Nana‘ulu line governed the Häna Coast and the ‘Ulu line con-

94
häna: from ko ‘ olau to k aupö

trolled the remainder of the island. At the time that High Chief ‘Umialïloa
ruled Hawai‘i, High Chief Pi‘ilani, from the ‘Ulu line of western Maui con-
quered the Häna coast and united the island under his rule.33 At the time of his
conquest, Pi‘ilani may have dedicated the great temple, Pi‘ilanihale, believed
to be the largest heiau in the Hawaiian Islands, at Honoma‘ele.34 To consoli-
date an alliance with the Maui chiefs, ‘Umialïloa arranged to marry Pi‘ikea,
the daughter of High Chief Pi‘ilani.
The history of the Häna district in East and South Maui also involves many
chiefs from Hawai‘i. Given Häna's proximity to Hawai‘i, it periodically served
as a residence and sanctuary for chiefs of both islands. The distance between
East Maui and ‘Upolu Point on Hawai‘i could be crossed in either direction
in a couple of hours. It was an ideal location for ruling ali‘i of either island.
Häna had an abundance of the type of wood used for making scaffolds and
ladders to scale fortresses, and it had the best smooth, round stones for war-
riors to use in slingshots. Fish and taro were plentiful. The surfing was fine.
The Hawaiian historian Samuel Kamakau described the features that made
Häna so attractive:

Häna was in those days a noted place famous for the fortified hill Ka‘uiki,
the surf at Puhele, the fresh-water bathing pool of Kumaka, the diving at
Waiohinu, the flying spray of Kama, the changing color of the fronds of
the ama‘u fern, the yellow-leafed ‘awa of Lanakila, the delicious poi of
Kuakahi, the fat shell fish (‘opihi) of Kawapapa, the fat soft uhu fish of
Haneo‘o, and the juicy pork and tender dog meat dear to the memory of
chiefs of that land, moistened by the ‘apuakea rain that rattles on the hala
trees from Wakiu to Honokalani.35

Altogether, it was then, as it remains today, a pleasant, bountiful, and beau-


tiful place to live.36 The fortress of Ka‘uiki also made Häna one of the most
attractive places for the Native Hawaiian chiefs to live.

Ka‘uiki and the Ruling Chiefs


The ‘olelo no‘eau “O Wananalua ia ‘äina; o Punahoa ka wai; o Ka‘uiki ka
pu‘u” (Wananalua is the land; Punahoa is the pool; Ka‘uiki is the hill)
identifies famous wahi pana of Häna.37
Of these, Ka‘uiki hill is the most famous and culturally significant. It fig-
ured prominently in the myths and history about Häna. Martha Beckwith
wrote about the hill’s mythical origins and significance in Hawaiian Mythology:

95
chapter three

Some say that it sprang from the navel of Hämoa. Others that it was born
to the parents of Pele, or to the hill Kai-hua-kala by his wife Kahaule.
Others relate how Ka-lala-walu (The eight-branched) brought the hill
from Kahiki as an adopted child, but grew tired of its nibbling at her
breasts and tried to leave it along the way, first at Kaloa, then at Kaena,
then at the Ka-wai-papa stream. Others tell of the wanderings and death
of Pu‘uhele, little sister of Pele . . . men say that formerly Käne and Kana-
loa planted a garden below the hill, and they point out two rocks below
the hill on the inaccessible sea side which are called “the coconuts of Käne
and Kanaloa” and the “root-stock” (kumu) of Kauiki . . . Here lived Hina-
hana-ia-ka-malama, she who worked at tapa making in the moon, and her
husband, father of Puna and Hema on the Ulu line of chiefs.38

During the epoch of Hawai‘i's ruling chiefs, Häna was reputed to be a


favorite and beloved district because of the fortress of Ka‘uiki and the ease of
living in that district.39 A chant for Ka‘uiki recalls, “Healoha no Ka‘uiki /Au i
ke kai me he manu la” (Ka‘uiki is beloved /Afloat on the sea like a bird).40
Ka‘uiki was famous for its strength as a refuge in times of danger. Its sum-
mit was approached by a ladder made of ‘öhi‘a wood from Kealakomo and fas-
tened with the ‘ie‘ie vine from Paiolopawa. The summit was covered with kana-
wao plants from Kawaipaka, furnishing a natural bedding for those defending
the fortress. The fishponds of Kihahale provided an abundance of fish. The
big ‘awa roots of Kualakila delighted the nostrils of the first-born chiefs with
their aroma. Below the fortified walls of Ka‘uiki was the excellent battlefield
of Wänanalua. A lookout was stationed at Mokuhana to warn those in the
fortress where to strike against their enemies.41
Numerous chiefs of Maui and Hawai‘i settled with their families and entou-
rages at and around Ka‘uiki. These included Hua, son of Pohukaina, descen-
dant of the Ulu line of chiefs; Kanaloa and Kalahumoku, sons of Hualani (the
wife of Kanipahu) and half brothers to Kalapana, who ruled Hawai‘i; Eleio;
Kalaehaeha; Lei; Kamohoali‘i; Kalaehina; and Ho‘olaemakua.
Ka‘uiki featured prominently in the rivalry between the sons and heirs,
Lonoapi‘ilani and Kihaapi‘ilani, of the ruler of Maui, High Chief Pi‘ilani.
Lonoapi‘ilani, as ruler of Maui, felt threatened by his younger brother Kiha-
api‘ilani and sought to kill him.42 Kihaapi‘ilani escaped and then conspired
to overthrow Lonoapi‘ilani. Kihaapi‘ilani married the daughter of the chief
Ho‘olaemakua, who controlled Ka‘uiki, in order to recruit his support for the
rebellion against Lonoapi‘ilani. Ho‘olaemakua remained loyal to Lonoapi‘i-

96
häna: from ko ‘ olau to k aupö

lani, and Kihaapi‘ilani was forced to seek the support of his brother-in-law,
‘Umialïloa of Waipi‘o. Kihaapi‘ilani and ‘Umialïloa spent a year preparing for
the invasion of Maui by building canoes and making clubs and other imple-
ments for battle. When the fleet of canoes reached Häna, Ho‘olaemakua com-
manded the battle from Ka‘uiki. The Hawai‘i warriors were unable to land at
Häna. Kihaapi‘ilani directed the Hawai‘i canoes to land on the shores of Wai-
luaiki and Wailuanui, where they were dismantled and set upright so that all
of the great fleet could beach. The Hawai‘i warriors marched over land to
Häna. Ho‘olaemakua fought against the Hawai‘i warriors for days, until the
warrior Pi‘imaiwa‘a discovered that at night, the fortress was guarded only by
a wooden image cunningly set up to look like a huge armed warrior. The
Hawai‘i warriors finally defeated the forces of Ho‘olaemakua and killed him
at Kapipiwai, near Nähiku. Pi‘imaiwa‘a assumed the chieftainship of Ka‘uiki.
The war fleet of Kihaapi‘ilani and ‘Umialïloa proceeded to attack Lonoa-
pi‘ilani at Wailuku. Hearing of their approach and fearing defeat, torture, and
death, Lonoapi‘ilani is said to have trembled with fear and died. Kihaapi‘ilani
became one of the greatest rulers of Maui. He oversaw the construction of
the Alaloa road: 138 miles long and paved with water-worn stones laid out
four to six feet wide between curbstones, it completely encircled the island of
Maui.43
Kaleikini, who also dwelt at Ka‘uiki, is remembered for sending canoes to
Lana‘i to gather logs of kauila wood to seal the many blowholes along the
Häna coast, which had previously sprayed salt water on the good agricultural
lands and natural vegetation of Häna. At Honokalani, remnants of those logs
can still be seen.44
In Kamehameha’s time, Ka‘ahumanu, the most influential and powerful
wife of Kamehameha I and kuhina nui or prime minister for Kamehameha II
and Kamehameha III was born at Ka‘uiki in 1768. Her parents, Ke‘eaumoku
and his wife, had sought refuge there from Kahekili. Ke‘eaumoku and the
Häna chiefs played a major role in Kamehameha's rise to power and in his
central government.45 On the eve of European contact, Häna continued to be
fought over by the ruling chiefs of Hawai‘i and Maui.
High Chief Kekaulike resided at Kaupö and in 1700 built the great heiau
of Pu‘umaka‘a at Kumuni and Kanemalohemo (Keakala‘auae) at Popoiwi near
Mokulau in Kaupö. High Chief Kekaulike also rededicated the Lo‘alo‘a heiau
in Kaupö around 1730. He left his youngest son, Kamehamehanui, as ruler
when he died in 1736 at Lelekea, near Kaupö.46
Later, Alapa‘inui, king of Hawai‘i, landed a large force at Mokulau near

97
chapter three

Kaupö to raid Maui. However, in 1736, when he discovered that his own
nephew, Kamehamehanui, was the new mö‘ï of Maui, he negotiated a peace
agreement.47
In 1754 Kalaniopu‘u became the ruling chief over Hawai‘i.48 In 1759 High
Chief Kalaniopu‘u went to war against High Chief Kamehamehanui to win
control of Häna and Kïpahulu. Kalaniopu‘u won control of the marvelous
fortress of Ka‘uiki. Kamehamehanui, with the support of chiefs from Molo-
ka‘i and Läna‘i, sought to regain Ka‘uiki, Häna, and Kïpahulu in the battle
known as Kapalipilo. The battle extended from Akiala to Honomaele, but
Kalaniopu‘u and the Hawai‘i warriors were victorious, and Häna and Kïpa-
hulu remained under the control of Hawai‘i. When Kalaniopu‘u returned to
Hawai‘i, he left the chief Puna in control of Ka‘uiki. Mahihelelima, indepen-
dent chief of Häna, Kïpahulu, and Kaupö deceived Puna and thus managed
to replace him as head of the fortress of Ka‘uiki. Nevertheless, Kalaniopu‘u
retained control over Häna and Kïpahulu.
In 1765 High Chief Kamehamehanui died. He was succeeded as the ruler
of Maui by High Chief Kahekili. From 1775 to 1779 Kalaniopu‘u continu-
ously waged battles against High Chief Kahekili of Maui from his stronghold
in Häna and Kïpahulu.49 In the battle of Kalaehohoa (“the forehead beaten
with clubs”), Kalaniopu‘u sent his warriors from Häna to raid Kaupö and take
control of the district. The Hawai‘i warriors abused the local people and beat
them over the head with clubs. Kahekili sent his warriors, led by Käne‘olae-
lae, to Kaupö to retaliate against Kalaniopu‘u and restore his control over that
district. Kamehameha distinguished himself as a strong warrior in this battle,
known as Kalaeoka‘ilio where it was fought. He rescued the warrior Keku-
haupi‘o, his instructor in the fighting arts, from death. However, as described
by Stephen Desha, the Hawai‘i warriors were defeated: “The Hawaii warriors
fled but many were slaughtered by the Maui people at that battle at Kaupö
which was named the Battle of Kalaeoka‘ilio. It was a battle in which the bod-
ies of the Hawai‘i warriors were heaped like kukui branches before Maui’s
exceptional warriors.” 50

Contact and the Rise of Kamehameha over Hâna and Maui


When Captain James Cook returned to Hawai‘i in November 1778, he first
anchored at Ha‘aluea, below Wailuaiki on Maui.51 At the time, Kalaniopu‘u
and his forces were in the Ko‘olau district of Maui fighting Kahekili. The

98
häna: from ko ‘ olau to k aupö

young Kohala chief, Kamehameha, and his mentor, Kekuhaupi‘o, boarded


Cook’s ship and sailed with Cook toward Hawai‘i Island until a canoe sent by
Kalaniopu‘u caught up with the ship and summoned it to return to Maui.52 In
1781 Kahekili made war on the Hawai‘i chiefs and their Maui allies in Häna.
Mahihelelima joined forces with Kalaniopu‘u against Kahekili. The fortress of
Ka‘uiki was impenetrable and held out against Kahekili’s warriors for one year
until Kahekili was able to cut off the water supply to Ka‘uiki. Thus, he finally
wrested control of Häna and Kïpahulu away from Kalaniopu‘u and his allies.53
In 1785 Kahekili vanquished Kahahana as chief of O‘ahu and went to
O‘ahu to consolidate his rule over that island, leaving his son, High Chief
Kalanikupule, to rule Maui. At that time, Kamehameha sent his brother Kala-
nimalokulokuikepo‘olani to retake control over Häna and Kïpahulu for the
Hawai‘i chiefs.54 He met with no resistance and instituted a policy of benev-
olent rule over the people, gaining the name Keli‘imaika‘i or “good-hearted
ali‘i,” which he kept until his death. When Kalanikupule learned that Häna
and Kïpahulu were under the rule of the Hawai‘i chiefs, he dispatched a great
army of warriors that defeated and ousted the Hawai‘i warriors. Keli‘imaika‘i
escaped back to Hawai‘i.
In 1790 Kamehameha invaded Häna and defeated the Maui warriors in a
battle called Kaua o Kawa‘anui (Battle of Great Canoes). Moving on from
Häna, he slaughtered the Maui chiefs at Iao Valley in a battle called Kaua i
Kepaniwai o Iao (Battle at the Dammed Water of ‘Iao; the name referred to
the damming of the ‘Iao stream with bodies of slain warriors). Despite these
victories, as discussed in chapter 2, Kamehameha had to return to Hawai‘i to
protect his lands and people from the assaults of his rival, Keouaküahu‘ula of
Ka‘ü. At that point, Kahekili formed an alliance with High Chief Ka‘eokulani
of Kaua‘i and returned to Maui. Together, Kahekili and Ka‘eokulani retook
control of Häna and launched an attack upon Kamehameha and his forces on
Hawai‘i. Landing at Waipi‘o, Ka‘eokulani viciously attacked the people of
Waipi‘o and desecrated the sacred heiau of the Hawai‘i chiefs, as described in
chapter 2. When Kahekili joined Ka‘eokulani, their forces were defeated by
Kamehameha off of Waimanu in the naval battle Kaua o Kepuwaha‘ula‘ula,
(Battle of Red-Mouthed Cannon).55 Kahekili retreated back to Maui in defeat,
having fought his last battle. He died on O‘ahu in 1793. Kamehameha finally
assumed the rule over all of Maui, including Häna, through his victory over
High Chief Kahekili’s successor, High Chief Kalanikupule, in the Battle of
Nu‘uanu on O‘ahu in 1795.

99
chapter three

Hâna in the Nineteenth Century


In the early 1800s missionaries made an estimate of the population in the var-
ious districts of Maui in order to determine the number of missionaries to
assign. The missionary Jonathan S. Green concluded that Häna, Kaupö, and
Kïpahulu were well-populated districts in need of missionaries, as he wrote in
December 1835:

• Kaupö district: similar in appearance to the Kïpahulu district, but larger


and more populous. Small vessels frequently anchor here. Missionaries
are needed here.
• Kïpahulu district: rough country, fertile, and populous. In need of two
missionary families and a missionary station.
• Häna district: large populous densely inhabited in need of four
missionaries.56

In 1837, seventeen years after the settlement of missionaries in Hawai‘i, a


permanent mission station was established in Häna. Prior to that date, mis-
sionary presence and influence in the district were weak because the mission-
aries ventured to the Häna Coast only once or twice a year. However, mission
schools were established in the Häna district by 1850.57 Churches in the area
were built in the 1850s and 1860s with beach rocks that were washed up to
shore by extraordinary storms.58 Even after the churches were established,
mission duty in the Häna district was considered to be among the most diffi-
cult because of its isolation.59 Thus, Hawaiian spiritual beliefs and practices
persisted without competition from Christian beliefs longer in Häna than in
other parts of Maui.
Continental diseases killed much of the Native Hawaiian population in
east Maui. Samuel Kamakau described how a measles epidemic in 1848 wiped
out a third of the population of the Islands. About the impact in east Maui he
wrote: “I know personally of two families in Kipahulu, those of ‘Ili-mai-hea-
lani and Kuku-‘ula, in which only three persons were left out of fourteen. In
Ka-pule’s home at Papauluna [Kipahulu] nine died out of thirteen. At this
rate more must have died than survived.” 60
Samuel Kamakau also described the impact of the smallpox epidemic that
swept through the Häna district in 1853: “The whole population was wiped
out from Wakiu, the uplands of Kawaipapa, Palemo, and mauka of Waika‘a-
kihi in the Hana district, and so for ipahulu and Kaupo.” 61
The Mähele in 1848 and the Kuleana Act in 1850 established a system of

100
häna: from ko ‘ olau to k aupö

private property ownership of all the lands on Maui and the Hawaiian Islands.
In the entire Häna district a total of 1,690.718 acres were awarded to 324
maka‘äinana. The awards ranged in size from 0.03 acres to 211 acres; the aver-
age size was 5.218 acres. Table 2 shows the number of Land Commission
awards granted in each subdistrict.

Hâna Mâhele Awards to Ali‘i


Ten awards of ahupua‘a or large sections of ahupua‘a were also made to ali‘i
and konohiki in the Häna district.62 In Ko‘olau, the ahupua‘a of ‘Ula‘ino was
awarded to James Young Kanehoa, son of Kamehameha's British military
adviser, John Olohana Young, and his first wife, Namokuelua. He accompa-
nied King Liholiho and Queen Kamämalu to England as an interpreter. From
1846 to 1847 he served on the Board of Commissioners to Quiet Land Titles.
In Häna, Julia Alapai Kauwa, the wife of Kanehoa’s brother, Keoni Ana
( John Young, Jr.), was awarded 873.89 acres of the ahupua‘a of Haneo‘o. The
ahupua‘a of ‘Aleamai, with 1,093.5 acres, was awarded to Enoka Kuakamauna
for his wife, Henrietta Kaleimakali‘i. According to the claim in the Native
Register, ‘Aleamai was his wife’s share from the mö‘ï. Kaleimakali‘i was a
descendant of Keawekuikekaai, son of Keakealanikane and Kaleimakali‘i I.
Her mother was a niece and namesake of John Papa Ii’s mother, Wanaoa
Kalaikane. The konohiki Kahanu was awarded half of the ahupua‘a of Hono-
maele — 990 acres. He relinquished his share in the other half of the ahupua‘a
to the government.
The ahupua‘a of Mü‘olea was awarded to Keohokalole, the mother of King
Kaläkaua and Queen Lili‘uokalani.
In Kïpahulu, the two ahupua‘a ‘Alaenui and Wailamoa were awarded to
Kekauonohi, who was married to Aarona Keali‘iahonui, governor of Kaua‘i
and hänai son of the kuhina nui Ka‘ahumanu. She married Levi Ha‘alelea in
1850, a year after Keali‘iahonui’s death. Kekauonohi was an heir of her uncle

table 2 land commission awards in häna


district no. of awards no. of acres average size

Ko‘olau 131 awards 361.573 acres 2.76 acres


Häna 111 awards 785.418 acres 7.075 acres
Kïpahulu 56 awards 354.53 acres 6.33 acres
Kaupö 26 awards 189.197 acres 7.276 acres

101
chapter three

Kalanimoku as well as of her mother’s husband, Kaukuna Kahekili. She also


received awards of Mäalo, Pu‘ulani, Popo‘ö‘io, and Kahuai in Kaupö. The
ahupua‘a of Ka‘apahu was awarded to William Charles Lunalilo, who later
became king. The ahupua‘a of Maulili was awarded to J. A. Kuakini.
In Kaupö, 12,140 acres in the ahupua‘a of Nu‘u was awarded to Kalai-
moku. He was a cousin of Kuhina Nui Ka‘ahumanu and descended from
Kamanawa I. His wife was the sister of Kamakahonu.
Five awards were also made to foreigners in the Häna district. In Ko‘olau,
471 acres were awarded to Stephen Grant, Pakea, and Poakea. In Häna, the
American Board of Commissioners for Foreign missions acquired 27.64 acres
in Wananalua. John Richardson and Company was awarded 211 acres in
Olewa. In Kaupö, 1.5 acres were awarded to a Dr. Baldwin in Pualaea, Kali-
anu, and 145 acres were awarded to William Harbottle in Kumunui 1.
The first sugar mill was established in Häna in 1849 by George Wilfong,
a haole sea captain who invested his profits from whaling activities in sixty
acres of land at Ka‘uiki.63 In 1852 the first Chinese laborers arrived in Häna
on five-year contracts, but Wilfong’s mill burned down soon thereafter. In

Map 2 The Coulter 1853 population map of Maui shows a population of 2,000 for the
Ko‘olau, Hâna, Kîpahulu, and Kaupò districts. o = 50 persons. Source: Coulter, Population and
Utilization of Land and Sea, p. 22.

102
häna: from ko ‘ olau to k aupö

1864 two Danish brothers, August and Oscar Unna, raised $47,000 to start
the Häna plantation. Beginning in 1868 they imported Japanese laborers
through the Hawaiian Labor and Supply Company.64 By 1883 there were six
separate sugar plantations on the Häna Coast— the Häna Sugar, Ka‘elekü
Sugar, and Reciprocity Sugar Companies, and the Hämoa Agricultural,
Kawaipapa Agricultural, and Haneo‘o Agricultural Companies.65
As the Chinese immigrant laborers who had been brought in to work on
the Häna Coast plantations either completed or broke their contracts in the
late 1800s, they looked for a place to settle. Along the Häna Coast, Ke‘anae
provided an excellent location, given its isolation and the traditional terracing
and irrigation of the land for taro. The Chinese were able to cultivate rice
wherever taro farmers were interested in earning cash by renting out their
traditional taro growing land.66
Throughout the nineteenth century and until the Häna road was built in
1927, the Ko‘olau district of Maui, between Ke‘anae and Häna, was practically
inaccessible by land. Travelers entering the district from Wailuku, usually
rode on horseback to Ke‘anae, and then journeyed by canoe to Häna, gener-
ally taking two days. However, if one traveled entirely by canoe from Wai-
luku, the trip took only five and a half hours.67 Through the 1860s, when the
Häna sugar plantation was established, Häna maintained a reputation of being
one of the most isolated places in all of Hawai‘i.68
At Nu‘u, in the 1800s, the king’s ship used to come from Lahaina to obtain
fish. The kua‘äina would catch mullet for the king in the fishpond. Originally,
the pond had an outlet to the ocean, but it was later blocked off. At Nu‘u
landing, fishermen who perished when their canoes overturned in the rough
surf as the men were trying to land were buried in a small graveyard.69

The District during the Territorial Period


In 1900 the average pay for the plantation workers was $15 a month for
twenty-six days of work, ten hours a day in the field or twelve in the mill. This
was less than 5 cents an hour. They were also provided with camp housing,
schooling, some medical care, and recreation programs. Many of the workers
supplemented their meager incomes by fishing, hunting, and gardening,
learning from the Hawaiians of the district how to best utilize the natural
resources available along the Häna Coast.70 By 1905 both Hämoa Agricultural
Company and the Haneoo Agricultural Company had been absorbed into the
Kaeleku Company, a subsidiary of C. Brewer.

103
chapter three

The principal limitation to the modernization of the Häna Coast was the
difficulty encountered in traveling over land to Häna for trucks and cars in
1927. The people of the Häna Coast overcame the obstacles to transportation
and communication by maintaining contact with the outside world through
the inter-island steamers that called at the district at least once a month. Häna,
the safest harbor for these steamers, developed as the major center of com-
merce for the Ko‘olau district.71
Although Häna was regularly serviced by steamers, because of the difficulty
of the voyage, visitors to the district were rare, and the manner of life among
the Hawaiians was quaint by comparison to that of urban Honolulu, O‘ahu,
or even Wailuku, Maui. In 1910 H. M. Ayres, a reporter for the Pacific Com-
mercial Advertiser, wrote a series of articles about an excursion through the
Häna Coast. He described the arrival of passengers from the SS Claudine at
Häna:

While the landing is all right for freight, it is mightily inconvenient for
passengers and when the swell is heavy it must be dangerous to life and
limb. Passengers leaving here have to jump into the arms of the boat boys,
and on occasion have to be grabbed and thrown from the boat to the
wharf.
The visitor to Häna put up at the comfortable club maintained by
L. Y. Aiona, a Chinese gentleman of parts who runs the club rather as an
accommodation for visitors than as a money-making proposition. A man
at the house of whom every visitor to Häna calls is W. P. Haia, the
Bismarck of the Maui County Board of Supervisors.72

Mr. Ayers was still able to observe a traditional lauhala thatched house in
the Häna area. Nearby, men were making fishhooks out of wire and women
cooked papayas in ti leaves.73
Even though Ke‘anae had a landing, Harry Kühini Mitchell, a Ke‘anae taro
farmer born in 1919 in Ke‘anae of Hawaiian and Chinese ancestry, recalled
that if the ocean was too rough for loading and unloading there, freight would
be dropped off at Häna, and the people of Ke‘anae would have to go to Häna
by horseback to pick up their goods. He also remembered how the Chinese
owner of a poi factory in Ke‘anae, Ah Lum, had a twelve-mule wagon train
that he used to deliver poi between Ke‘anae and Häna. When he reached Häna
he would go down to the pier to look for freight for the Ke‘anae people. He
would then take the freight back to Ke‘anae for a small fee. According to
Mitchell:

104
häna: from ko ‘ olau to k aupö

Häna was the major place. The same boat come Ke‘anae. Too rough . . .
they cannot land . . . so they come Häna, unload our goods. And we got a
horse trail eh. And we come all the way Häna, pick up our freight, come
home.
And the Chinese, Ah Lum, he had the poi factory . . . He pound his
own poi. Deliver the poi. Come Häna . . . pick up, he go down the pier.
Look for freight for the Ke‘anae people. You know, then he charge you
so much, so much, for bringing the freight, eh.
The Ke‘anae landing—too small, eh. And, uh, where they land, you
know, the whaler come in. The big boat then come outside. Then the
rowboat come inside. The whaler deliver passengers, deliver freight, we
get donkey engine. The old way only up and down. Then, uh, gotta get
one man with the rope, one on each side, yeah. Then pull ’em. Come on
land then they— only can go up and down. And steep.74

Sugar freighters also made port calls to transport the sugar produced in the
district to mainland markets.
The population of the Häna district in 1930 was 2,436, of which 1,177
people were Hawaiians, accounting for 48 percent of the population. In
Kaupö and Kahikinui the total population was 185, of which 160 people were
Hawaiian, accounting for a percentage of 86 percent.75
A few Native Hawaiians worked for the sugar mill, and some worked as
cowboys for the Kaupö Ranch, but the majority of Hawaiians were subsis-
tence farmers and fishers, occasionally selling or exchanging their labor, agri-
cultural products, or fish for money, cloth, or tools. They farmed, hunted, and
fished to provide their families with food and basic necessities. When money
was needed, the men would usually work for a day’s wage shoeing horses,
pouring cement, clearing land, harvesting rice, mending fences, catching wild
cattle, or hauling goods overland.76

table 3 population of häna district in 1930


district total population no. hawaiian % hawaiian

Häna Town 1,585 536 34


Kïpahulu 147 118 80
Kaupö-Kahikinui 185 160 86
Total 2,436 1,177 48
Source: U.S. Bureau of the Census, Fifteenth Census of the United States: 1930, Occupation
Statistics Hawaii (Washington: Government Printing Office, 1931), p. 72, table 22.
The precincts were identified in “Governors’ Proclamations,” 1926–30, pp. 6–21.

105
chapter three

The ‘Ohana of Hâna

Native Hawaiian households of the Häna coast tended to be larger in size


than those of the urban Native Hawaiian families on O‘ahu. They were usu-
ally multigenerational, that is, they consisted of grandparents, parents, and
grandchildren. Married children were encouraged to live at home. The tradi-
tion of hänai was actively and extensively practiced among the Häna Coast
Hawaiians.77
The homes of the Häna Hawaiians were generally small two- or at most
three-room wooden structures with outhouses. The large families usually
crowded together on the floor to sleep. Cooking was usually done outside over
open wood fires.
Along the Häna coast, the extended family or ‘ohana was still a viable social
and economic unit. Food was constantly shared and exchanged within the
‘ohana. Relatives were always ready to share whatever they had with relatives
and to take in destitute relatives and friends.
Douglas Yamamura, who became a professor of sociology and chancellor
of the University of Hawai‘i at Mänoa, was born and raised on the Häna plan-
tation. In 1939 he conducted interviews with families from Häna, Kïpahulu,
and Kaupö for his master's thesis in sociology at the University of Hawai‘i.
Yamamura made a number of observations about the Native Hawaiian ‘ohana
of the Häna Coast on the eve of World War II, in 1939–41:

The Hawaiian household is still a closely knit group. The bilateral kin or
family group functions today as a cooperative and jointly responsible unit
in the household organization. This is a survival of the old Hawaiian
culture in which the “large family” or ‘ohana, a kin group — the bilateral
kinship grouping of all those related by blood, marriage, or adoption, and
including a number of households—functioned as a unit in economic and
social affairs of the community.
The average Hawaiian depends largely on relatives and friends for
support when he is destitute. Relatives are always ready to share what they
have or to take others into the household, even though there are no visible
means of supporting the enlarged group.78

Nearly 75 percent of the households along the Häna Coast had horses or
mules for transport between the communities of the coast and for traveling
outside the district. However, most people just traveled on foot within their
community and ahupua‘a or between ahupua‘a. The churches, one-room

106
häna: from ko ‘ olau to k aupö

schoolhouses, and general stores, which were commonly located at the center
of each of these communities, were all readily accessible on foot. Whatever
jobs were available were also within walking distance. The communities were
loosely organized units composed of households scattered over a wide area.
The churches and schools served as unifying forces for the community. The
churches brought the community together for the Sunday service, and the
schoolhouse brought the children together for daily classes.79
Family obligations often put a heavy strain on ethnically mixed marriages.
It also made it almost impossible for a Hawaiian to survive in any storekeep-
ing business, because the Hawaiian owner was expected to automatically
extend unlimited credit to his family and friends.80
Yamamura took note of Hawaiian attitudes toward money and work in his
study of factors affecting the success of Häna Hawaiian children in school:

Under the old culture and economy of Hawaii, there was no need for the
individual to accumulate wealth, for the land was always present and pro-
vided all the necessities of life. Thus periods of hard work were followed
by long periods of relaxation. These values of the ancient culture to some
extent still condition the life of the average Hawaiian in the Häna district.
To have an adequate amount of leisure the Hawaiians may be contented
with earnings that put them on a subsistence level.81

The Hawaiian children of Häna were required to attend school through


the eighth grade, as were children throughout Hawai‘i between 1900 and
1930. However, Hawaiian children tended to be indifferent to Western edu-
cation. The culture of the classroom did not match the culture of the family.
The conflict between the Hawaiian values the children lived day by day and
the Western values the schools taught them tended to alienate Hawaiian chil-
dren from achieving in the school. One manifestation of Hawaiian children’s
poor adjustment to school was their high rate of absenteeism.
Yamamura observed this problem and attributed it to the conflict between
Western values and Hawaiian values. He even suggested that Hawaiians’ absen-
teeism from school may have been one form of resistance to a culture that sub-
merged native culture. He believed that the indifference of Hawaiian parents
to the school system was a form of rejecting domination by Western values:

The economic status of the Hawaiian families, their passive resistance to


the ideals of the west, and the conception of responsibility of older for
younger children sponsored by the solidarity of the kin group are impor-
tant causes for the frequent absences of the child from school.

107
chapter three

Even before the child comes to school, he has acquired, to a certain


extent, an immunity to the teachings of the western culture. Therefore,
much of the school work is passed off as inconsequential. Such patterns
of action make possible the continuance of ideals and values of the native
culture, hindering the complete assimilation of the Hawaiians into the
American culture.82

Harry Mitchell explained the conflict he experienced between what was


taught in school and what was practiced at home. He also described the pun-
ishment he received for speaking Hawaiian in the school.

You go school, learn. You come home, you try apply. They [grandparents]
scold you. You gotta live their lifestyle. Not the one they teach you from
the school. The school, junk eh. We only play in the damn school. No
study. Play hookey, eh. Go fishing. Forget it. Go hunting. They punish
you. You pull weeds they catch you speaking Hawaiian. They make you
stand in the corner on one leg till you fall asleep. No way, eh.83

The conflicts the Hawaiians in the Häna district experienced between


Western and Hawaiian values were mitigated in different ways. According to
the oral accounts of küpuna who were interviewed, some family members left
Häna to assimilate into urban living in Wailuku or Honolulu, while others
chose to stay in Häna as their parents had before them and to maintain their
Hawaiian way of life.
Yamamura also reported that the Häna Hawaiians, lacking access to West-
ern medicine, continued to rely upon lä‘au lapa‘au, or traditional Hawaiian
healing practices, for their health care. One of his informants described how
her father had fallen off a small cliff and smashed his face. Thought to be dead,
he was nonetheless taken to a kahuna lä‘au lapa‘au, or Hawaiian herbal healer,
who pounded medicine from native Hawaiian mountain plants to heal the
woman’s father. A second informant related the experience of a man whose
foot was smashed by a rock. A kahuna lä‘au lapa‘au was engaged to nurse the
man back to good health. The sister of a third informant fell from the rock
near the Häna lighthouse and broke her hip. The kahuna lä‘au lapa‘au prayed
for her to get well without suffering too much pain. The young woman recov-
ered and suffered very little pain while recuperating.84
These traditional approaches to healing were commonplace in Häna and a
natural part of the people’s lives. Yamamura’s informants also provided
accounts about akua lele, large balls of fire with tails, which were regarded as

108
häna: from ko ‘ olau to k aupö

omens of misfortune; and lapu, the wandering spirits of the dead, who usually
bother persons traveling at night with food such as pork or fish.85 The inform-
ants also believed in menehune, guardian spirit ‘aumakua, and night marchers
on Po Käne. They also shared a list of dos and don’ts to avoid misfortune,
from not whistling at night to not picking a lehua flower to not eating bananas
when one goes fishing.86

Subsistence Resources of Hâna


Overall, life in the district of Häna, thanks to the abundance of natural
resources available to the Hawaiian people who lived there, was pleasant and
healthy. Several of the küpuna interviewed by Mary Kawena Pukui confirmed
this observation. For example, Mrs. Kapeka Ka‘auamo concluded: “Maika‘i,
maika‘i këia ‘äina, maika‘i këia. Ho‘opiha mau i ka ‘öpü, ‘a‘ale pilikia ka ‘öpü”
(Excellent, excellent this land, this place is fine. The stomach is always filled,
the stomach is not troubled).87
The numerous streams that flowed down from Haleakalä to the ocean
along the rugged Häna Coast not only provide the Häna Hawaiians with an
abundant supply of fresh water, they also serve as a habitat for native Hawai-
ian aquatic life, which was an important source of food for the Häna Hawai-
ians. Küpuna from the Ko‘olau sections of the district all described to Pukui
the importance of ‘o‘opu (gobbi fish), ‘öpae (shrimp), hïhïwai (limpet), and wï
(limpet) to their regular diets.
There were many types of ‘o‘opu in the streams of Häna. Harry Mitchell
named five that he was familiar with catching and eating—näkea, näpili, ‘owau,
hi‘ukole, and ‘alamo‘o.88 Küpuna Josephine Medeiros named two additional
types, the ‘akupa and the ‘apohä. The näkea type was either fried or salted and
then baked with ti leaves for a dish called läwalu. The ‘owau was usually good
in soup. The small ‘o‘opu näpili was usually dried and then eaten.89
The streams were filled with ‘o‘opu between 1900 and 1930. Harry Mitch-
ell recalled catching two or three 30-gallon buckets full of ‘o‘opu overnight
when flood waters washed them downstream to the ocean. The first heavy
rains usually arrived in August or September, carrying the ‘o‘opu to the ocean
where they spawned. Once she laid her eggs, the mother ‘o‘opu died. The
baby ‘o‘opu, called hinano, would hatch and develop in the salt water from
August or September through November. The salt water made them strong
enough to swim upstream, where they would mature. About November, the
hinano began to make their way upstream to the large freshwater pools in the

109
chapter three

mountains. Their migration upstream coincided with the arrival of the migra-
tory birds from the north, which fed upon the hinano as they made their per-
ilous journey to the uplands.90
Being intimately knowledgeable about the life cycle and habitats of the
‘o‘opu and other aquatic animals enabled the Häna kua‘äina to plan out what
types of food sources to gather at different times of the year. Mrs. Pü, of Pa‘u-
wela, recalled how Wailua was full of ‘öpae, or black mountain shrimp. They
used to eat it with a mountain fern that the Maui people called pohole but that
is more widely known as hö‘i‘o. Mrs. Ka‘auamo also recalled the abundance of
‘öpae kuahiwi or mountain shrimp in the Ke‘anae and Wailua area.91 “Mullet
also spawned in the Wailuaiki district. They would usually go where the
stream met the ocean to spawn, around December. At that time, millions of
baby mullet could be seen swimming around. Hïhïwai and wï were also plen-
tiful in certain streams at the points where they flowed into the ocean.” 92
While the aquatic life that thrived in the Häna streams was an important
source of protein to the Häna kua‘äina, the stream waters also made possible
the widespread cultivation of the Hawaiian “staff of life,” or kalo. Mrs. Ka‘au-
amo of Wailua expressed the importance of taro to the Hawaiians of Häna:
“Taro is perhaps most important here. If it wasn’t for taro perhaps all the peo-
ple here would soon perish.” 93
Aside from taro, the patches also sustained important aquatic life, which
the Häna kua‘äina relished. These included the püpü pake, or snails, which
resemble escargots, and the ‘ula‘ula goldfish.
In the drier areas of Kaupö, Hawaiians planted sweet potatoes, pumpkins,
and dry land taro for home consumption. Patch after patch of sweet potato was
planted. Just as taro was the staple food for the people of Ke‘anae and Wailua,
the sweet potato was the staple food for the Kaupö families. The plantings did
especially well during the rainy season. As their ancestors had taught them,
they followed the annual cycle of the rain. About one month before the rainy
season began, they would carry dirt down from the mountains to the coast in
lauhala baskets and fill holes in the lava in preparation for planting. Children
also helped to carry some dirt in lauhala bags. While on the coast, the Hawai-
ians would subsist on fishing and various gourd plants such as Hawaiian water-
melon, ipu oloolo, ipu nühoulani, pumpkin, and pohä or ipu ‘ala, which were cul-
tivated in the pockets of lava and nurtured by the rain. When the vegetables
matured they were consumed. After a period of about six months, just when
the climate became dry, the families would make the return journey to their
upland habitation sites.94

110
häna: from ko ‘ olau to k aupö

If there was a drought, the families of Kaupö visited their relatives in Kïpa-
hulu to gather and get food. Most of the families in the two districts were
related to each other, and exchange and sharing were common and expected.95
A Hawaiian riddle for Kaupö and Kïpahulu linked the two districts and noted
the importance of sweet potato cultivation as a staple food for the people of
the districts: “Pö nä maka, a i ka pahulu ke ola” (The eyes become dim with
hunger, the hunger is appeased by old food patches). The saying originated
when a famine developed during a drought and the starving people of the dis-
tricts became hollow-eyed. They were able to survive by eating broken pieces
of sprouting potatoes left in the mounds by sweet potato harvesters to grow
to maturity.96
Pigs and goats thrived in the forested uplands throughout the Ko‘olau dis-
trict, and the Häna kua‘äina would hunt them from time to time and salt, dry,
and smoke the meat. Many families also raised a few pigs, cows, and chickens
for home consumption.97
As with their ancestors, the ocean along the entire district, from Ko‘olau
to Kaupö, continued to provide the Häna kua‘äina with fish, crab, shellfish,
and seaweed. They would gather shellfish and limu along the shore; go deep-
sea fishing in canoes; lay nets, including large hukilau nets in the bays; dive;
pole fish; and even cultivate fish in ponds or by feeding the fish at designated

Figure 18 About 2,000 fish each averaging 8 inches in length were harvested by the Hâmoa
community hukilau in Hâna. 1936. Harold T. Stearns, Bishop Museum.

111
chapter three

ko‘a or traditional fishing grounds. Salt for the district continued to be annu-
ally gathered at Nu‘u.98
Throughout the 1930s Hawaiian women continued the practice of weav-
ing hats and mats out of lauhala—both the light and dark varieties. Küpuna
informants also identified many other plants used for weaving, including the
‘ekaha or bird nest fern; the pämoho fern; the red ti plant known as “Kaupö
Beauty”; the häpapa pueo kalo, a white taro top; banana bark, especially the
‘ele‘ele or black variety; stems of pili grass; large leaves of the ‘ulu or bread-
fruit tree; the nänaku sedge; the ‘iwa‘iwa fern; and the ‘ie‘ie vine.99

Ke‘anae and Wailuanui in the Territorial Period


The ahupua‘a of Ke‘anae and Wailuanui are the first of the Häna districts to
be reached on the road to Häna by travelers setting out from Ha‘ikü. The
famous and picturesque taro patches of Ke‘anae have been continuously cul-
tivated by the families who lived there from the time that their ancestors
created them by carrying soil to cover the barren lava penninsula.100
Wailuanui was favored with more fertile and extensive agricultural lands.
Three streams provided the water for the patches —Wailuaiki, Waiohue, and
Hanawï.
In 1910 H. M. Ayres, a reporter with the Pacific Commercial Advertiser,
hiked through Ke‘anae to Häna along the East Maui Irrigation Company’s
Ko‘olau ditch trail. He reported what he observed in the feature section of the
Advertiser on September 4, 1910. The beauty of Ke‘anae’s lush forests and the
waterfalls cascading through ravines down to the magnificent ocean below
impressed Ayres, and he was also interested in the lifestyle of the Hawaiian
and Chinese families who had homesteads in the district:

At the house of Halemano we were made very welcome, supper being


ordered by our host at a Chinese restaurant nearby. He naively remarked
that poi and fish were not good for haoles. Halemano, who is postmaster
and political boss of the precinct, is a dignified old native. His house is on
the campaign circuit and when election time rolls round there are stirring
times at his residence. His daughter, Annie, is easily the belle of the district.
Keanae is a sugarless settlement, rice being the main industry of the
place. The natives live contentedly in their homesteads and are unusually
well informed on matters of the world, for dwellers in such an out-of-the-
world place. It is one of the prettiest settlements on Maui. Across from

112
häna: from ko ‘ olau to k aupö

Keanae is Wailuanui, a place well worth a trip over the ricefields, if one
has the time. Back of Halemano’s house is a natural bathing pool formed
by a cascade and large and deep enough to allow of a really good swim.
Many of the Keanae girls have Chinese husbands and appear to be quite
happy with them. They are better providers than the Hawaiians and this
probably accounts for the phenomena.
Before leaving Keanae we offered to buy a squid stone from Halemano
but the old man refused to part with the relic, declaring that it was his
wife’s and that he didn’t need the money.101

Harry Mitchell observed that when he was young the Hawaiians usually
leased out the lowlands downstream, where the water tended to be too warm
for the taro to grow well and the taro often spoiled. The Hawaiians contin-
ued to cultivate taro in the cool uplands for home consumption and for sale.
Many of the Chinese men who moved to Ke‘anae intermarried with Hawai-
ian families of the district and helped to cultivate the Hawaiian lands.102 Ke‘a-
nae became a center for rice cultivation through 1927, when, as in Waipi‘o,
low prices for California rice drove the Ke‘anae rice growers out of the busi-
ness. At that point Ke‘anae farmers shifted back to the cultivation of taro for
market.103
The Ke‘anae kua‘äina developed a system of exchanging labor for tools and
materials with the Chinese rice farmers. The enterprising Chinese imported
lumber, pots, shovels, sickles, muslin, and other tools and materials for use in
the production of rice. Occasionally, Hawaiians would perform work for the
Chinese, such as plowing and hauling bags of rice in exchange for tools and
materials rather than for money. According to Mitchell, his grandfather and
others worked for the Chinese in order to get lumber and metal pots:

Chinese bought lumber, eh, and they got a big pot. Yeah, you know for
cook their rice. And the Hawaiians like for cook taro . . . So now you gotta
work, eh, maybe three, four months before they give you one pot, eh.
Maybe every month they give you one lumber, eh. You know 1 x 12 x 8
feet long . . . And those Hawaiians, they tough, they work for the Chinese
because they look, see, that pot—so everybody start getting pot.104

Prior to using pots, the Hawaiians cooked their taro in the imu. The taro
came out dry and hard and had to be pounded while it was still hot. Boiled
taro was moist and could be easily pounded even when cool.
The Hawaiians also exchanged work for clothing and canvas. If they

113
chapter three

worked two days they could earn a pair of pants or a raincoat. The Chinese
would coat canvas with linseed oil to make waterproof raincoats.105 Work was
also exchanged for shovels, picks, and saws. In addition to exchanging work
for materials, with the Chinese, work was also exchanged between households
or ‘ohana, as in traditional times. The families would help one another in the
taro patches and exchange labor and food:

Every weekend, lü‘au someplace. You know why? They laulima. This
family go help the other family clean taro patch. Then if I go clean your
taro patch, you kill the pig. Next week maybe me. Keep on going like that,
huh. We go on the horse—go this house, they singing and everything. Play
mandolin, they get guitar, ukulele. My uncle, he play the clarinet. He was
good too, wow.

The Ke‘anae kua‘äina also had a system of barter and exchanged with Kona
and Moloka‘i. Taro in the form of pa‘i‘ai would be exchanged for ‘öpelu or
akule from Kona and squid from Moloka‘i:

Most time they exchange. You know, then they get boats. They put pa‘i‘ai
in the box— maybe four or five hundred pounds — put ’em on the boat.
They go Moloka‘i. That same crate come home full with dry squid. Or
they go Kona, eh — that same crate come home full with dry ‘öpelu. And
they distribute. How many pa‘i ‘ai you get four? Five? Well here — one
pound ‘öpelu, all dry, you know — one kauna — forty. You get more? Then
two kauna. That’s how you know— divide. That’s how they do—trade.106

Subsistence work and exchange for goods was the dominant form of work
in Ke‘anae until the Häna road was constructed by prison labor based at the
Ke‘anae Prison Camp. From 1890 till 1920 Ke‘anae’s commercial directory
listed most of the Hawaiians in the district as taro planters. In 1920 the direc-
tory listed three Hawaiians as laborers. However, in the 1930 directory,
Hawaiian taro planters and homesteaders were listed as laborers and truck
drivers employed by the East Maui Irrigation Company, the road department,
and Ke‘anae Prison Camp.107
In 1934, when the anthropologist E. S. Craighill Handy conducted a survey
of the natural resources of the Häna Coast, he observed that the Ke‘anae Val-
ley mauka of the road was a water reserve with no inhabitants. The wetland
taro patches for which Ke‘anae is famous were entirely located on its broad,
flat peninsula of lava, which extended about a mile out into the sea from the

114
häna: from ko ‘ olau to k aupö

base of the cliffs. Polaukulu Stream provided the water for the patches, which
were still under active cultivation in 1934. However, at Wailuanui, only half
of the taro terraces originally developed by Hawaiians of the district were still
being cultivated by Hawaiians in the 1930s.108
Josephine Kauakeaohana Roback Medeiros described the livelihood of the
people of Ke‘anae:

Ke‘anae people there make a living by fishing, gathering oysters, sea shells,
catching shrimps from the mountain streams, clams, catfish, water shells
and planting their own sweet potatoes and taro which they used for starch
and vegetable. Also they hunt for wild pigs, fowl, etc. Very few of them
work for the county as road clearing gang, and only work for 15 days out
of a month at a time, even only 8 days a month sometimes—so they depend
a lot by the ocean and mountain for their livelihood. They also market
some of their catch and harvesting.109

A prison camp was built at Ke‘anae in 1926 to house the prisoners who
would construct the road, including several bridges from Kailua to Häna. As
described above, when the road was completed in 1927, men from Ke‘anae to
Häna town were hired to maintain the road, especially during the rainy season.
In 1934 the prison camp was converted into quarters for the Civilian Conser-
vation Corps. This federal program, created by President Franklin D. Roose-
velt to provide jobs to get the United States through the depression, brought
in men from other parts of Maui and other islands to plant thousands of euca-
lyptus and other introduced trees throughout the Häna coast. Eventually, in
1949, the camp was acquired by the YMCA. Part of the land area continued
to be used as a base yard for the Maui County public works projects.110

Nâhiku to Hâna Town, Under the Territory


Nähiku, like Wailua, was a fertile ahupua‘a where Hawaiians had cleared and
terraced the lands for irrigated taro cultivation. The land to the east of Nähiku
sloped gently down to the ocean. There were no large streams or gulches.
Along the shore was a hala forest that extended from ‘Ula‘ino to Häna. Hawai-
ians living in this part of the Häna district also raised dry-land taro around
their homes, together with other food plants.111
Other agricultural ventures were attempted along the Häna Coast. In 1899
the Nähiku Rubber Company planted thousands of rubber trees makai of the

115
chapter three

road. After some initial experimentation in producing rubber, the company


incorporated in 1905. The American and Ko‘olau Rubber Companies also
established rubber plantations in the district. At one point there were more
than 25,000 rubber trees of different varieties growing in and around Nähiku.
By 1912 the rubber companies had begun to phase out their operations. A
former field worker recalled that a ten-hour day with a thirty-minute lunch
break netted him only 50 cents. At that rate for labor, the quality and quantity
of rubber produced on the wet Häna Coast was too low to make a profit.112
At the height of rubber production, Nähiku had a Chinese grocery and post
office; a plantation general store; Protestant, Mormon, and Catholic churches;
and a schoolhouse attended by twenty children. As shown above, there were
only 182 people living in Nähiku in 1930. By 1941 only fifteen Hawaiian fam-
ilies and two non-Hawaiian families lived in Nähiku, clustered around a one-
room school and the churches.113 After the rubber plantations closed, the res-
idents planted bananas as a cash crop and subsequently planted roselle for
jelly. Eventually, when these economic ventures failed, the population shifted
out of Nähiku.114
In 1910, when Ayres hiked through Nähiku, the community bustled with
activity related to the production of rubber. Although planted in rubber trees,
the surrounding tropical forest was still impressive to the visitor:

Every place has its peculiarities and characteristics; so with Nähiku. It is


rubber, first, last and all the time there. And while the population of the
place tend the trees, the owners work and dream of what the future holds
in store, for they are handling a new thing and it may yield them fabulous
profits or only a pittance.
There is a very good class of Hawaiians at Nähiku, industrious and con-
tented. The rubber affords them more or less constant employ and fish are
very plentiful off the shore. The natives working for Mr. Austin regard him
as a friend. He speaks their language fluently and both he and his mother
have, by their helpful attitude, endeared themselves in the hearts of the
Hawaiians of Nähiku.115

As the rubber companies phased out their operations in 1912, some of the
residents moved out of Nähiku. Those who remained resumed the cultivation
of native crops—bananas and taro. Some of the terraces below the settlement
were still under cultivation in 1934. In addition, each of the Hawaiian fami-
lies also cultivated dry-land taro patches around their homes.116

116
häna: from ko ‘ olau to k aupö

Hâna Town during the Territorial Years


Despite its isolation, Häna was a bustling little town in the 1920s and 1930s.
Beginning in 1919 there were two movie theaters, one at Ka‘elekü and the
other in Häna town, where there were two showings each night in addition to
a Saturday matinee. There were also fifteen different stores, three barber-
shops, a pool hall, and a choice of several restaurants. The small general stores
in Häna emerged as centers for commercial and social exchange.117 The Haiku
Fruit and Packing Company planted pineapple in Mü‘olea in the Kïpahulu
district in 1922 and built a cannery in Häna town in 1924. However, by 1927
the pineapple cannery had closed.118
In 1927 the Häna road, constructed by prison labor, linked the Häna Coast
to the “outside” world. After completion, it was maintained by Häna kua‘äina
hired by the territorial government. During the rainy winter season, mud-
slides, downed trees, and flooding streams frequently made the road impass-
able. County road work enabled many Hawaiian families to earn cash to sup-
plement their subsistence lifestyles.119 Eventually, trucks replaced trains as the
method of transporting the harvested cane to the mills, and road work
declined.
In 1930 Häna, as the center of commerce and plantation operations, had
65 percent of the population of the Ko‘olau district, the majority of which
were non-Hawaiian. Most of the Hawaiian families enumerated as living in
Häna in the 1930 census lived in the small self-sufficient communities outside
the town, from Ke‘anae through Wai‘änapanapa and from Hämoa to Wailua,
and outside the control and organization of the plantation camps. The Häna
planters had to import Chinese and later Japanese and Filipino laborers to
produce sugar in the district.
In 1934 E. S. Craighill Handy observed that in north Häna, above the sea
cliffs and lava caves of Wai‘änapanapa, was the native settlement of Hono-
kalani. Here and on the forest land called Helani, Hawaiians raised dry-land
taro primarily for home consumption. Given the lack of streams at Häna and
Hämoa, wetland taro was not cultivated. Hawaiians who lived there raised
dry-land taro on forested lowlands about two miles inland. In south Häna,
past Hämoa, streams begin to cut across the landscape again. Hawaiian home-
steads at Maka‘alae, Waiohonu, Pu‘uiki, Pohue, Pukuilua, Hä‘ö‘ü, Hulihana,
Müolea, and Koali had extensive wet taro patches, many of which had been
originally constructed by the ancient Hawaiians.120

117
chapter three

Kîpahulu, Territory of Hawai‘i


Traveling southeast from Häna, Wailua was the first valley outside of Häna.
It had extensive wetland taro terraces at four different levels sustained by
water from three streams that converged at Kaumakani. The Kahalawe Stream
waters provided water for patches on the steep slopes of Paehala. Most of the
patches in the valley and along the slopes were still being cultivated in the
early 1930s. Dry taro was also planted in unirrigated terraces where the steep-
ness of the slope allowed sufficient aeration of the soil without running
water.121
Historically, Kïpahulu was a district with rich and diverse agricultural
resources scattered throughout its vast valley. The moku of Kïpahulu included
many smaller ahupua‘a or valleys fed by streams. Historically, these little val-
leys were terraced and planted in taro. A sugar plantation and mill were estab-
lished approximately ten miles from Häna at the turn of the century and oper-
ated through the 1920s.
In 1922 Thomas Maunupau and Kenneth P. Emory of the Bishop Museum

Figure 19 Joe Kahaleuahi, shown here with his granddaughter, is a kua‘âina fisherman and
farmer of Kîpahulu, Maui. 1971. Franco Salmoiraghi.

118
häna: from ko ‘ olau to k aupö

traveled through Häna and Kïpahulu on their way to Kaupö. Their trip was
chronicled in the Nüpepa Kü‘oko‘a of June 1, 1922:

We arrived at Kïpahulu and saw the plantation on the upper side of the
road and the homes of the laborers. I saw the laborers cutting cane and the
cane cars bearing them to the mill. Now and then we passed a Hawaiian
house. Some people were farming, some pounding poi and some of the
Hawaiian mothers were plaiting mats. Our car continued and came to a
place called Kukui‘ula.122

Kukui‘ula Stream is where the ridge-and-valley trail to Kaupö began.


Twelve years before Emory and Maunupau made the trek to Kaupö, a staff
correspondent with the Hawaiian Gazette hiked the trail between Kukui‘ula
and Kaupö. His observations about the district were published in the Gazette
on September 6, 1910. His descriptions of the scattered Hawaiian households
and the sights, sounds, and flavors of the resources of the area provide an
excellent picture of life in the district.

The road was good till the first gulch was arrived at. The down-go wasn’t
so bad, but going up again was a caution. I rested a good quarter of an
hour in the shade of a lauhala tree half-way up.
On the Kïpahulu side of this gulch is a grass house, the occupants of
which, judging by the number of bamboos about the place, do a good deal
of fishing. This house is very picturesquely situated. For a way the trail led
by the sea, and by its side I passed an ancient canoe, a sad reminder of
other days, when the stone fences enclosed prosperous kuleanas, and
when the natives were thick upon the countryside . . .
The third gulch is waterless, but in the fourth, close to the house of
Inaiana, is a pool of fresh water, supplied by the stream above.
Passing up the far side of the fourth gulch, the smell of orange blossoms
was borne to me, and I soon located several trees full of excellent fruit, to
several of which I did ample justice. The guavas on this hill, by the way,
are very sweet and finely flavored.123

In 1922 the Haiku Fruit and Packing Company planted pineapple in Mü‘o-
lea in the Kïpahulu.124
In 1934 Handy found the sugar plantation closed and replaced by a cattle
ranch. Native Hawaiians lived in homes in the lower kula lands above the sea
and raised dry-land taro around their homes for consumption by their own

119
chapter three

households. Several small groups of taro terraces were still cultivated by the
Hawaiians who lived in the district. The Lolokea and Kalepa Valleys were not
in active cultivation in 1934. Hanawï Valley was watered by the ‘Alelele
Stream, and some of the terraces there were being replanted in wet taro in
1934. Nuanualoa, the last valley before Kaupö, had a handful of Hawaiian
households in 1934. They cultivated wet taro in the traditional terraces.125

Kaupò, Territory of Hawai‘i


Five miles down the coast from Kïpahulu is Kaupö. The district is arid along
the seacoast but receives a moderate amount of rainfall three to four miles
inland at an elevation of 2,000 feet, where the land ascends to Haleakalä
Crater. Manawainui is the large stream that drains the higher slopes east of the
Kaupö Gap, an access point into Haleakalä Crater. The valley of the stream is
canyonlike. The Hawaiians of Kaupö got their water from numerous springs
in the area, such as Punahoa and Waiü. In Manawainui Valley, there were sev-
eral large springs until landslides caused by heavy rains and earthquakes cov-
ered the springs over and broke the pipelines that had carried the water to
households.126
At one point in the early 1900s attempts were made to cultivate wheat at
Kaupö.127 A reporter with the Hawaiian Gazette who traveled to Kaupö in
1910 described life in the area:

At Kaupö are to be found Hawaiians living industriously and contentedly,


as they do in the few places on the Islands which one has to go off the
beaten track to reach. Sweet potatoes are largely grown and they have to
take the place of poi, of which there is none here. There are plenty of fish,
however, and some of the finest ‘opihis procurable anywhere. There are
several grass houses here and the garden of every house is overrun with
geraniums, carnations and beautiful roses. A fine new schoolhouse is being
built here to provide for the large number of children in the district. This
evening a single party, gaily decorated with leis, is going from house to
house serenading.128

The reporter also sent his story to the Pacific Commercial Advertiser, which
ran the report as a series of articles on September 5 and 9, 1910. The Septem-
ber 9 article carried a description of the living conditions of Hawaiians who
lived between the new schoolhouse and the Kaupö Gap. It described the tra-

120
häna: from ko ‘ olau to k aupö

ditional houses that the reporter observed along the way as well as the types
of foods cultivated and animals raised by the native Hawaiians of the district.
Problems that the kua‘äina had with marketing the fruit were also identified:

It is nearly two miles from the schoolhouse to Marciel’s house, the trail
running past several pohaku houses, grass thatched, and all-grass houses.
The occupants raise pigs and sweet potatoes and working a little and rest-
ing a great deal, appear to drift in from day to day happily enough.
All take a turn at fishing at times and the toothsome aweoweo is so
abundant hereabouts that it helps out the commissary problem materially
during the moonlight season.
Excellent oranges and limes are grown at Kaupo, the former being
sweet and finely flavored. The Kaupoans could earn many a dollar by ship-
ping their fruit to Honolulu, could they rely on a regular steamer. As it is
impossible to tell when a steamer is going to call. The steamer Claudine
used to call here regularly once a month but the service was discontinued
last July, the steamer now going direct from Hana to Hawaii and return.
Several shipments of limes and oranges from here have rotted on the
wharf waiting for a steamer to call. The Kaupö people suffer in another
way by lack of a regular steamer service. Many of them order their house-
hold supplies from Honolulu and are often reduced to famine rations as
far as some of the necessities of life are concerned.129

In 1922 Maunupau and Emory took note of the majesty of Kaupö: “Kaupö
is indeed a green land and so is Häna. They look so open and pleasant to live
in because the wind is always blowing. The coast is good to look at and fine
for inshore fishing. The whole of Kaupö faces West Hawaii. Looking upward
one sees the majestic Haleakalä mountain, the Kaupö Gap and many small
waterfalls.” 130 Nu‘u is an ahupua‘a within Kaupö. In 1922 Maunupau and
Emory observed five Hawaiian households in the area: “When we came to
level land, that was Nuu proper. It is a seaport and cattle is [sic] shipped from
here. This was a landing place for fishermen in the olden days and even down
to the present. There are about five houses at Nuu and the inhabitants are all
Hawaiians.” 131
In the 1930s the federal Works Progress Administration financed the con-
struction of a motor road from Kïpahulu to Kaupö. In 1934 Handy observed
that almost the entire area of Kaupö was ranch land, although Hawaiians who
worked for the ranch still raised sweet potatoes for home consumption.132

121
chapter three

Working as a ranch cowboy was the type of job Hawaiians enjoyed. It involved
intense periods of strenuous activity followed by periods of relaxation. One
such intense period was when cattle were shipped out to be slaughtered and
marketed from the landing at Nu‘u. Mrs. Marciel, an informant for Mary
Kawena Pukui in the 1960s, was born and raised in Kaupö, near Nu‘u. The
Marciel family once owned Kaupö Ranch, but during a great drought, they
sold it to the Baldwin Estate. Marciel described what was involved in shipping
the cattle from the landing:

The boat used to come over there. We had a shipping pen. Where we
shipped our cattle before. The boat came outside and we shipped the pipi
out to the boat— swim, swim out. Then they swing them up into the boat.
You have to get good horses to take the pipi out. Huki me ka lio—ho‘au i ka
pipi (tug with the horse—make the cattle swim).133

Marciel explained that her family planted sweet potatoes there during the
rainy season when the earth was soft. They used the planting enclosures left
behind by the ancient Hawaiians. They planted the piko variety, which bore
potatoes in four, five, or six months; and the möhihi potato which took several
months to bear but could be stored for several months after harvesting. Her
family also planted the ipu ‘awa‘awa calabash gourds and pumpkins at Nu‘u.

Figure 20 Kaupò Ranch hired Hawaiian men living in the district to work as cowboys. 1923.
Kenneth P. Emory, Bishop Museum.

122
häna: from ko ‘ olau to k aupö

The sweet potato, gourd, and pumpkin plants bear well in dry land and flour-
ished at Nu‘u. The people of the district also wove hats using nänaku sedge
as well as ‘iwa stalks for material.134
Marciel described an active Hawaiian community at Nu‘u and Kaupö that
was involved in fishing, making salt, and planting; some worked on the ranch.
According to Marciel, there were five boat houses and two canoe houses at
Nu‘u prior to the 1946 tidal wave. There was a medicine house, right where
Kaupö Landing is, in a small hale (house). Lapa‘au, or medicinal plants, are
everywhere in the area; they were tied in bundles and kept in the building.
This was also the salt house. Great schools of akule frequented the bay, and
the salt was used to dry the akule. Right against the pali, where the landing
and the medicine house was, was a canoe house that had Hawaiian canoes
in it.

Figure 21 Kaupò Ranch shipped


out cattle from Nu‘u. May 1922.
Bishop Museum.

123
chapter three

At Nu‘u, all the stones with small hollows were put on the edge of the pond,
and people would put ocean water in them. The sun would evaporate the
water, leaving the salt. They used wooden spoons to scoop out small amounts
of salt. Nu‘u Bay had a nice pond, behind, which is now all overgrown. Peo-
ple used to keep it clear of hau trees. The Navy used kaili hau for rope on an
almost daily basis, and this kept the hau in check.
Nu‘u Bay has a black stone beach. The Hawaiians used the stones for
weighting their nets with palu, or bait. The stone bait consisted of a flat weight
with two ears; they tied it and put the sweet potato peelings inside, covered
the bag, rolled and twisted it, and laid it in the net. They also used pumpkin,
squash, ipu, and sweet potato as bait. When they made sour potato mash,
which looked like a big pudding, they kept the liquid part, fermented it, and
drank it as liquor. There was a cave with a water well in front if it and also a
Hawaiian house. Beach equipment was stored there as well. Photos record
that Hawaiians in malo or loincloths launched canoes in the bay. The fishing
houses consisted of Japanese-style skiffs, with miles of net. People that lived
there were part of a hui or organized group. They fixed nets, made floaters, and
made salt. Nu‘u was also famous for holoholo he‘e, a very big squid with short
tentacles and a large body. The uhu, a fish that Hawaiians like to eat fresh,
would come in schools, turning the sea red beyond the bay. As many as thirty
people would gather, start a fire, make coffee, and go out fishing, one group
at a time. Even before the fishing was over, they would start loading these fish
up on donkeys and try to get mauka and give it away. This was the ancient
Hawaiian way of life: when there was something to be had, everybody shared.
Hukilau nets were used more often in those times.

Changes during and after World War II


The forces of change for the Häna coast included World War II, the phasing
out of sugar plantations, and the 1946 tsunami.
The war came to Häna in January 1942, when a Japanese submarine tor-
pedoed the transport ship General Royal T. Frank in the ‘Alenuihähä Channel.
Twenty-nine persons were killed, but thirty-three survivors were taken to
Häna and cared for at the Häna School gymnasium.135
In December 1942 Governor Ingram Stainback tried to assist the war effort
by sending forty inmates from O‘ahu Prison to the Ke‘anae Prison Camp to
revive the old Nähiku rubber plantation in the hope of yielding 20,000 to
50,000 pounds of crude rubber annually.136 The venture was not successful.

124
häna: from ko ‘ olau to k aupö

Residents in Ke‘anae and Wailuanui reported that they cached explosives in


two caves at Nua‘ailua to use to blow up the Häna Road in the event of a Japa-
nese invasion.137 By far the biggest impact of the war was the induction into
military service of the Häna Coast’s young men, some attracted to high-pay-
ing jobs at Pearl Harbor. Many never returned to their subsistence livelihoods
on the Häna Coast when the war ended.
The San Francisco entrepreneur Paul I. Fagan acquired the Häna Sugar
Company from the Unna brothers in the 1930s. In 1944 he closed the Häna
sugar plantation and began to transform the former plantation lands into a
cattle ranch. After the war, sugar workers organized under the International
Longshoremen and Warehousemen’s Union ( I.L.W.U.), increasing the cost
of producing sugar, and C. Brewer closed Kaelekü Plantation.138 In 1946 Fagan
opened the Hotel Häna-Maui; from then on he focused his marketing efforts
on wealthy travelers who could afford to spend luxury time in a world-class
resort.
A huge tidal wave hit the islands on April 1, 1946. Of the fourteen persons
who died on Maui, twelve died along the Häna Coast. In the Ke‘anae Penin-
sula, several buildings were damaged, and two persons died.139 Residents whose
homes on the peninsula were destroyed resettled along the Häna highway.
During the 1950s the population of Häna dropped to 500, the low point
for the twentieth century. In 1956 Fagan built the Häna Ranch Center, which
included a post office, a bank, a barber shop, and a lunch room. He also built
a center at the harbor, which he named Helene Hall in honor of his wife, and
donated it to the county for the community. With the money from Helene
Irwin Fagan’s estate, the Fagans established a trust fund for the nonprofit
Häna Community Association to oversee the recreational needs of Häna’s
children.
The paving of the fifty-five-mile Häna highway in 1962 opened the coast
to tourists who usually drove in and out for the day. However, the curves and
one-lane bridges saved the Häna coast from the full-scale development seen
in other parts of Maui such as Lahaina, Ka‘anapali, and Kïhei.
Over the years, some of the clients of the Hotel Häna-Maui bought land
along the Häna Coast for retirement or vacationing. These wealthy new res-
idents of the Häna Coast have included Sam Pryor, a vice president of Pan
American Airways; the aviator Charles Lindbergh; the actors Jim Nabors and
Kris Kristofferson; and the singers George Harrison and Willie Nelson.
When Paul Fagan died in 1970, the community erected an impressive lava
rock cross upon a grassy knoll overlooking Häna Town as a memorial.

125
chapter three

Hotel Häna-Maui and Häna Ranch were sold to the Rosewood Corpora-
tion of Dallas, Texas, in 1984. In 1989 Rosewood sold the hotel and ranch to
Keola Häna-Maui, an international investment group made up of Japanese,
British, and Hawaiian investors.140
By the end of the twentieth century there were 2,000 residents along the
Häna coast.
Changes to Kïpahulu developed with the acquisition of Kïpahulu Valley by
the U.S. Department of the Interior for inclusion within Haleakalä National
Park. The pristine upper valley was incorporated into the park in 1951. The
lower valley and coastline, including the pools of ‘Ohe‘o, were added in Jan-
uary 1969. Haleakalä National Park is recognized by the United Nations as
an international biosphere reserve. Park policy now limits development in
Kïpahulu to park facilities and shapes the lives of the kua‘äina of Kïpahulu.
The Kïpahulu section of the park attracts 500,000 visitors annually.141 In 1974
the world-famous aviator Charles Lindbergh decided to spend his last days in
Kïpahulu, where he died and is buried. His grave in the Palapala Ho‘omau
Church cemetery has become an attraction for tourists who venture to drive
along the winding Häna highway. The road built by the Works Progress
Administration during the depression is usable, but subject to periodic clo-
sures after storms and landslides. Sections of the road have been paved and
some of the one-lane bridges replaced with modern two-lane bridges. Kïpa-
hulu was also designated an International Biosphere Reserve in 1980 in recog-
nition of its unique and diverse ecosystem. The area from Kïpahulu to Kaupö
is rural. Commercial horseback riding stables and local farming are the pri-
mary business in the district.
Kaupö is largely enjoyed by the local people of Maui as a pristine and
unspoiled refuge. The Kaupö General Store at the center of the district is the
only private commercial store between Häna and ‘Ulupalakua. It serves the
small pool of local residents, tourists who pass by in rental cars, and hikers
coming out of Haleakalä Crater through the Kaupö Gap. Kaupö Ranch is the
main employer for a handful of residents in the district.
The panoramic view of the southern slopes of Haleakalä Crater, the island
of Kaho‘olawe, and Hawai‘i Island is unobstructed, uncluttered even by util-
ity lines and poles. The primary impact upon the traditional cultural landscape
comes from cattle and goats. The small pond behind Nu‘u Bay was fenced by
the U.S. Fish and Wildlife Service in 1997 to keep the cattle out, and it has
become a reserve for native birds such as the koloa (native duck) and Hawai-

126
häna: from ko ‘ olau to k aupö

ian stilts and coots. Huakini Bay and Nu‘u Landing are regularly frequented
by weekend campers. During the summer, families camp there for weeks at a
time; the setting engenders a feeling of self-reliance and independence. Here,
Maui residents can feel in contact with nature, physically and spiritually.
Native Hawaiian campers and fishermen feel a connection to their cultural
roots.

Ke‘anae-Wailuanui in the Late Twentieth Century


As late in the twentieth century as 1994, the State of Hawa‘i Department of
Land and Natural Resources recognized the Ke‘anae-Wailuanui ahupua‘a as
an Historic Cultural Landscape with unbroken continuity to the original pre-
contact uses of its lands for Native Hawaiian subsistence and cultural activi-
ties. This led to a study titled Kalo Kanu O Ka ‘äina: A Cultural Landscape Study
of Ke‘anae and Wailuanui, Island of Maui, in which I was privileged to partici-
pate as historian. The study documented the importance of protecting Ke‘a-
nae and Wailuanui because of the persistence of Native Hawaiian cultural
practices in relation to taro cultivation and subsistence activities. I will close
this chapter about the Häna district with the mo‘olelo shared with me by the
descendants of the earliest Native Hawaiian inhabitants of the Häna Coast,
who, as modern kua‘äina, continued to rely upon their ancestral knowledge to
provide for their ‘ohana. These experiences verify that key elements of Native
Hawaiian culture centered around subsistence activities have persisted in
Ke‘anae-Wailuanui. In fact, Ke‘anae-Wailuanui exemplifies the perpetuation
of Native Hawaiian culture into the twenty-first century within the Häna dis-
trict as a whole.
In 1990 the area known as Keanae-Wailuanui had a population of 241 peo-
ple living in sixty-seven households.142 In the 1994–95 school year, the Ke‘anae
Elementary School had eleven students in kindergarten to third grade. Adult
education programs were held at the school in the evening. During the day,
the community organized senior citizen programs at the school so the küpuna
could interact with the children.143
According to the kua‘äina who were interviewed, the original region which
continued to be used and cared for by generations of Native Hawaiian kua-
‘äina of Keanae-Wailuanui, extended from Makapipi in the East to Hono-
manü in the West, and all the way mauka to Pohaku Palaha on the northern
rim of the Haleakalä Crater.144 The Makapipi stream and forest access road

127
chapter three

forms the boundary between Keanae-Wailuanui and Nähiku on the east. The
Kaumahina ridge above the Honomanü stream forms the Keanae-Wailuanui
boundary on the West.
Although the traditional ocean boundary for an ahupua‘a is the reef, or one
mile out where there is no reef, most of the Keanae-Wailuanui residents who
regularly fished went out as far as the offshore buoy, seven miles out to sea:
that is how far they consider their traditional access area in the ocean to
extend. The Ke‘anae-Wailuanui kua‘äina fished and gathered ‘opihi or limpets
by boat along the coast from Kailua in the west and over as far as the Häna
airport, to a place called Honomae‘ele in the east (also popularly called Pine
Trees). The entire shoreline and nearshore waters have abundant marine
resources.
Along the mauka section of the ahupua‘a, the common practice was to fol-
low the ditch trail looking for signs of pigs and then to follow the tracks into
the forest. The 2,000-foot elevation was usually as high as most hunters and
gatherers needed to venture to get pigs. Nevertheless, those who hunted and
gathered could go as high as 3,000 feet, into the Waiakamoi area and up to
Olinda. Some (Moki Day, Doug Chong, Paul Sinenci, Harry Pahukoa, and
Keola Hueu) have hiked up the mauka trail all the way into Haleakalä Crater
through the Ko‘olau Gap.
The families who live in Ke‘anae-Wailuanui are part of ‘ohana from Kailua,
Nähiku, Häna, and Kaupö. They also have continued to exercise their access
and gathering rights in those ahupua‘a. For example, Awapuhi Carmichael, a
kua‘äina of Ke‘anae, said, “All I know is that my mom said that because they
came from Kaupö all the way back, we use from Kaupö to Punalu‘u, that’s near
Kailua. And we have always practiced our gathering rights that way, from
Kaupö all the way to Punalu‘u, by Kailua.”
The area included within the boundaries described, therefore, constitutes
the true traditional Keanae-Wailuanui cultural landscape, as established
through custom, use, and practice for cultural, religious, and subsistence pur-
poses by the people of Keanae-Wailuanui. The resources used for traditional
customs and practices in the Ke‘anae-Wailuanui area are located at all eleva-
tions, from the ocean to sea level and up to over 10,000 feet at the summit of
Haleakalä. Rainfall in the area varies from 40 inches a year to over 300 inches
a year in the wet rainforests. The forests above Ke‘anae-Wailuanui are part of
the east Maui watershed, which is the largest single source of surface water in
the state, with an average harvested flow of 60 billion gallons a year. In addi-
tion, artesian water is harvested from wells within the watershed.

128
häna: from ko ‘ olau to k aupö

The area mauka of the Häna highway is dominated by the state-owned


Ko‘olau Forest Reserve and Hanawï Natural Area Reserve. Vegetation varies
according to elevation. The upper elevations have grass and shrublands. The
forest begins at approximately 7,200 feet and extends down to the coastline.
It is dominated by ‘öhi‘a but includes native koa and introduced trees such as
eucalyptus, strawberry guava, common guava, paper bark, rose apple, and java
plum. Some of the most intact and extensive native forests occur here and sup-
port the state’s greatest concentration of endangered forest birds. The forests
are home to twelve species of native Hawaiian birds, native insects, snails, and
other invertebrates. The native species and ecosystems provide a stable and
beautiful watershed which would be nearly impossible to replace if destroyed.
Clearly, Native Hawaiian subsistence practices have resulted in a sustainable
use of the natural resources of these ahupua‘a.

Traditional Practices and Way of Life


What is special about the way of life in Ke‘anae-Wailuanui? Here is what the
modern-day kua‘äina of Ke‘anae-Wailuanui have observed:

Life is easier here than the outside world. Here, can hunt, can fish, can
farm. (Charmain Day, 30s)

Over here, I got free things to eat . . . That’s how we live over here, ‘öpae,
o‘opu, hïhïwai, ‘opihi. The sea clear, we go get ‘opihi, fish. (Enos Akina, 94)

We’re lucky. This is the place. We’re blessed, we took such good care of
everything we had. Awapuhi said, and til today I remember the word she
used is “selfish.” The word selfish is for the people over here, to protect the
area. Selfish doesn’t mean that we going to make money, selfish is because
the area, we don’t want it to get spoiled. This is the difference between
selfish and stingy. We have to be selfish. (Kaipo Kimokeo, 50s)

Wailuanui they have everything. They have the taro patch, they have the
fishing ground, they have the ‘öpae, they have the o‘opu, they have the
shell in the patch, you know, they have everything what you want. It’s that,
you have to do it yourself, you see . . . you got to work hard for everything
that you want. (Mary Ka‘auamo, 82) 145

The ability to make a living from taro cultivation, fishing, gathering, and
hunting is the highlight of life in Ke‘anae-Wailuanui.

129
chapter three

Family members of all ages engaged in some level of gathering activity in


the Ke‘anae-Wailuanui district in the 1990s. Küpuna such as eighty-three-
year-old Helen Nakanelua still went out and gathered ‘öpae with her home-
made ‘a‘aniu net in the ‘auwai that ran through her property at Lakini. Waio-
kamilo Stream still has ‘öpae that is accessible to the küpuna. The Ka‘auamo
family is best known for its gathering activities. Awapuhi Ka‘auamo Car-
michael, in her fifties, still went to gather ‘öpae, hïhïwai, and ‘opihi from Kai-
lua and over through Kühiwa. Paul Sinenci, an ‘öpio or young member of the
Ka‘auamo family in his twenties, did extensive gathering with friends of his
age. He had access to a broader area because he worked for the East Maui
Irrigation Company (EMI). Awapuhi Carmichael identified some of the areas
they regularly accessed for gathering of ‘öpae, hïhïwai, and o‘opu:

We have our own names. Kapa‘ula, gather ‘öpae. We use Puaaka‘a, we call it
Kaunoa. Above the road, the ditch above the road, we use that stream, and
then it branches off. Even Makapipi, we use Makapipi stream. We use all
the way to the tunnel. We use it. Kühiwa gulch is used by our family.
Kühiwa gulch we use also. Makapipi is just mauka. Kühiwa is mauka.146

Gathering from a variety of places is important in order to conserve


resources. The choice of place to gather is determined by the weather and
other natural signs. Awapuhi Carmichael described the factors that affected
her decision as to where to gather on a particular expedition:

It depends on what we’re getting, and how we feel, the ocean. We never go
to the same place. You know how the Hawaiians used to do, they don’t go
back to the same place, so can restore. It depends on the weather, and then
we go by the moon, the stars. If use one place, then go to another place,
depends on the moon and the stars. We go up far. Especially for hunting
too, we go all the way up. We all go to the same places, although each of us
have our favorite hole, places, where we go for ‘öpae, you know. All mauka
for ‘öpae. And then below have the ‘o‘opu and the prawns, they introduced
the prawns, and hïhïwai. Above the road is more the ‘öpae. Above the road
is where all the ‘öpae are. Above the main highway. And then below the
road has hïhïwai, ‘o‘opu, you know.147

Within the traditional cultural landscape area for Ke‘anae-Wailuanui,


unoccupied areas with pristine flowing streams and forested areas are integral
to the livelihoods of the families in the district. For example, nobody lives in
the area from Wailuaiki to Kopili‘ula and over to Hanawï, and there are many

130
häna: from ko ‘ olau to k aupö

gulches and streams flourishing with hïhïwai and ‘o‘opu that are routinely
harvested by the Ke‘anae-Wailuaiki kua‘äina.

‘Òpae, O‘opu, and Hîhîwai


Ke‘anae-Wailuanui is one of the few remaining areas in the Hawaiian Islands
where ‘öpae could be gathered. Virtually every stream had ‘öpae at some time
during the year. However, it was easier to gather in large amounts in the tun-
nels of the EMI ditch system. The irrigation ditch itself was an excellent
breeding area for the ‘öpae because it had flowing water year round. Some
streams below the ditch, however, did not have enough flowing water to sus-
tain the ‘öpae year round when water was diverted into the ditch system.
Commercial sale of ‘öpae was prohibited under a state law that went into
effect in 1993. Prior to that, certain families gathered ‘öpae for sale in Hono-
lulu. The küpuna explained that when they were growing up, the gathering of
‘öpae for sale was an important source of income for their parents and grand-
parents who did not hold full-time jobs. Helen Nakanelua, for example, would
accompany her grandmother to gather and dry ‘öpelu for sale to the Asahi
market in Honolulu:

And I used to go along with my grandma, with a five gallon can, you know
those tall ones, and I pack some wood, and I pack salt, so that whenever
my grandma goes with the ‘upena net, do you have an idea what the ‘upena
net looks like and they have a little bag there? Some of the bags are small,
but she used to have these long bags. And then she cleans that where I am,
she takes that out, we clean it and we cook it in this can. Salt it and cook
it there, with the wood that I take we cook it. And after it’s cooked, I begin
spreading it on an oil tablecloth and a mat I used to pack along and then
she leaves me there. I attend that ‘öpae while it’s drying. By the time she
comes back here, it’s partly dried, I gather that ‘öpae again, and separate
it in another bag, because that's partly dried, and we continue on, she gets
another bag to do the same thing, cook, so that by the time she ends up
her day, most of the ‘öpae, except the last one she has is partly half dried
already. Do you know how the ‘upena look like? I show you, cause I have
made some for me, because I use it.148

Although ‘öpae cannot be gathered for commercial sale, it is still a popu-


lar delicacy among the families in the district. They also gather ‘öpae to share
with family and friends outside and on different islands. ‘Öpae, the ‘a‘aniu net

131
chapter three

used to gather it, and the methods of preparing it continue to be a distinctive


aspect of the cultural lifestyle for which Ke‘anae-Wailuanui is known and dis-
tinguished.
O‘opu and hïhïwai, which require pristine and flowing stream waters, are
becoming increasingly scarce in the Hawaiian Islands. Certain species of o‘opu
are endangered, and others are rare. Ke‘anae-Wailuanui is one of the few areas
where they still thrive in sufficient size and abundance to be occasionally
caught for subsistence food.
The gathering of hïhïwai is also carefully managed. The location of the
hïhïwai is knowledge that has been passed down from one generation to the
next for protection and proper management. It is not made available to the
general public.

Plants and Trees


Traditionally, the mountain area above Ke‘anae, Wailuanui, and Kopili‘ula is
forested with native trees and plants such as koa and ‘öhi‘a lehua. There is also
sandalwood at Nähiku. The plateau looking from Ke‘anae and over toward
Honomanü has grasses, pia, and ‘ölena. Beginning in 1934, the federal Civil-
ian Conservation Corps planted the non-native trees that became dominant
in the lowland forests.
Most Native Hawaiian plants have some medicinal property. Native plants
throughout the district are still used for medicinal purposes, such as the maile
hohono, ‘ie‘ie, laua‘e, koali, noni, ‘uhaloa, pöpolo, and even watercress.
James Hueu, the eighty-year-old caretaker of the Ke‘anae YMCA camp,
recounted how he had been recruited by two küpuna, his father and brother,
when he was just eleven years old to climb the small islet called Keöpuka to
collect an Hawaiian herb called mokou:

But way back in 1931, two old men from Honolulu, they came up and they
wanted this medicine. One of the old patients he used to live here, his
name was Kalilimoku. So he knew, when he left here he was a young boy,
he knew where the medicine was. My dad was living yet, my brother was
supposed to climb that moku. You see that moku down there [points toward
Häna side, in the ocean]? We call that Keöpuka, but it’s Puahakumoa Bay,
but we call it Keöpuka. When we went down there, I was young, I was only
about 11 years old. So my brother was to climb, so I told my brother I
think I better climb because you have a family and I’m a single boy. So I

132
häna: from ko ‘ olau to k aupö

did climb and I brought that herb. And til today I ask all these lapa‘au
people. They don’t know the name. The medicine they told me was
mokou. So I told Larry about it and then he made a song about it. It was
about me going up on that moku. About that mokou and that island. The
herb is something like a dahlia, it grows like a dahlia plant. It grows some-
thing like a taro or a lily, and then certain times of the year, it dies. Then,
certain times of the year, it grows. So when I went it just was growing,
that’s why I knew, but I didn’t know what kind of herb that was, but my
grandfather said that’s the one. Anyway I tried, I taste, oh boy, it burn in
my chest. So what for, I don’t know. They came to collect that for a doctor
by the name of Kaonohi. He was the first Hawaiian herbalist, from down
Käne‘ohe or somewhere. That’s what they came for. It grows only on that
kind of moku. This thing is like a dahlia. A dahlia grows something like a
lily. The color is green. I never seen the flower. But the leaves are similar
to taro. And in the bottom has a potato. I got the whole thing. The potato,
but I don’t know what part he used. That grows just like the ‘ölena, certain
time of the year it dies.149

A Native Hawaiian black banana, as well as ‘ie‘ie and the Hawaiian bamboo,
was at one time used for weaving mats and hats of natural fiber by weavers in
Ke‘anae-Wailuanui. Pohole, a native fern, was popular for eating as a salad.
One type of pohole is native to Ke‘anae-Wailuanui. Another type of pohole
was introduced in the district from the Big Island by Mrs. Kanoa, but the
older people preferred the taste of the native variety. They only gathered it in
the morning.

Fishing
The entire shoreline, reef, and nearshore area have abundant marine
resources. Those who have boats dive, fish, and gather marine life from Kai-
lua to Häna, going out as far as the buoy, seven miles offshore. They gather
‘opihi, crab, and other shellfish from the rocks; dive for squid, lobster, and reef
fish; bottom-fish for ono and uku; and troll for aku, ‘ahi, and mahimahi. They
occasionally surround akule in the bays such as at Honomanü. It is possible to
use the hukilau method of surrounding and catching fish in any of the bays,
but the practice has been discontinued because it depletes marine resources.
Launching points for fishing are Ke‘anae Point, Honomanü Bay, Wailuanui
Bay, and Häna Harbor.

133
chapter three

The bays, where fresh water mixes with ocean water, are important spawn-
ing grounds for fish. Moki Day, in his fifties, described how the bays are
important breeding grounds that deserve protection:

You can consider all the shoreline area between here and Kaupö as breed-
ing grounds for all these shoreline species of fish. They come into our
rivers here because we have the fresh water, and they come in here and
breed here and lay their eggs here. You go around here in certain seasons
or certain times of the year and you see them, they come out in schools. I
remember when we were small, before we even started grade school we
used to go out with my grandfather and set net, and do all this fishing, you
know, pole fishing, we used to hook moi. We seen fifty-pound moi already.
They look like sharks in the water. But the only difference is that you see
them as a school, a great big school, and there are these big huge monsters
in there, and they’re all in the waves, and I seen them. My brother and I
have been fortunate to see them. And that’s what I’d like to see again. And
the only way to do that is to do this. And so I’d like to see the fishponds
open up so we can restock the moi, and clamp down on fishing, block it up
for a couple of years. They’re hooking babies.150

The men and young boys could fish in any number of areas, all along the
shoreline. For this reason, fishing down at Ke‘anae, Honomanü, and Wailua-
nui was reserved for the küpuna. It was easy for them to access these two areas.
Unfortunately, outsiders who are unfamiliar with the variety of fishing and
gathering spots also went to these easy areas, and made it more difficult for
the küpuna to get a good catch. Moki Day explained how the easily accessible
areas were reserved for the küpuna:

It seems like everyone here has their spot, where they go and harvest.
Like Wailua Bay to Wailuaiki, this is an area that we keep for our kupunas
because it’s easy to go to. If you are an ‘öpio or mäkua, well you have to
hoof it. You haven’t reached that age or to be honored in such a manner.
That’s why we have problems with the outside people because they come
here and this is where they go. Because they see the opihi. Little do they
know that it’s for our küpuna.151

There is interest among the members of the community in restoring the


fishpond down near Pauwalu Point. On the map it is identified as Pu‘uolu
fishpond; however, James Hueu said that the fishpond is called Po‘ulu fish-

134
häna: from ko ‘ olau to k aupö

pond. It has brackish water and is swampy, and the wild ducks used to go there
every year. The fishermen believe that it would be worth restoring.

Hunting, Swimming, and Trails


The area above the highway is accessed for the hunting of wild pigs. On the
Wailuku side, hunters usually go as far as Honomanü. However, the access
road is through Nua‘ailua and up to Pi‘ina‘au. From above Ke‘anae, the access
road starts above Ching’s Pond and goes up to the forest reserve and to the
ditch road. It is easy to go along the ditch road and look for pig tracks and
then go deeper up into the forest once the tracks are sighted. Hunters go to
Wahinepe‘e and up as far as Waiakamoi. Kupa‘u is a popular hunting spot.
However, above Wailuanui, the gulches get deeper and hunting grows more
difficult. The hunters go as far as Makapipi, then accessed along the four-
wheel-drive road at Makapipi and go up mauka as far as Kühiwa. There used
to be a cabin above Kühiwa for hunters to use for overnight stays. A hunting
license is required. Pigs are tracked using hunting dogs, which pin the pig
down so it can be killed by the hunter using a knife.
Freshwater ponds are popular swimming areas. Probably the most popular
is Ching’s Pond, originally known as Pähoa Pond. As the stream continues
down to the ocean, there are pools and additional swimming places. Through-
out the district there are waterfalls that cascade into great swimming ponds.
Along the shoreline, Wailuanui Bay and the Ke‘anae landing are popular
swimming areas. Nua‘ailua is a beach for both swimming and surfing.
The Pi‘ilani Highway Trail is the primary shoreline access route for gath-
ering and fishing all around the island of Maui. In the Ke‘anae-Wailuanui dis-
trict it is still used, particularly from Wailuanui over to Wailua Iki and to
Kalia‘e and over to Kopili‘ula.
There are several makai-to-mauka access trails and four-wheel-drive roads
between the highway and the shoreline. Access to these trails and roads is vig-
ilantly maintained by the community. Once, the use of a trail at Kalia‘e run-
ning makai from the highway was blocked by a new landowner from outside
the community. He put up a fence, bulldozed his land, and tried to cut off
access. The community insisted on maintaining their access and retained the
services of the Native Hawaiian Legal Corporation to assert their Native
Hawaiian access rights.
Awapuhi Carmichael described some of the trails used by her family:

135
chapter three

We go all over here. From the road, we go down, on the ridge, not the
gulch, (makai), right down to Waiohue, right down in the bay. And then
there’s the island there, and from here you go to Kopili‘ula too. All over
here is the King’s Highway, you can walk all along the shoreline. We have
gathering rights all over here, along the stream. You can go either way. But
for our family, we walk ’em. We walk, we walk from the main highway
down and along the coast. We used to walk with my mom up until 1988
we walked this whole area, from the main highway down to the beach and
to the rivers. And we used to take like a caravan. My brothers were in high
school and they would carry the provisions and the ‘opihi. And on the way
back to the highway we would stop for ‘öpae. And my dad, he used to cut
all the trails. Like he used to say, oh I think we should go to this part, like
to Kopili‘ula. So then a day before he would go down and cut the trail.
And then, you know, he had a four wheel jeep and we would go just up
to Kalia‘e and then walk from there to Kopili‘ula.152

Four-wheel-drive roads were the primary access routes from the highway
to mauka hunting and gathering areas. One of the main four-wheel-drive
roads from the highway to the ditch trail began at Pahoa by Ching’s Pond.
Until the original bridge fell down in 1958–59, this access road used to start
by the arboretum. Helen Nakanelua described how she and her grandmother
accessed the streams mauka of the highway using this route:

You know where the arboretum is? There used to be a bridge there. But
because the bridge was broken, the county did not want to build another
bridge. So they brought the bridge to where the swimming pool is. So
that’s the road you go up to Pi‘ina‘au they call it. So looking from the
school, you can see that trail that goes up on that mountain there. That’s
were we used to go on horseback until we get to Kopili‘ula, Wailuaiki,
Wailuanui. Way up Pi‘ina‘au on the mountain, that’s where they had a
camp for all the irrigation company to live, and across there, you cross that
kahawai, and there is that trail there to go. If you go by the school and you
look, you can see it. Pi‘ina‘au is toward the mountain, you take the road to
this way, you go to Wailuaiki, Wailuanui and all that, that the trail you
take. To Pa‘akeke from Pi‘ina‘au, but I went with my grandma, Wailuaiki,
Kopili‘ula, that’s where the end of the trail, when people go to Häna, we
go there and we stay there in a cave to get a car to go to Häna. Now we
would climb up that river, to get ‘öpae, every week we were doing that my

136
häna: from ko ‘ olau to k aupö

grandma and I because my brothers and sisters, they didn’t want to go to


that kind of place.153

The East Maui Irrigation (EMI) ditch road was the popular route for access
in the mauka area for hunting and gathering. Before the main Häna road was
constructed, the ditch road was used to travel from Ke‘anae-Wailuanui to
Häna or to Wailuku. According to the ninety-four-year-old küpuna Enos
Akina, “If you go Häna, telephone to the one who get car to go Kopili‘ula,
wait for you. Go with horse along ditch trail to Kopili‘ula, leave horse, go on
car. You like go outside, you go on the horse until Kailua. From Kailua on, you
get car to go Wailuku. Come back, get the horse, come back. Bumbye they
open this road, get car. Like go outside, go outside.” 154
The EMI ditch road is referred to as the old road since it was regularly used
for travel between Kopili‘ula and Kailua by the residents’ parents and grand-
parents. Residents are therefore accustomed to using it for their hunting and
gathering activities.

Water
Ke‘anae-Wailuanui is part of the east Maui watershed, where forty-eight
streams originate. Of these, thirty-five are perennial (they flow to the sea year
round) and thirteen are intermittent (they flow year round at higher eleva-
tions and intermittently at lower elevations). East Maui Irrigation (EMI) col-
lects, stores, and transports water from over 50,000 acres of the east Maui
watershed, including from the Ke‘anae-Wailuanui cultural landscape. The sys-
tem, owned and built by EMI, consists of approximately 400 intakes (stream
diversions), seven reservoirs, seventy-five miles of aqueduct (fifty miles of
tunnel and twenty-five miles of open ditches), sixty miles of four-wheel-drive
roads, and many miles of trails. The system provides irrigation water for agri-
cultural plains in central Maui, county water to upcountry Maui residents
and farmers, and domestic water for Kula, Pukalani, Makawao, Hali‘imaile,
Ha‘ikü, and Peahi.
Fresh water is an integral part of the cultural landscape for taro cultivation,
the gathering of aquatic and marine resources, recreation, and domestic use.
Awapuhi Carmichael explained the importance of the streams: “All of our
streams need to be protected for recreational and gathering purposes, because
if we don’t, then they’ll divert all of the water, and then the streams will be

137
chapter three

empty. Fishes need water too. Certain fishes need water to spawn. That’s why
we’re lucky, we have enough water for the fishes.” 155
Water flows throughout the Ke‘anae-Wailuanui landscape from both
streams and springs. The taro patches of Ke‘anae and Wailuanui are prima-
rily fed by spring waters. Some of both the stream and spring waters have
been diverted into the EMI system. Awapuhi Carmichael, the genealogist for
the Ka‘auamo family, discovered a letter written in 1881 to two commission-
ers, A. P. Carter and J. S. Walker, by her ancestors opposing the turning over
of the water rights in Honomanü to Claus Spreckels for construction of the
irrigation ditch now controlled by EMI. The letter reads in part:

Keanae, Koolau
September 12, 1881
Hon. A. P. Carter a me Hon. J. S. Walker
Na Komisina (Commissioners)

‘O makou o na komimike, ke nonoi aku nei i ko olua ‘a‘ole lilo ke kahi pono
wai o na aina lei ali‘i o Honomanu, Keanae, Wailua i na ona miliona.

J. W. Kehuhu (Kahuhukaunihiakami Halemano), K. Makaena J. K. Hueu


( John Kalawaianui Hueu), S. Kamakahiki, K. E. Maiailua, D. W. Napihaa,
M. Kaleba, B. B. Kalilimoku, Kamanele, J. S. Lono, J. Kuluhiawa, Keli‘i
(Keli‘iaukai), J. B. Kaakuamoku

[We are the committee members, we request of the two of you not to turn
over the water rights of the crown lands of Honomanu, Keanae, Wailua to
the millionaire (Claus Spreckels).]

[Signed] J. W. Kehuhu (Kahuhukaunihiakami Halemano), K. Makaena


J. K. Hueu ( John Kalawaianui Hueu), S. Kamakahiki, K. E. Maiailua,
D. W. Napihaa, M. Kaleba, B. B. Kalilimoku, Kamanele, J. S. Lono,
J. Kuluhiawa, Keli‘i (Keli‘iaukai), J. B. Kaakuamoku]

There is continuing concern about the impact of water diversion upon


marine life and taro cultivation. Without spring water, Ke‘anae and Wailua-
nui would have be cut off from free-flowing water when the streams were
diverted into the EMI system. The water for Wailuanui originates way up at
Plunket Spring and flows down to Wailuanui through Lakini. At Lakini, an
‘auwai carries the spring water together with water drawn in from the Waio-
kamilo Stream down into Wailuanui. In Wailuanui, the water is distributed
through three separate channels to different sections of taro land.

138
häna: from ko ‘ olau to k aupö

The water for the taro patches in Ke‘anae comes from the Palahulu stream
and the Waikuna spring above Ching’s Pond. Another ‘auwai carries water
into Ke‘anae from a creek named Waihaoawa. James Hueu told the story of
how J. A. Chamberlain dug the ‘auwai for his taro patch in Ke‘anae:

The guy dug an ‘auwai from there, irrigated his taro patch down Ke‘anae
and has his name under, J. A. Chamberlain. The name of that creek is
Waihaowa. Waihaowa iki is the one above, so there must be a Waihaowa
nui. Waihaowa means the separating water. So it was separated up there
and then they met again up there. So this man, he dug a ditch to use in his
taro patch and he named the ditch, the ditch name is ‘auwai ho‘omanawa-
nui, and he put his name under, J. A. Chamberlain.156

The ‘öhi‘a spring is in one of the ‘ili between Ke‘anae and Wailuanui. James
Hueu provided a description of the spring waters that feed this ‘ili:

Where the watercress is, that’s the ‘öhi‘a spring. That’s Waikäne and Kana-
loa. Well they call it ‘öhi‘a spring. Waikäne and Kanaloa is right under the
road. There’s two holes where the water comes out, right when you make
that turn. Right under there is where.
That stream, the Palauhulu stream, on the east side, until the other way,
they call that Pähoa-Waianu. Waianu is at that spring, Waikäne-Kanaloa,
and Pähoa is that river.157

At the time the interviews were conducted, hau bushes were overrunning
the streams and clogging up the water (during heavy rains, debris would get
caught in the bushes and block the free flow of water), a problem that affected
the EMI system, state lands, and the taro farmers. There was general interest
in having the streams cleared through a cooperative effort between the state,
EMI, and the taro farmers.

Taro
The Ke‘anae-Wailuanui district is famous for its taro, and the community is
proud of this legacy. The Ke‘anae-Wailuanui kua‘äina proudly shared tradi-
tional family accounts of how their district was never touched by the wars
between the Maui and Hawai‘i chiefs during the time of Kamehameha I.
According to them, Ke‘anae-Wailuanui was very important to the chiefs as a
source of taro, poi, and water. To ravage the landscape would have been reck-
less and wasteful. Moki Day provided the following insights:

139
chapter three

Thousands of canoes came in for water and for poi. But Ke‘anae-Wailuanui
were never touched by war. The people took food to Häna and to Nähiku
by the Pi‘ilanihale heiau where the war was. They carried the food to
Nähiku and came back to mahi‘ai (farm).
This is the bread basket of Maui, this area here all the way to Kïpahulu.
There are more taro patches here than anywhere else on our island. So
there’s any valley that you can go in here and see taro patches, lo‘i forma-
tions. These things are still here.158

As noted above, some of the historic taro patches in the district were con-
verted to rice paddies between the 1880s and the 1920s. During that period,
taro was primarily grown for home consumption rather than for commercial
sale. James Hueu shared his insights on the origins of some of the taro patches
in the district:

They say, the Chinese make the taro patch. I say not. The taro patches
were there before the Chinese ever came to Hawai‘i. But because the Chi-
nese came, they planted rice in the taro patches and they marry into the
Hawaiian women. But they never did the taro patches. The taro patches
were there long before that. I have two taro patches down here [at the
YMCA camp]. It wasn’t made by the old Hawaiians, I made ’em. I made
that on the dry land. But over here, when we started the CCC camp, I saw
terraces over here. So I think they had taro patches, long time ago. There’s
one taro patch over here on that land, Chamberlain’s land, now it’s Kep-
pler’s, it’s a small taro patch. If you don’t go around that taro patch, you
didn’t see Ke‘anae. The name of that taro patch is Ke‘anae. And they have
some taro patches named Kalihi, they have Maka‘iwa. The taro patches had
a name. Even me I don’t know some of the names. But for Ke‘anae, I knew
that from way back. In the back of the Keppler house, that area is called
Ke‘anae. My son takes care of the patches. I told him, you take care of that
patch, even if you not even get half a bag of taro, but that patch has history,
so don’t let it get filled up with grass.159

At the time that the 999-year homesteads were established at the begin-
ning of the twentieth century, each house lot also had a taro patch assigned
to it. According to Moki Day, “Everyone who chose a homestead got two acre
lots, a house lot and taro land.”160 Each house lot continued to have a taro
patch. Taro began to be produced for commercial sale to Häna and to Wai-
luku beginning in the 1920s. However, the families of many of the kua‘äina

140
häna: from ko ‘ olau to k aupö

who are now producing taro for commercial sale mainly began to sell taro for
poi after World War II. Before that, most of the taro in the district was grown
for home consumption. Many of the families continued to sell taro to supple-
ment their salaries from employment with the Maui county road crew, EMI,
or jobs outside in Kahului or Wailuku. However, there were a few families
whose sole source of income was from taro production.
Many of the current owners of the house lots still cultivate their own taro
patches; however, some of the patches have been passed on to family members
living in Ke‘anae or outside. In some cases, the taro patch was being leased to
other residents in the community. Many of those cultivating taro sold their
taro to the Moloka‘i poi mill.
Taro was also raised in the Ke‘anae arboretum by the State Department of
Land and Natural Resources. At one point the taro collection included sev-
enty varieties, but disease reduced the collection to ten. The taro was grown
in restored taro terraces and patches.
Throughout the district, old taro terraces could be found with wild taro in
the valleys along streams. Some families went out to gather lü‘au leaves from
the wild taro because they had good flavor, distinct from that of the cultivated
varieties. Some of the areas where wild lü‘au was gathered included Pi‘ina‘au,
Nua‘ailua, Kupa‘u, Waipi‘o, Awiowio, Pohole, and Pähoa.
The growing demand for poi made the production of taro a profitable ven-
ture, and there was a lot of interest in taro production. Taro was farmed by
küpuna, mäkua, and ‘öpio. As the küpuna grew old, their children and grand-
children carried on the work. As the mäkua generation reached retirement
age, they returned to their family lands to maintain and open up ancestral taro
patches. In addition, being able to cultivate the taro lands of their ancestors
provided cultural and personal satisfaction to the kua‘äina of Ke‘anae-Wailua-
nui. Kaipo Kimokeo described the spiritual connection he felt to his ancestors
when he worked with the taro:

Anyway, I got this job at the arboretum, we started to open the taro patch,
spiritually, I can see my family around me, because, that was the main food
. . . So now, spiritually, I’m involved . . . spiritually motivated this way [to
grow taro], because what I can see tells me. All these valleys. We supplied
an army. We had all our taro patches. We supplied Kamehameha when he
came here. Aunty Mary Ka‘auamo told me, I was asking her about the
‘apowai taro. She told me that water is spiritual water, it doesn’t touch the
ground. That water, it can be used for medicine and things like that.161

141
chapter three

Persistence of Lifestyle
Ke‘anae-Wailuanui has persisted into the twenty-first century as a center of
traditional Hawaiian culture and lifestyle. Although they live in distinct house-
holds and cultivate individual patches of taro for their own families and for
commercial sale, the residents also share in the abundance of natural resources
from Honomanü to Makapipi. Together, they respect and care for the springs
and streams that feed their individual lo‘i kalo and sustain an abundance of
resources. Together, they continue to balance the conservation of natural
resources with their own needs, taking only what is necessary for their fam-
ily’s daily sustenance and respecting the reproductive cycles of the animal and
plant life in the valleys along the Ko‘olau coast of Maui. Their children play
together, attend school together, and mature to adulthood, sometimes inter-
marrying.
The Native Hawaiian way of life continues to thrive among the households
and taro patches of Ke‘anae-Wailuanui, despite the influx of retired Chinese
plantation workers who farmed at the turn of the twentieth century and the
post-statehood settlement of outsiders seeking a quiet lifestyle in rural Maui.
The persistence of the Native Hawaiian lifestyle in Ke‘anae-Wailuanui pro-
vides an important source of continuity and connection for all Native Hawai-
ians, not just those living there, to their rich heritage. We will more fully
appreciate the role of Ke‘anae-Wailuanui and the Häna district as a cultural
kïpuka in chapter 6, where I discuss the role of the kua‘äina of Häna in the
revitalization, beginning in 1976, of the cultural and natural resources of
Kaho‘olawe.

142
 four 

Puna: A Wahi Pana Sacred to Pelehonuamea

Puna is the land section that inspires hula creation because of the natural
movements of wave, wind and trees. Puna is the source of regenerative
power. Some examples are the rising of the sun, volcanic creation of new
land and the growth of new vegetation on this new formed land.
—pualani kanaka‘ole kanahele, Ka Honua Ola: The Living
Earth, 1992

he interplay of many dynamic primal natural elements in Puna make

T it one of the most sacred areas in all of Hawai‘i. The regenerative power
inherent in the lands and atmosphere of Puna are also reflected in the
role and contributions of the kua‘äina of Puna to the perpetuation of Native
Hawaiian culture through the twenty-first century. “Puna, mai ‘Oki‘okiaho a
Mäwae” (Puna from ‘Oki‘okiaho to Mäwae): as this ‘ölelo no‘eau says, the
Puna district spans from Mäwae on the northern boundary with Hilo south
to ‘Oki‘okiaho on the southern boundary with Ka‘ü.1 Comprising 311,754
acres, the island of Kaua‘i (354,112 acres) could almost fit within the district.
Puna is located in the easternmost part of the easternmost island of the
Hawaiian chain, so all of Hawai‘i’s days begin there. The ‘ölelo no‘eau that
reminds us of this daily phenomenon goes: “Mai ka hikina a ka lä i Kumukahi
a ka welona a ka lä i Lehua” (From the rising of the sun at Kumkahi to the
fading of the sunlight at Lehua).2
The northeast tradewinds, with their rain-infused cloud formations and
rainfall, first reach Hawai‘i in Puna. A Hawaiian proverb, “Ka makani hali ‘ala
o Puna” (The fragrance-bearing wind of Puna), speaks of how these winds
grow fragrant as they travel over Puna, luxuriant with maile, lehua, and hala.3
The name Puna means wellspring and derives from observations by Native
Hawaiian ancestors of how the forests of Puna attract the clouds to drench
the district with its many rains, such as “ka ua moaniani lehua o Puna” (the
rain that brings the fragrance of the lehua of Puna).4 The rains refresh and

143
chapter four

enrich the Puna water table and sustain the life cycle of all living things in
Puna and the entire island of Hawai‘i.
The waters of Puna are believed to originate with Käne, the Hawaiian God
of freshwater sources. His domain is traditionally in the east, where the sun
rises. Käne is a guardian of the Pelehonuamea fire clan, the family of deities
who migrated from the south to Hawai‘i and are manifest in the Puna dis-
trict’s volcanic activity. Käne protects the subsurface waters, the main source
of the volcanic steam that forms the bloodstream of the volcano deity, Pele-
honuamea.
The steam is believed to be the mana, the life force and energy of Pele-
honuamea. When Pelehonuamea does not actively erupt, the steam is the main
form in which she manifests herself. When there is steam in the forest, Pele-
honuamea is thought to be there. That is her identity, her imagery, and her
manifestation. Throughout the district of Puna, traditional chants tell of warm
pools in caves and under ground, such as Kaukala and Punahakeone. These

Figure 22 Pelehonuamea is the dynamic creative energy that shapes the lives and livelihoods
of the kua‘âina of Puna. K. Maehara, Bishop Museum.

144
puna: a WA H I PA NA sacred to pelehonuamea

are the sacred bathing places of Pelehonuamea. Pelehonuamea practitioners


believe that the waters of the Puna district are sacred to Käne and that the
steam generated by the heat of Pelehonuamea is sacred to her.
Puna is where new land is created and new growth and new life sprout.
The new land is sacred, fresh, clean, and untouched. After vegetation begins
again to grow upon it, it is ready for human use.5 Puna is also the center of
the ongoing creation of new land through volcanic activity. It is where new
vegetation comes to life on the newly formed land, repeating a sequence of
evolution that is millions of years old.
A chant translated by Pualani Kanahele describes the primal elements and
features of Puna that Hawaiians celebrate and honor in legend, chant, and
hula.6

Ke Ha‘a La Puna I Ka Makani

Ke ha‘a la Puna i ka makani Puna is dancing in the breeze


Ha‘a ka ulu hala i Kea‘au The hala groves at Kea‘au dance
Ha‘a Hä‘ena me Höpoe Hä‘ena and Höpoe dance
Ha‘a ka wahine The woman dances
‘Ami i kai o Nänähuki [She] dances at the sea of Nänähuki
Hula le‘a wale Dancing is delightfully pleasing
I kai o Nänähuki At the sea of Nänähuki
‘O Puna kai kuwä i ka hala The voice of Puna resounds
Pae i ka leo o ke kai The voice of the sea is carried.
Ke lü la, i na pua lehua While the lehua blossoms are being scattered.
Nänä I kai o Höpoe Look toward the sea of Höpoe
Ka wahine ‘ami i kai o Nänä The dancing woman is below, toward
huki Nänähuki
Hula le‘a wale Dancing is delightfully pleasing
I kai o Nänähuki At the sea of Nänähuki

Pualani Kanaka‘ole Kanahele’s elaborate and compelling interpretation of


this chant in her manuscript “Ka Honua Ola: The Living Earth” reveals its
true significance. According to Kanahele, “Ke Ha‘a La Puna” is the first
recorded hula in the Pele and Hi‘iaka saga. Hi‘iaka performed a hula to this
mele or song to please her older sister, Pelehonuamea. The hula was performed
at Ha‘ena and represented the birth of the hula sacred to Pele. The chant
refers to weather phenomena, movements of nature, and the natural imagery
for which Puna is famous —the breeze, the hala groves, the sea, and the vol-

145
chapter four

canic eruption. The rain together with the rising sun of Puna are the nurtur-
ing substances that induce the growth of vegetation on the new land.
Proceeding from the context of Puna for the chant, Kanahele interpreted
the meaning of the lines in the chant. In the first line, “Ke ha‘a la Puna i ka
makani” (Puna is dancing in the breeze),

Hi‘iaka, the youngest sister of Pele is asked by Pele to do a ha‘a and a mele.
She satisfies her older sister’s request with “Ke ha‘a la Puna i ka makani.”
The ha‘a or dance which she exhibits is a creative exposition in praise of
the environment around her and a celebration of the regenerative power
of the coupling of land and flora. Hi‘iaka’s own kinolau or body forms are
the flora which readily grows on new lava flows. Therefore it is Hi‘iaka’s
place to celebrate this newly made land upon which her body forms are
given life.

Kanahele goes on to elaborate that the line “Puna kai i ka hala” (The voice
of Puna resounds) refers to the beating of the sea on the cliffs of Puna.

This sound is magnified through the groves of hala. The hala grove
becomes the resonator. The sea movements of Puna as it heaves, rolls,
dashes, splashes, sprays and vibrates, produces various distinct sounds and
chords. The various sounds emanating from the hala grove are symbolic
of the sounds reproduced by the hula implement which excites and provokes
movement for the dancer.7

According to Kanahele, the chant explains the roles of the deities Pele-
honuamea and Hi‘iaka in hula:

Imitation of nature gives praise to those Deities responsible for different


aspects of nature. Pele's energy, her explosive, dramatic creative tactics of
land birth deserve praise. The dualistic nature of Hi‘iaka and her pro-
creative powers of vegetable growth also deserve praise. The land and
vegetable manifestation of these sisters provide initial movement and
energy in creating hula. The hula associated with these deified sisters
are pure original movement and pure sound.8

Having reviewed the major primal forces and deities associated with the
Puna district, as well as the singular importance of Puna as the birthplace of
the Pelehonuamea forms of hula, we will look at the patterns of change in the
landscape reflected in myth and legend.

146
puna: a WA H I PA NA sacred to pelehonuamea

Puna’s Mythical Era


The myths and legends of Puna are dominated by Pelehonuamea, Hawaiian
Goddess of the volcano, and the members of her fire clan, who migrated from
their distant homeland through the northwest islands of Hawai‘i until they
settled in Puna, Hawai‘i. The Hawaiian proverb “Ke one lau‘ena a Käne”
(The rich, fertile land of Käne) was interpreted by Mary Kawena Pukui as a
reference to the idea that Puna was a beautiful and fertile land loved by the
God Käne. According to Pukui, Pelehonuamea changed it into a land of lava
beds, cinder, and rock when she settled there from Kahiki.9
The legends, myths, and chants that describe the early development of the
Puna district relate the dynamic tension between the deities of the Pelehonua-
mea fire clan and the deities honored by other Hawaiian families as their
ancestors. Each of the deities represents a different elemental force in the nat-
ural landscape of the Puna district. Throughout all of the folklore for Puna,
Pelehonuamea and her family of deities emerge as the natural primal elements
that dominate and shape the lives of the chiefs and the people of Puna. Here
are two traditional mo‘olelo that reflect the conflict between Pelehonuamea or
the volcanic fire and deities representing other natural elements in Puna, the
mo‘o or dragon lizards who dwelled in mountain pools and shoreline ponds
before the Pelehonuamea clan came to Hawai‘i, and the Pig God who dwelled
in the old-growth forests. The volcanic deity conquers the mo‘o but reaches
a compromise with the Pig God.

Waka the Mo‘o


Traditional mo‘olelo describe Ka‘ü and Puna as beautiful lands without lava
beds. It is said that there was only earthen soil from one end to the other. The
mo‘olelo reveal the existence of a very long sandy stretch called Keonelaue-
naakäne (“Käne's great sand stretch”) in the district of Puna. The lava cov-
ered the earth and sand and transformed Puna into a land of lava rock.
The mo‘o, Wakakeakaikawai and Puna‘aikoa‘e, were destroyed by Peleho-
nuamea of the eternal fires. According to this legend, the fight between these
mo‘o and Pelehonuamea began in Punalu‘u in Ka‘ü, continued in Puna, and
ended in Waiäkea in Hilo. Through the course of the battle, a long stretch of
sand extending from Waiäkea, Hilo, to Pänau, in Puna, called Keonelau‘ena-
akäne, was covered with lava. Because Waka ran through Puna, with Peleho-

147
chapter four

nuamea in pursuit, most of the land in Puna became covered with rough and
smooth lava and remains so to this day. The famous stretch of sand disap-
peared. Only traces of it can be seen in small pockets, scattered here and there,
from Waiäkea to Puna.10

Pele and Kamapua‘a


Legends also relate the dramatic struggles between Pelehonuamea and the
forces of the storm and forest represented as Kamapua‘a, the hog-man or Pig
God. Kamapua‘a goes to the crater of Halema‘uma‘u and courts the Goddess
in the form of a handsome man. Her sisters attract her attention to him. Not
at all deceived, Pelehonuamea refuses him with insults, calling him “a pig and
the son of a pig.” His love songs change to taunts, and the two engage in a con-
test of invective. He attempts to approach her, but she sends her flames over
him. Each deity summons its own God. Pelehonuamea’s brothers encompass
Kamapua‘a “above and below” and would have smothered him had not the
lovemaking God of Kamapua‘a lured them away with a beautiful woman.
Kamapua‘a threatens to put out the fires of the pit with deluges of water, but
Pelehonuamea’s uncles and brothers and the fire tender Lonomakua keep
them burning. The reigning chiefess of Makahanaloa sends fog and rain to
support her brother against the fire Goddess. Hogs run all over the place.
The pit fills with water. The lovemaking God sees that if Pelehonuamea is
destroyed Kamapua‘a will be the ultimate loser. The fires are all out; only the
fire sticks remain. These the God decides to save. Pelehonuamea yields, and
Kamapua‘a has his way with her. They divide the districts between them, Pele-
honuamea taking Puna, Ka‘ü, and Kona (districts that are periodically over-
run with lava flows) and Kamapua‘a ruling Kohala, Hämäkua, and Hilo (the
windward districts, always moist with rain).11

Settlement and Expansion


Archaeological excavations in Puna set the date for the earliest settlements in
the district between 300 and 600 a.d.12 Very little subsurface excavation has
been conducted in the Puna district; as such studies are conducted, it is pos-
sible that the date for earliest settlement could be revised. Because of the lack
of running streams in Puna, early settlers first lived along the shoreline, where
they had access to the ocean, freshwater springs along the shoreline and in the
ocean, and arable land.

148
puna: a WA H I PA NA sacred to pelehonuamea

Legends possibly set in this era document the trials of Puna chiefs and their
followers with Pelehonuamea and her fiery temper. Chief Kanuha of Kona
shared the legend of the Puna chief Keliikuku with the French explorer Jules
Remy in the nineteenth century.13 The event was believed by Kanuha to have
occurred in the 1600s. The legend of how the young chief Kahawali and his
hula students perish after rebuffing a challenge from Pelehonuamea to com-
pete at hölua or mountain sledding is also related in many sources and prob-
ably occurred around the same time.14

The Legend of Keli‘ikuku


According to the chief Kanuha, up until the 1600s the district of Puna was
renowned as magnificent country, with smooth, even roads and a sandy soil
that was favorable to vegetation. Native Hawaiians at the time of Kanuha had
grandparents who related stories of the great volcanic floods they had person-
ally witnessed in Puna during their lifetime.
A certain high chief reigned in Puna. He journeyed to the island of O‘ahu
where he met a prophet of Kaua‘i named Käneakalau, who asked him who he
was. “I am,” replied the chief, “Keli‘ikuku of Puna.” The prophet then asked
him what sort of country he possessed. The chief said: “My country is charm-
ing. Everything is found there in abundance. Everywhere are sandy plains
which produce marvelously.” “Alas!” replied the prophet, “Go. Return to your
beautiful country. You will find it overthrown, abominable. Pelehonuamea has
made of it a heap of ruins. The trees of the mountains have descended toward
the sea. The ‘öhi‘a and pandanus are on the shore. Your country is no longer
habitable.” The chief made answer: “Prophet of evil, if what you now tell me
is true, you shall live. But if, when I return to my country, I prove the falsity
of your predictions, I will come back on purpose and you shall die by my
hand.”
Unable, in spite of his incredulity, to forget this terrible prophecy, Keli‘i-
kuku set sail for Hawaii. He reached Hämäkua, landed, and traveled home by
short stages. From the heights of Hilo at the village of Makahanaloa, he beheld
in the distance his entire province overwhelmed in chaotic ruin, a prey to fire
and smoke. In despair, the unfortunate chief hung himself on the very spot
where he first discovered this sad spectacle.
This tradition of the mountain of Keli‘ikuku and Käneakalau continued to
be told and retold among Hawaiian storytellers. It was even put to meter and
sung by the ancients. According to the chief Kanuha, whether the prediction

149
chapter four

was made or not, the fact that Puna had been ravaged by volcanic action had
come to pass.

Kahawali and Pelehonuamea


The handsome young chief Kahawali lived near Kapoho in the Puna district
of Hawaii in the days of the chief Kahoukapu. He had a wife and two chil-
dren named Paupoulu and Ka‘ohe. His mother lived at Küki‘i, and he had a
sister, Ko‘ae who lived at Kula. His father and another sister named Käne-
wahinekeaho lived on O‘ahu. Kahawali was an expert in the hula dance and in
riding the hölua. At the time of the Makahiki festival, when the hula pupils
gathered for a public appearance, a sled race was arranged with his friend
Ahua. Pelehonuamea in the guise of an old woman also offered to compete
with Kahawali and he laughed at her impertinence. Angry at the chief ’s rebuff,
Pelehonuamea pursued him down the hill in her fire form. Kahawali fled first
to the hill Pu‘ukea, then hastened to bid good-bye to his wife and children.
He paused to say farewell to his favorite pig, Aloipua‘a, and had just time to
greet his sister at Kula before he escaped to the sea in a canoe his brother had
opportunely brought to land. Upright lava rocks are said to mark the fate of
members of Kahawali’s family and of his favorite pig. The famous tree molds
(Papalauahi) above Kapoho are identified as a group of hula pupils caught in
the track of Pelehonuamea’s wrath.15

Migration, Ruling Chiefs, and ‘Ohana


For the period of the Tahitian migration to Hawai‘i Island between 1100 and
1300, Puna is prominent in legends as the district where the high priest Pa‘ao
made his first landfall and built the Waha‘ula heiau for his God. The Hawai-
ian historian Samuel Kamakau provides a brief account: “Puna on Hawai‘i
island was the land first reached by Pa‘ao, and here in Puna he built his first
heiau for his God Aha‘ula and named it Aha‘ula [Waha‘ula]. It was a luakini.
From Puna, Pa‘ao went on to land in Kohala, at Pu‘uepa. He built a heiau
there, called Mo‘okini, a luakini.” 16
Pa‘ao, according to Hawaiian mo‘olelo, was a powerful priest and prophet.
According to Kamakau, he originated from Wawau and ‘Upolu, lands in the
mythical Polynesian homeland, Kahiki.17 In Hawai‘i he established a new order
of religious priesthood and practices that included human sacrifices at the lua-

150
puna: a WA H I PA NA sacred to pelehonuamea

kini heiau whose form of construction he introduced in Hawai‘i. The priest-


hood of Pa‘ao served the ruling chiefs of Hawai‘i until the time of Hewahewa,
high priest of Kings Kamehameha I and II who collaborated with Kameha-
meha II in the abolition of the traditional chiefly kapu in 1819.
Abraham Fornander gives a description of Waha‘ula heiau:

It was built in the quadrangular or parallelogram form which characterized


all the Heiau built under and after the religious regime introduced by Paao,
and in its enclosure was a sacred grove, said to have contained one or more
specimens of every tree growing on the Hawaiian group, a considerable
number of which, or perhaps their descendants, had survived when last the
author visited the place in 1869.18

According to Kamakau, Hawai‘i Island was without a chief when Pa‘ao


arrived in Hawai‘i. Evidently the chiefs of Hawai‘i were considered ali‘i maka-
‘äinana or just commoners, maka‘äinana, during that time.19 Pa‘ao sent back
to Tahiti for a new ruler for Hawai‘i, thereby ushering in a new era of ruling
chiefs and kahuna in the Hawaiian archipelago. The new ruler was Pilika‘aiea,
ancestor of the future King Kamehameha I. Kamakau, Fornander, and Thrum
place Pa‘ao in the eleventh century, sixteen generations from Heleipawa. Cart-
wright places Pili, the chief brought to Hawai‘i by Pa‘ao, in the twenty-fifth
generation before 1900 — that is, 1275 ce.20
In the legend of the migration of Mo‘ikeha to Hawai‘i, his party first
touched at the easternmost point of Hawai‘i, Cape Kumukahi, and his younger
brothers Kumukahi and Ha‘eha‘e remained in Puna. Of the others in his fam-
ily, the kahuna Mo‘okini and Kaluawilinau made their home at Kohala; Honu-
aula landed in Häna on Maui; and his sisters Makapu‘u and Makaaoa landed
on O‘ahu. The rest of the party went on to Kaua‘i.21
In the Kumuhonua legend about the migration of Hawai‘iloa, also known
as Kekowaihawai‘i, he, his family, and his followers migrate to Hawai‘i. He
alone takes his wife and children. They are credited with being the ancestors
of the Hawaiian people. Hawai‘iloa named the island of Hawai‘i after himself,
the other islands after his children, and various land divisions after the naviga-
tors who sailed with him. From time to time Hawai‘iloa voyages south to bring
back mates for his children from the family of his brother Ki. He brings Ki’s
oldest son, Tunuiaiateatua, as husband for his favorite daughter, O‘ahu. Their
son, Tunuiatea, is born at Keauhou on Hawai‘i. Hawai‘iloa names the district
of Puna for the birthplace of his nephew Tunuiaiateatua, Punaauia, in Tahiti.22

151
chapter four

Ruling Chiefs of Puna


“Hilina‘i Puna, kälele iä Ka‘ü” (Puna leans and reclines on Ka‘ü) refers to the
common origin of the people of Puna and Ka‘ü. The ancestors of these two
districts were originally of one extended family. The time came when the peo-
ple of each district decided to have a name of their own, without breaking the
link entirely. Those in Ka‘ü referred to themselves as the Mäkaha, meaning
fierce, savage, ferocious. Those in Puna called themselves Kümäkaha, or stand-
ing fierce, savage, ferocious. Both names are related in chants of the chiefs of
Puna and Ka‘ü.23 Again referring to the common origins of the Mäkaha of
Ka‘ü and the Kümäkaha of Puna is the rallying cry “E ala e Ka‘ü, Kahiko o
Mäkaha; e ala e Puna, Puna Kümäkaha; e ala e Hilo na‘au kele!” (Arise, O
Ka‘ü of ancient fierce descent; arise, O Puna, stand fierce; arise, O Hilo of the
water-soaked foundation).24 The distinction between the families of Ka‘ü and
Puna may have occurred during this period.
Puna’s political history throughout this period is bound up with the for-
tunes of the ruling families of Hilo or Ka‘ü. No one single Puna family
emerges upon whose support either the Hilo or Ka‘ü chiefs seeking power
could depend upon for success. Thus, the political control of Puna did not rest
upon conquering Puna itself, but rather upon control of the neighboring dis-
tricts of Ka‘ü and Hilo.25
Nevertheless, there were two notable Puna chiefs in this era, Hua‘a and
‘Imaikalani, who were identified as enemies of High Chief ‘Umialïloa and
were killed by he and his warriors. During the time of High Chief Lïloa,
approximately 1475 ce, the chiefs of the six districts of Hawai‘i, including
Puna, were autonomous within their own districts, but they acknowledged
Lïloa as their paramount chief. Hakau, son of the sacred wife of Lïloa, suc-
ceeded him. According to Kamakau, Hakau failed to look after the well-being
of the people under him: “But in the later years of his rule he was lost in pleas-
ure, mistreated the chiefs, beat those who were not guilty of any wrongdoing,
and abused the priests of the heiaus of his God and the chiefs of his own gov-
ernment.” 26
The chiefs and priests conspired with ‘Umialïloa, Hakau’s half brother, and
killed Hakau. Hakau's death left ‘Umi in possession of Hämäkua. The chiefs
of the remaining districts of Hawai‘i declared their independence from ‘Umi.
‘Umi went to war, conquered those chiefs who resisted him, and reunited the
districts of the entire island under his rule. According to Kamakau, Hua‘a, the

152
puna: a WA H I PA NA sacred to pelehonuamea

chief of Puna, was conquered by ‘Umialïloa: “Hua-‘a was the chief of Puna,
but Puna was seized by ‘Umi and his warrior adopted sons, Pi‘i-mai-wa‘a,
‘Oma‘o-kamau, and Ko‘i. These were noted war leaders and counsellors
during ‘Umi’s reign over the kingdom of Hawai‘i. Hua-‘a was killed by Pi‘i-
mai-wa‘a on the battlefield of Kuolo in Kea‘au, and Puna became ‘Umi-a-
Liloa’s.” 27
‘Imaikalani is the first chief of Ka‘ü who is said to have control over parts
of Puna. In the time of ‘Umialïloa, circa 1500 ce, he reconditioned the heiau
of Waha‘ula. This is an indication that he held supreme authority over the
ahupua‘a of Pülama in Puna. He was a chief of power and prestige and can be
found in several chiefly genealogies, including that of Queen Emma. Accord-
ing to Barrere, ‘Imaikalani may well have been one of the chiefly ancestors of
the Mäkaha and Kümäkaha lines of Ka‘ü and Puna.28
Kamakau provided the following account of the conquest of chief ‘Imai-
kalani by the ‘Umi warrior Pi‘imaiwa‘a:

‘Umi-a-Lïloa feared I-mai-ka-lani. Although he was blind and unable to see,


his hearing was keen. He had pet ducks that told him in which direction a
person approached, whether from in front, at the back, or on either side.
All depended on the cries of the birds. In former days I-mai-ka-lani was
not blind, and ‘Umi was never able to take Ka‘ü. The war lasted a long time.
‘Umi went by way of the mountains to stir up a fight with I-mai-ka-lani and
the chiefs of Kona . . . I-mai-ka-lani was never taken captive by ‘Umi, but
Pi‘i-mai-wa‘a was crafty and studied the reason for his great strength and
skill with the spear . . . All these men were destroyed by Pi‘i-mai-wa‘a, and
the blind man was at a loss for the lack of helpers. Well could Pi‘i-mai-wa‘a
say in a boast, “Death to him from Pi‘i-mai-wa‘a.” After I-mai-ka-lani’s
death Ka‘ü became ‘Umi-a-Lïloa’s.29

In the next generation, Kahalemilo, son of ‘Imaikalani and Lililehua, and


the son of Hua‘a were both killed by ‘Umi's son, Keawenuia‘umi, when he
gained control of Hawai‘i. According to Barrere, this seems to have extin-
guished the lines of ‘Imaikalani and Hua‘a as autonomous chiefs of Ka‘ü and
Puna. From the time of Keawenuia‘umi, Ka‘ü was ruled by the Kona chiefs
descended from Keawenuia‘umi. Puna is linked with Ka‘ü until the time of
Keaweikekähiali‘iokamoku, when the ‘I family of Hilo controlled parts of
Puna. While control over the other part of Puna is not specifically mentioned,
it can be inferred that it continued to be linked with Ka‘ü.30

153
chapter four

Puna on the Eve of European Contact


On the eve of European contact, Puna seemed to have enjoyed a brief resur-
gence of semiautonomous rule. In the time of Kalani‘opu‘u, the chief ‘Imaka-
koloa of Puna became powerful enough to attract the wrath of the ruling chief.
‘Imakakoloa was probably a descendant of ‘Imaikalani through the ‘I fam-
ily. Kalaniopu‘u, having gained control of all Hawai‘i, found his latter days
troubled by a suspected rebellion in Puna and Ka‘ü. Kamakau gives the follow-
ing account:

Meanwhile rebellion was brewing. It was I-maka-koloa, a chief of Puna,


who rebelled, I-maka-koloa the choice young ‘awa [favorite son] of Puna.
He seized the valuable products of his district which consisted of hogs,
gray tapa cloth (‘eleuli), tapas made of mamaki bark, fine mats made of
young pandanus blossoms (‘ahu hinano), mats made of young pandanus
leaves (‘ahuao), and feathers of the ‘o‘o and mamo birds of Puna.
Nu‘u-anu-pa‘ahu, chief of Ka-‘u, was also in the plot to rebel, but he
was at this time with Ka-lani-‘opu‘u, and Ka-lani-‘opu‘u feared Nu‘u-anu-
‘opu‘u.31

Kalaniopu‘u first disposed of Nu‘uanupa‘ahu by conspiring with his kahuna


to have sharks devour him. Although Nu‘uanupa‘ahu successfully killed the
attacking sharks, he died from the mortal wounds that he sustained in the
struggle with them.32 After disposing of Nu‘uanupa‘ahu, Kalani‘opu‘u hunted
down ‘Imakakoloa. Kamakau, again, offers an excellent account of this event:

Ka-lani-‘opu‘u the chief set out for Hilo with his chiefs, warriors, and
fighting men, some by land and some by canoe, to subdue the rebellion
of I-maka-koloa, the rebel chief of Puna . . . The fight lasted a long time,
but I-maka-koloa fled and for almost a year lay hidden by the people of
Puna . . . Puhili went until he came to the boundary where Puna adjoins
Ka-‘u, to ‘Oki‘okiaho in ‘Apua, and began to fire the villages. Great was
the sorrow of the villagers over the loss of their property and their canoes
by fires. When one district (ahupua‘a) had been burnt out from upland to
sea he moved on to the next . . . Thus it was that he found I-maka-koloa
where he was being hidden by a woman kahu on a little islet of the sea . . .
I-maka-koloa was taken to Ka-lani-‘opu‘u in Ka-‘u to be placed on the altar
as an offering to the god, and Kiwala‘o was the one for whom the house of
the god had been made ready that he might perform the offering . . . Before

154
puna: a WA H I PA NA sacred to pelehonuamea

he had ended offering the first sacrifices, Kamehameha grasped the body of
I-maka-koloa and offered it up to the god, and the freeing of the tabu for
the heiau was completed.33

The stage was therefore set for the usurpation of Kïwala‘ö as heir to his
father, High Chief Kalani‘opu‘u, by Kamehameha, in the period after Euro-
pean contact.

European Contact in Puna


The surgeon David Samwell and Lieutenant James King, British officers on
the Cook voyage, provided the first written accounts of Puna. According to
King: “On the southwest extremity of Opoona the hills rise abruptly from the
sea side, leaving but a narrow border, and although the sides of the hills have
a fine verdure, yet they do not seem cultivated.” 34 Samwell observed: “Many
people collected on the Beach to look at the Ship . . . many canoes came off
to us . . . [with] a great number of beautiful young women.”
Soon after Kalani‘opu‘u died in 1782, Kïwala‘ö was killed by the forces of
Kamehameha in the battle of Moku‘ohai. For the next ten years, Kamehameha
fought the chiefs of Hawai‘i for control of the island. The districts of Kona
and Kohala and portions of Hämäkua acknowledged Kamehameha as their
ruler. Hilo, the remaining portion of Hämäkua, and a part of Puna acknowl-
edged Keawema‘uhili as their ruling chief. The lower part of Puna and the
district of Ka‘ü supported their chief, Keouaküahu‘ula. The battles among
these three chiefs culminated in the triumph of Kamehameha.35 “He moku
‘äleuleu” (district of ragamuffins) was a description the followers of Kameha-
meha I had for the people of Ka‘ü and Puna. According to Hawaiian scholar
Pukui, this was because the people of these two districts were hard-working
farmers who dressed most of the time in old clothes.36 The saying indicates
that the people of Puna were not among those who prospered under the reign
of King Kamehameha.

Puna in the Nineteenth Century


Table 4 shows the population of the Puna district throughout the nineteenth
century. Descriptions of the Puna district in the nineteenth century paint an
image of the living conditions of the Native Hawaiians of Puna.
The first missionary to journey through Puna was William Ellis in 1823. In

155
chapter four

his published journal he described the natural resources available to the resi-
dents of the district and some of their living conditions and subsistence and
exchange practices. He estimated that there were approximately 725 inhabi-
tants at Kaimu and another 2,000 Hawaiians in the immediate vicinity along
the coast. At Kauaea, about three and a half miles from Kaimu, he reported,
300 people gathered to hear him preach.37 The journal entries excerpted below
describe the diversity of conditions he observed traveling through Puna, from
Kïlauea through Kealakomo, toward Kalapana, over to Kapoho, and finally to
Kea‘au.
In the area between Kealakomo and Kamoamoa more people lived along
the coast, close to where they could fish for subsistence, than inland. The
resources of the land alone were not sufficient to allow the ‘ohana to subsist:

We saw several fowls and a few hogs here, but a tolerable number of dogs,
and quantities of dried salt fish, principally albacores, and bonitos. This
latter article, with their poe [poi] and sweet potatoes, constitutes nearly the
entire support of the inhabitants, not only in this vicinity, but on the sea-
coasts of the north and south parts of the island.

table 4 census of the puna district


population
year estimate source

1823 142,050 Ellis, 1823, in Journal of William Ellis


1832 12,755 Jarves, History of the Hawn Islands (1872), p. 202 (North Hilo
& South Hilo included)
1834 4,000 American Board of Commissioners for Foreign Missions
1835 4,807 Ke Kumu, April 13, 1836
1854 2,702 Lyman, letter to Armstrong, Jan. 14, 1854
1860 2,158 Anderson, Hawaiian Islands, p. 278
1866 1,932 Jarves, History of the Hawn Islands (1872), p. 202
1872 1,228 Thrum’s, 1876
1878 1,043 Gen’l Supp. of the Census (G.S.P.), Dec. 27, 1878
1884 944 G.S.P., Dec. 27, 1884
1890 834 Bureau of Public Instruction, G.S.P., Census, 1890
1896 1,748 Department of Public Instruction, G.S.P., 1896
Source: Robert C. Schmitt, The Missionary Census of Hawai‘i, Pacific Anthropological Records 20
(Honolulu: Bernice Pauahi Bishop Museum Press, 1973).

156
puna: a WA H I PA NA sacred to pelehonuamea

Besides what is reserved for their own subsistence, they cure large quan-
tities as an article of commerce, which they exchange for the vegetable
productions of Hiro [Hilo] and Mamakua [Hamakua], or the mamake
and other tapas of Ora [Olaa] and the more fertile districts of Hawaii.

The area past Kamoamoa and toward Kaimu was verdant with gardens and
groves of coconut and of kou trees. There were approximately 725 people liv-
ing at Kaimu. The fine sandy beach afforded a safe landing for fishing canoes.
Leaving Kehena, the village of Kamaili, in a gently sloped valley, was culti-
vated and shaded by large coconut trees. The lava around Puala‘a was pictur-
esque. While some areas had soil, here only grass and trees ornamented the
landscape. Between Puala‘a and Kapoho the lava was barren and rugged until
they reached Kapoho, which Ellis described as charming:

We soon left this cheerful scenery, and entered a rugged tract of lava, over
which we continued our way till about two p.m., when we reached Kapoho.
A cluster, apparently of hills three or four miles round, and as many hun-
dred feet high, with deep indented sides, overhung with trees, and clothed
with herbage, standing in the midst of the barren plain of lava, attracted
our attention . . . On reaching the summit, were agreeably surprised to
behold a charming valley opening before us. It was circular, and open
towards the sea. The outer boundary of this natural amphitheater was
formed by an uneven ridge of rocks, covered with soil and vegetation.
Within these there was a smaller circle of hills, equally verdant, and orna-
mented with trees. The sides of the valley, which gradually sloped from the
foot of the hills, were almost entirely laid out in plantations, and enlivened
by the cottages of their proprietors.

Kea‘au was the last Puna village visited by Ellis. As the one ahupua‘a with
a stream, it was well populated and intensely cultivated, as Ellis noted: “It was
extensive and populous, abounding with well-cultivated plantations of taro,
sweet potatoes, and sugar-cane; and probably owes its fertility to a fine rapid
stream of water, which, descending from the mountains, runs through it into
the sea.” 38
It was not until 1836 that the next missionary, Titus Coan, traveled through
Puna. He preached to villages throughout the district, creating a Christian
revivalist atmosphere wherever he went. Following his visit, some of the Puna
Hawaiians formed Christian congregations. In the 1840 Annual Station
Report for Hilo it was noted that six new “meeting houses” had been built and

157
chapter four

that fifteen congregations were meeting in houses in the districts around


Hilo. When Chester Lyman, another missionary, toured Puna with Coan in
1846, he described visiting a meeting house in Kamoamoa and a “church” in
Kalapana.39
In 1840 a Catholic priest, Father Walsh, was assigned to the island of
Hawai‘i, and in 1841 he baptized Hawaiians in Puna and Ka‘ü. Soon thereafter
a resident priest was assigned to Ka‘ü, and he made periodic visits to Puna.
However, it was not until a Belgian priest, Father Damien de Veuster, was
assigned to Puna in 1864 that more Hawaiians were baptized into the Catholic
faith and regular services were held. During the year Father Damien spent
there before being assigned to Moloka‘i, the number of Catholics in Puna
increased from 350 to 450. Damien built several thatched grass churches and
began to build a mortar church with a thatched roof. Originally called St.
Joseph’s, it later came to be known “Father Damien’s Church.”
In 1841 Coan estimated the Hawaiian population of Puna at 4,371. He
wrote that most of the inhabitants of Puna lived along the shore, although
there were hundreds also scattered inland.40 The same year Captain Charles
Wilkes of the U.S. Exploring Expedition explored the Kïlauea volcano and the
East Rift Zone in Puna. He observed agricultural activities in the Puna Forest
Reserve in the vicinity of Kahauale‘a: “We left Pänau after half-past eight
o’clock, and passed on towards the east. After traveling about three miles, we
came in sight of the ocean, five mile off. Our course now changed to the
northeast, and before noon we reached an extensive upland taro-patch.” 41
In 1846 Lyman traveled through Puna with Coan and reported on agricul-
tural activity in what was probably the interior of the Puna Forest Reserve
near Kahauale‘a:

Our route from Kahauale‘a [village] lay northerly, gradually rising. By half
past 2 p.m. we had reached a plantation in an unsettled region where a good
old man had been at work all day putting up a small neat house of ti leaves,
in expectation that we would stop here for the night. Plantains, pawpaws,
taro, etc. were growing around . . . We went on about 5 miles further, or
10 miles from Kahauale‘a [village] over an exceeding rough and jagged path
and through a dense miry thicket to a small grass shanty.42

Ka Mâhele of Puna
Puna is distinguished as the district on Hawai‘i with the smallest amount of
private land awards under the 1848 Mähele and Kuleana Act. It is remarkable

158
puna: a WA H I PA NA sacred to pelehonuamea

that in a district with 311,754 acres, only nineteen awards of private land were
granted. Of these awards, sixteen grants of 50,876 acres, four ahupua‘a, and
two portions of a third ‘ili were given to ten chiefs who lived outside of Puna.
Three small parcels totaling 32.33 acres were granted to commoners, Bara-
naba, Hewahewa, and Haka. The bulk of the Puna lands were designated as
public lands either to the monarchy, as Crown lands, or to the government of
the Hawaiian kingdom.43 This means that the interests of the majority of the
Native Hawaiians in Puna were never separated out from the lands of Puna
and remained vested in the lands held by the Crown and the government.
Among the chiefs who received lands in Puna was William Charles Luna-
lilo, who later reigned as king from 1873 to 1874. He received 26,000 acres in
Kahauale‘a, 5,562 acres in Keahialaka, and 64.275 acres in Kea‘au. His father,
Charles Kana‘ina, received 4,060 acres in Kapoho. Victoria Kamämalu, the
daughter of High Chiefess Kaho‘anoku Kina‘u and Mataio Kekuanaoa, was
the sister of Kamehameha IV and Kamehameha V. She received 1,568 acres in
Kauaea, 1,822 acres in Kauwalehua, and 2,869 acres in Kahuwai. Keohokalole
Ane, mother of King Kaläkaua and Queen Lili‘uokalani, received 4,919 acres
in Puua and an ahupua‘a in Puna. Miriam Kekauonohi received the ahupua‘a
of Pänau and Waiakahiula. Hakaleleponi (Hazaleleponi) Kalama, the wife of
Kamehameha III, received 2,902 acres in Puna (kula). Kale Davis, daughter of
Isaac Davis, the second British military advisor (with John Young) to Kameha-
meha I, received an apana in Waikahekahe. Gina Lahilahi, daughter of John
Young, received a portion of Waikahakahe. William Leleiohoku, whose first
wife was Nahi‘ena‘ena and whose second wife was Princess Ruth Ke‘elikolani,
received 1,110 acres in Puala‘a. Mary Kaoanaeha, wife of John Young and
niece of Kamehameha I, received the ahupua‘a of Kamoamoa.
Among the Puna residents, Barenaba was a school superintendent at the
time of the Mähele. He was one of the first converts to Christianity and the
first to teach the Hawaiian language to Titus Coan. Given his position, he was
probably aware of the process and had the money needed to conduct the sur-
vey. He received 11.32 acres in Kalaihina. Hewahewa filed for a 13.64-acre
coffee patch in Hapaiolaa, Kea‘au, which he had received in 1842. Haka
received six fields totaling 7.37 acres in the ‘ili of Pakalua. He was possibly a
former house servant of Coan’s who kept a house for the minister at Ke‘eke‘e
near Kehena.44 In 1854, four years after the Kuleana awards were granted, the
estimated population of Puna was 2,702. Why then, were only three of the
inhabitants of Puna awarded land?
An examination of the possible reasons (aside from those discussed in chap-

159
chapter four

ter 1) that almost the entire population of Puna did not apply or receive a land
award illustrates the plight of Native Hawaiian kua‘äina who lived outside of
the mainstream of Hawai‘i’s economic and social development. First, Puna was
isolated from the mainstream of communication and transportation networks.
It is very probable that the kua‘äina of Puna were not aware of the process or
did not realize the significance of the law proclaimed in February 1846 to
“No Na Mea kuleana ‘äina a Pau Ma Ko Hawai‘i Pae ‘äina” or “All Claimants
of Lands in the Hawaiian Islands.” Second, it is possible that the Puna Hawai-
ians did not have a way to raise the cash needed for the land surveys, which cost
between $6 to $12. Wages at the time were normally between 12 1 ⁄2 cents and
33 cents a day. There were few wage-earning jobs in Puna. Cash would have
to be raised from selling extra fish or other products, which was difficult given
the people’s subsistence level of living. Third, continuing volcanic activity in
Puna may have discouraged claimants from filing for a particular lot. It is also
possible that some Native Hawaiian families believed that the lands of Puna
were the domain of Pelehonuamea and her family of deities and could not be
claimed for ownership by individuals. Fourth, at least some of the Puna
Hawaiians filed their land claims after the deadline. In an 1851 petition to the
legislature, several Puna residents asked to be issued land grants without
penalty because they had filed their claims after February 14, 1848.45
Between 1852 and 1915, 526 land grants and patents were issued in Puna.
Out of this number, 275 were issued for the ahupua‘a of Ola‘a. Some of these
grants represent kuleana claims that were not awarded. Eventually more pub-
lic lands were opened for homesteading in Puna. However, large tracts
remained in the public domain and continued to be openly accessed for hunt-
ing, gathering, and spiritual practices by Native Hawaiians with a long history
of settlement in Puna.
The 1858 tax records for Puna shows how many men over twenty and how
many men under twenty were living in each ahupua‘a and paid taxes. There
were a total of 894 males over the age of twenty who paid poll taxes in Puna
in 1858. A hundred and thirty males under the age of twenty paid taxes. This
would have been after the devastating measles epidemic of fall 1848 that,
according to Samuel Kamakau, claimed the lives of one-third of the popula-
tion, and also after the smallpox epidemic of 1853 and the epidemic of colds
in 1857. Very definitely, in February 1848 there were substantially more than
three Kanaka ‘öiwi who would have qualified as applicants for land.
With the break-up of the traditional land and labor system by the estab-
lishment of private property, Hawaiians were pushed into the market econ-

160
puna: a WA H I PA NA sacred to pelehonuamea

omy to earn cash to purchase, lease, or rent land and to pay taxes. In Puna the
primary resources for commercial sale were the coastal fisheries, salt, pulu
(the hairy fibers from the hapu‘u fern), ‘öhi‘a timber, and open land for cattle
and goat grazing. Isaac Davis traveled around Hawai‘i to conduct an assess-
ment of the Crown lands. Of the Crown land in Puna he wrote:

Kaimu ahupua‘a in Puna, was the first land that I saw. Cocoanuts and pan-
danus are the only things growing, there is sand on the sea shore, and rocks
are the most. Waiokolea, and Ili in Kaimu, is of the same quality, but there
is a fish pond in Waiokolea, it is a good pond, and I have leased it for
$909.00, and R. Keelikolani has it.
Apua, Ahupuaa in Kau, I do not know the extent of this land, not at the
sea shore, but, on making observation, there is a lot of stone on that land,
Kapaakea’s man told me that salt is the only product on this land, but it is
very little. And I called the natives to lease it, but there was no one wanted
it, and no one made a reply.46

Pulu processing became an industry in Puna in 1851. Pulu was used for
mattresses, pillows, and upholstery. At its peak, in 1862, Hawai‘i exported
738,000 pounds of pulu worldwide to San Francisco; Vancouver; Portland,
Oregon; and Australia. It sold for 14 to 28 cents a pound.
In 1860 Abel and C. C. Harris and Frank Swain leased the ahupua‘a of
Pänau for the hapu‘u on the land. Kaina and Heleluhe requested government
leases on Lae‘apuki and Panauiki. Kaina maintained two pulu picker camps,
one near Makaopuhi Crater and the other near the present Keauhou Ranch
headquarters. Pulu was collected, processed, and dried at these camps and
then hauled down the pali to Keauhou Landing on mules.
In an article about pulu in 1929, Thomas Thrum suggested that the pulu
industry broke up homes and dispersed the Hawaiians:

The sad part of the story lies in the fact that the industry caused homes in
various sections to be broken up, the people moving up into the forests to
collect the pulu. In many cases whole families were employed, who provided
themselves with rude shelter huts meanwhile, to live long periods at a time
in damp, if not actually rainy quarters, without regular and proper food,
that resulted in colds and illness.

As a survey report by H. L. Lyman in 1865 showed, much of the unsold


government lands of Puna were covered by lava, barren of resources and
unsuitable for agriculture:

161
Figure 23 Mules were popular for traveling on the unpaved roads and rugged country-
side of Puna. 1894 or 1895. H. W. Henshaw, Bishop Museum.
puna: a WA H I PA NA sacred to pelehonuamea

1 Makuu to Kaohe, a large tract mauka, rocky land, worth little.


2 Kalapana, about 200 acres, mauka, rocky land.
3 Kaapahu, about 300 acres or 400, mauka, rocky land.
4 Laeapuki, about 200 acres, mauka, rocky land.47

A description of the Ola‘a area at the end of the nineteenth century gives
an insight into the changes in the way of life of the kua‘äina of Puna during
the nineteenth century:

Some fifty years ago about 1,000 natives were living on the margin of the
virgin forest and Pahoe-hoe rock along the trail connecting Hilo town with
the crater of Kilauea, island of Hawai‘i, in a spot corresponding to the
present 22-mile point of the Volcano road. Making of “kappa” [native bark]
out of “mamake” bark [Pipturus albidus], of olona fiber for fishing nets out
of Touchardia latifolia, and capturing “O-U” birds for the sake of the few
precious yellow feathers under the wings, of which luxurious royal gar-
ments were manufactured—those were the industries on which they lived.
For the reasons common to all the native population of the islands, viz.,
the introduction of new germs of disease—syphilis, leprosy, tuberculosis,
smallpox, etc. —this settlement gradually dwindled away, and in 1862 the
few surviving members migrated to other localities. At present only patches
of wild bananas, taro, and heaps of stones scattered in the forest indicate
the places of former habitation and industry. I have heard, however, that
as late as the seventies Kalakaua still levied a tax on olona fiber from the
natives of Puna and Olaa districts, which fiber he sold at high prices to
Swiss Alpine clubs, who valued it for its light weight and great strength.48

Throughout this period, subsistence fishing, ocean gathering, hunting, and


forest gathering were still the primary livelihoods for the kua‘aiana living in
the Puna district. Despite strong economic and social forces pushing to dis-
perse the ‘ohana, Native Hawaiians still maintained strong family ties and
obligations. They continued to look after the welfare of their relatives and
friends. Native Hawaiians who had to move away to earn a living were peri-
odically able to return to visit and find refuge among relatives and respite from
the drudgery and alienating social conditions of wage labor. Some left their
children to be raised by grandparents in the traditional rural setting rather
than in a port town, thus perpetuating the practice of hänai or the raising of
children by relatives.
In June 1873 the Boundary Commission conducted hearings to settle the
boundaries of the privately held lands in the ahupua‘a of Kea‘au in the district

163
chapter four

of Puna. Uma was a kama‘äina (native-born) expert witness. He described him-


self as a Native Hawaiian who had been born at Keauhou in Kea‘au “at the time
of the return of Kamehameha Ist from Kaunakakai, Molokai.” He provided
testimony that described the natural features and resources in the area and the
traditional cultural and subsistence activities of Native Hawaiians in the dis-
trict. According to Uma, the inland forest of Puna was used for bird catching
and for the gathering of sandalwood and olonä. Uma also described caves that
had been used for shelter during the wars between the Hawai‘i chiefs:

I have always lived there and know the boundaries between Keaau and
Waikahekahe. My parents pointed them out to me when we went after birds
and sandalwood. Waikahekahe Nui joins Keaau at the sea shore at Kaehuo-
kaliloa, a rock that looks like a human body, which is between two points,
the point on Waikahekahe is called Kaluapaa and the one on Keaau Keahu-
okaliloa, thence the boundary runs mauka to place called Koolano, the
pahoehoe on the North side is Keaau and the good ground where cocoanut
trees grow is on Waikahekahe. In past days there was a native village at this
place. Thence mauka to Haalaaniani (Ke Kupua) when the old road from
Kalapana, used to run to Keaau thence the boundary runs to Wahikolae,
two large caves, the boundary runs between them thence mauka, to another
cave called — Oliolimanienie, where people used to hide in time of war . . .
Keaau on the Hilo side of the road running mauka, thence to Kikihui, an
old Kauhale [living compound] for bird catchers, thence to Hoolapehu,
another old village, thence to Alaalakeiki, which is the end of Waikahekahe
iki and Kahaualea joins Keaau. This place is at an old Kauhale manu [bird
catcher’s compound] . . . From the Hilo Court House to the Government
School house, thence mauka to KeeKee; Kauhale kahi olona [olona fiber
combing compound] in Olaa, the boundary is a short distance from the
Government road, on the South East side . . . the sea bounds Keaau on the
makai side. Ancient fishing rights, including the Uhu which was konohiki
fish extending out to sea.49

Puaa was another kama‘aina expert witness who testified on the boundaries
of Kea‘au. His testimony reveals areas in Kea‘au where there were breadfruit
trees, plots cultivated by Native Hawaiians, marshy areas, springs, and ‘öhi‘a,
orange, and banana trees:

The boundary between Keaau and Waikahekahe is the land of, or place
Keahuokaliloa, thence mauka along Waikahekahe to pahoehoe, on Hilo
side of a place called Kukuikea (where the natives cultivate food, and where

164
puna: a WA H I PA NA sacred to pelehonuamea

bread fruit trees grow), thence to Hilo side of Waiamahu a large place that
fills with water in the rainy season, thence to Koolano, the pahoehoe on the
Hilo side of it is Keaau the soil is on Waikahekahe nui thence mauka along
the road to Halaaniani, Keaau on the Hilo side of road; Halaaniani is a
puupahoehoe, in a grove of ohia trees, called Keakui . . . below Kahopua-
kuui’s houses, to a place called Kilohana where Oranges are growing there
the boundary of Keaau and Olaa leaves the Volcano road, and runs mauka
above these Orange trees, thence to an ohia grove called Puaaehu, thence
to Waiaele . . . A water spring with banana trees growing near it used to
be an old kauhale.

Kenoi, originally from Kapapala in Kaü, provided testimony on the bound-


aries as he had learned them from companions with whom he went gathering
in the forests. He spoke of going after the ‘ö‘ö bird in Keauhou; gathering
sandalwood in Kahauale‘a and at Pu‘ukea; and catching the ‘uwao at Nama-
mokale, opposite Kauanahunahu. He also spoke of two ponds, Nawailoloa
and Kilohana, on the road to Pänau from Palauhulu. Nailima, a kua‘äina from
Ola‘a, also provided testimony. He verified the accounts of those who went
before him and also identified in his descriptive testimony a hill covered with
pu‘uhala by Kilohana, an old village at a place called Ka‘aipua‘a, and a pond
with aweoweo growing in the water at Waiaele on the old road from Ola‘a to
Po‘ohölua. Waipo, a kua‘äina from Waiäkea, identified a small cave where
natives worshiped idols at a place called Kawiaka‘awa and a place called
Na‘auo, between Mäwae and Waiaele, that people used to flee to and live in
during times of war.
A description of the land use pattern and practices in Kapoho and Keahi-
alaka in the late nineteenth century was recorded in a brief filed by attorneys
Hitchcock and Wise and filed with the Boundary Commission for the Third
and Fourth Circuits of Hawai‘i on March 20, 1897. Of significance is the fact
that where two ahupua‘a were owned in common by the same family, in this
case by Charles Kana‘ina and his son William Charles Lunalilo, the bound-
aries between the two parcels lost significance. The brief also speaks of an iso-
lated section of land that belonged to no one and was therefore open to all,
similar to the “Kamoku” in Hämäkua. In part it stated:

The two ahupuaas of Kapoho, and of Keahialaka, were practically held by


one family. By the great Mähele, Kapoho was confirmed to C. Kanaina,
while Keahialaka was confirmed to his son W. O. Lunalilo. The influence
remains that the laws and customs which in the case of adjoining ahupuaas

165
chapter four

under different owners would have held and trespass, the one to the other
thereby enjoined, were in this instance permitted to lapse. It is furthermore
probable, and the presumption is given force by the subsequent isolation of
Kaniahiku so-called that it was an Okana “a no man’s land,” similar to the
Kamoku of Hamakua. This trend of the Puna coastline on both sides of the
East Point with ahupuaas extending back rectangularly from the sea coast,
would naturally bring about an irregular shaped remnant in the interior
similar to those in the North Kohala District, and the upper Keauhou lands
of Kona.50

In Puna, Joseph Nawahï, a founder of the Hui Aloha ‘äina (Hawaiian Patri-
otic League), had a strong following of royalists. On May 23, 1893, four
months after the overthrow of the Hawaiian monarchy, Rufus A. Lyman,
patriarch of the Lyman Estate, which now owns substantial landholdings in
Keahialaka and Kapoho, wrote to his colleague, M. Whitney in Honolulu,
suggesting that the provisional government open up government and Crown
lands for homesteading by Native Hawaiians. He felt that such a gesture would
win the support of Native Hawaiians for the illegal provisional government
and undermine the influence of the royalist Joseph Nawahïokalaniopu‘u in the
district:

Here in Puna there are only three Crown Lands Ola‘a, Kaimu and Apua
next to the Kau boundary. The Govt. lands are scattered all through
District, and large tracts near the villages especially Opihikao, Kamaili,
Kehena, and not under lease. And there are quite a number of young men
there with families who own no land, who will probably remain in Puna
and cultivate coffee, kalo, oranges, etc., if you get them settled on land
they can have for homes for themselves.
Nine of them have commenced planting coffee on shares for me.
Puna has always been Nawahi’s stronghold, and I want to see his hold on
natives here broken. And I think it would help do it, if we can show natives
here that the Govt. is ready to give them homes, and to improve the roads.

In 1894 the provisional government set up the Republic of Hawai‘i, which


instituted a program of opening up government lands for homesteading under
the Land Act of 1895. In Puna, as Lyman had predicted in his letter to Whit-
ney, homestead grants were quickly purchased and cultivated in coffee. Cof-
fee acreage expanded from 168 acres in 1895 to 272.5 in 1899 in Ola‘a and
Pähoa.51

166
puna: a WA H I PA NA sacred to pelehonuamea

Puna, Territory of Hawai‘i


Economic development in Puna centered on the scarcely populated inland
forest areas around the towns of Pähoa and Ola‘a in the twentieth century. A
multiethnic plantation community also developed in and around these towns
as immigrant Japanese, Puerto Rican, and Filipino laborers were imported to
work on the developing sugar plantations. Hawaiian families continued to live
along the coastal areas in lower Puna, particularly around Kalapana.
The Puna Sugar Company was established in 1900 in Kapoho. The low-
land forest was cleared for cane fields, and railroads were built. Puna Sugar
expanded around Pähoa and Ola‘a.
At the turn of the century coffee was still an important agricultural indus-
try in Puna. Cattle ranching was also significant. The Shipman family, a major
landowner in the district, ran the Shipman Ranch in Kea‘au. Pineapple was
started for export to California.
In 1908 the Hawaiian Mahogany Company erected a lumber mill in Pähoa
and sent out its first shipment of 20,000 ‘öhi‘a log ties to the Santa Fe Rail-
road. In 1910 the company became the Pähoa Lumber Mill and obtained cut-
ting rights to 12,000 acres of territorial forest in Puna.52 Finally, in 1911 the
territorial government designated 19,850 acres as the Puna Forest Reserve to
protect it from logging. In 1928 the forest reserve was expanded to include a
total of 25,738 acres.53
Charles Baldwin’s Geography of the Hawaiian Islands, published in 1908,
provides a glimpse of the Puna district at the turn of the century. Efforts to
actively develop agriculture in the midst of old lava flows, seismic activity, and
heavy rainfall included the cultivation of vanilla, tobacco, pineapples, rubber,
and sugar cane.

table 5 censuses of puna, 1900–1960


year population source

1900 5,128 Twelfth U.S. Census: 1900


1910 6,834 Thirteenth U.S. Census: 1910
1920 7,282 Bureau of Health Statistics, Board of Health, pop. est.
1930 8,284 Fifteenth U.S. Census: 1930
1940 7,733 Sixteenth U.S. Census: 1940
1950 6,747 Seventeenth U.S. Census: 1950
1960 5,030 Eighteenth U.S. Census: 1960

167
chapter four

The rainfall is so great in parts of the district that this lava has been rapidly
decomposed, and the heaviest of forests are to be found, as in Olaa and the
region about Pähoa [the Puna Forest Reserve]. A large part of the soil of
upper Olaa is ash which probably came from Kilauea; the great fertility
of this soil is due to the decayed vegetable matter which has been added
to it . . .
The Olaa section of Puna is a fine agricultural region, but, owing to
the want of a market, small-truck farming does not pay. However, vanilla,
tobacco, pineapples, and bananas grow well; and the rubber industry is
destined to be an important one, as the climate is particularly well adapted
to the growth of rubber trees. The cultivation of coffee in Olaa has been
abandoned, as the trees did not thrive there.
All the lower lands of Olaa are planted with the cane of the Olaa Sugar
Company. This is one of the largest plantations on Hawaii, and occupies
nearly all of the available cane land of the Puna district, including the
Kapoho and Pähoa tracts . . .
A long section of the Puna coast, thirty or forty miles, shows evidence
of having sunk: cocoanut trees are found below the tide level, or their dead
stumps stand out in the sea.
At Kapoho there is a warm spring . . . Other interesting features of
Puna are: the lava tree casts found in the forest above Kapoho; the bowlders
strewn along the coast near Pohoiki by the great 1868 tidal wave; the heiau
of Wahaula in farthest Puna.54

In 1913 the Hilo Board of Trade published a guidebook called The Island of
Hawai‘i, by Henry Walsworth Kinney, to promote tourism around the island.
The Kïlauea Volcano and its spectacular sites, trails, and forested areas are
prominently featured. Ola‘a and Pähoa were described as the centers of eco-
nomic development for the Puna district:

The district of Puna may, for the sake of clearness, be divided into two
sections, the Olaa region, the north half, and Puna proper. The former
consists in the main of the great Olaa sugar plantation, and forest which
has been partially cleared, while some tracts are used for cattle. The middle
part of the district, with Pähoa as the center, is used for extensive lumber
operations. The remainder, Puna proper, is covered by forest and old lava
flows, most of them covered with vegetation. In spite of its exceptional
beauty and the fine opportunity it offers for seeing the typical Hawai‘i,

168
puna: a WA H I PA NA sacred to pelehonuamea

which is so rapidly disappearing in the march of progress, it is compara-


tively little known.55

Ranching and sugar plantations flourished at Ola‘a. ‘Öhi‘a and koa lumber
operations were established at Pähoa. Kinney described lower Puna as a tradi-
tional Hawaiian subsistence area. Kaimu and Kalapana were the main Hawai-
ian villages in lower Puna:

At the beach the road enters first the village of KAIMU, exclusively
Hawaiian, with a large grove of cocoanut trees surrounding a fine semi-
circular sand beach. Care should be exercised in bathing on account of
the undertow. Less than a mile further on, westwards, lies the village of
KALAPANA, one of the largest Hawaiian villages in the Islands. There
are no white inhabitants, and only a couple of Chinese stores . . . KALA-
PANA still supports quite a large population, and is a very pretty village,
having like all the Puna coast villages, a fine growth of cocoanuts, puhala
and monkeypod trees. The landing is so rough that it is used now only
for canoes.56

Kalapana’s tourist attractions included Pu‘u o Hakuma, a cave used as a


place of refuge during war; the Niukukahi heiau; the ranch and Hawaiian vil-
lage at Kahaualea; a mineral bathing pool called Punalu‘u; and the Waha‘ula
heiau.
Kinney also described tourist attractions in Kapoho, including the Waia-
Pelehonuamea crater, famous as the first residence of Pelehonuamea in Puna,
and the three craters mauka of that, created as Pelehonuamea searched for a
suitable home before reaching Kïlauea. Green Lake was said to be situated
within a ring of five craters. Küki‘i heiau, the hot springs of Puna, and a pretty
Hawaiian village called Koa‘e were also featured. Along the coast between
Kapoho and Kaimu Kinney described Cape Kumukahi; the almost deserted
Pohoiki village; Opihikao, with its hot-spring cave, and the small villages of
Kamaile, Kehena, and Kaueleau.

Mo‘olelo of Kalapana: The Territorial Years


The Kalapana Oral History Project, completed in 1990 by the University of
Hawai‘i at Hilo anthropologist Charles Langlas and student researchers, is a
primary source of information about the life of Native Hawaiian kua‘äina in
Lower Puna during the territorial period. I have drawn upon this source,

169
chapter four

together with accounts by E. S. Craighill Handy and documents from the Vol-
cano National Park archive, to describe the way of life of the kua‘äina of
Lower Puna.
Throughout the territorial years, the majority of the food of the kua‘äina
in Puna was produced at home. ‘Uala (sweet potatoes), kalo (taro), and ‘ulu
(breadfruit) were the main staples. Seafood, especially fish, ‘opihi (limpet),
and limu (seaweed), was the main protein. Chickens, pigs, and cattle were
raised. Wild pigs and goats were hunted, and their meat was usually smoked.
Some households kept cows for milk and even made butter. When cash was
earned, special items from the store such as flour, sugar, tea, coffee, and rice
could be bought.
Sweet potatoes were usually grown around the home. Families also grew
chili peppers, onions, and sometimes pumpkins, watermelons, tomatoes, or
cucumbers. Families in Kalapana usually had a taro patch in the uplands as far
as the forest, which was as much as three or more miles from their house lots.
Handy wrote that in 1935, when he toured Puna to appraise the old native
horticulture, “one energetic Hawaiian of Kapa‘ahu had cleared ‘öhi‘a forest,
at a place called Kaho‘onoho about 2.5 miles inland, and had a good stand of
taro, bananas, and sugar cane in two adjacent clearings.” 57
Pigs were allowed to run free, but to keep them tame and near the home,
they were fed sweet potato vines and tubers after harvesting, papayas, man-
goes, or breadfruit. Each family had its own way of marking its pigs by notch-
ing or slitting the ears or cutting the tail. Some pigs went wild and wandered
up the Kïlauea mountain, even above the zone where the families cultivated
taro. These were hunted with dogs.
The kua‘äina in Kalapana utilized many methods of fishing during this
period. Net fishing for ‘öpelu (mackerel) was the highest-yielding method.
The fish was usually dried for later consumption or for sale. Aku was also
caught for subsistence and for sale in season. As late as the 1930s, ‘öpelu fish-
ing in Kalapana was conducted in accordance with traditional and customary
rituals and was a community effort:

The ‘öpelu season began in the summer months, after a first-fruits sacrifice:
a fish from the first catch was placed on the kü‘ula rock at the beach. The
kü‘ula rock was kept by a guardian, who brought it out for the ritual, and
then took it away for safe-keeping. Traditionally the year was divided into
two seasons, a period from approximately February to July, when aku could
be caught and ‘öpelu was taboo, and a period from approximately August to

170
puna: a WA H I PA NA sacred to pelehonuamea

December, when ‘öpelu could be caught but aku was taboo. The opening of
the ‘öpelu season was marked by a fish sacrifice.58

‘Öpelu fishing went from daybreak to evening. The canoes from a village
generally went out together and kept each other in sight in case one should
get into trouble. When they returned, people would be waiting to help carry
the canoe up, and everyone would get a share of fish. Later in the day or at
night the canoes might go out for ‘u‘u or kawele‘a. On dark nights, if the
fishermen went out, children would gather at the beach and keep bonfires of
coconut leaves going to guide the fishermen back to shore. Until 1926 the
nets were made of olonä from the wet uplands of Puna. After that they were
replaced by store-bought cord.
A one-room house with a separate cookhouse was the usual style in Kala-
pana around 1900. By the 1920s several families still lived in such dwellings,
but the majority of the families were already living in sizable multiroom board
houses built in the Western style. Many families had also installed kerosene
stoves in their houses. Since there was no running water in the Kalapana area,
families had outhouses for toilets. Water barrels were used to collect water
from the roof for drinking and cooking. In times of drought, they had to drink
brackish water from the ponds. Brackish ponds were used for bathing, for
doing laundry, for rinsing off saltwater after coming from the ocean, and for
watering stock.
Through the 1920s and 1930s families still made their own poi from bread-
fruit or the taro they grew in the uplands. They usually made enough poi to
last the whole week. After this, there was only a limited amount of daily cook-
ing to do, mostly broiling fish on the fire or salting shellfish to eat with the poi.
Sometimes the family might cook a pig in the imu, stew dried meat, or make
rice over the fire or kerosene stove.
Weaving lauhala mats for home use and for sale was a large part of a
woman’s work during this period. Lauhala grew all along the coast, but women
often went to Kehena to gather good-quality leaves.
Lü‘au continued to be held for family gatherings to celebrate special occa-
sions and life cycle events such as birthdays, weddings, anniversaries, and
funerals. The lü‘au for Christmas usually lasted through New Year’s. ‘Ohana
relationships remained strong. Even the practice of hänai (adoption between
family members) continued in Kalapana.
Through the 1920s and early 1930s relations with the outside were limited
by distance and the difficulty of travel. The outside world was primarily rep-

171
chapter four

resented in Kalapana in the form of schoolteachers, ministers from Hilo, Chi-


nese stores (which sold goods from the outside), and campaigning political
candidates. Automobiles were introduced during this time, and by the late
1930s most families had a car. Still, most people went to town only once or
twice a month to shop for cloth, kerosene, and food items they didn’t grow.
Rice and flour were purchased in big bags. Since they seldom went to town,
the kua‘äina of Kalapana did not go to Western doctors and hospitals. Kala-
pana kua‘äina relied on Native Hawaiian medicine, using herbal remedies for
sickness and broken bones.
Even though cars became common in the 1930s, it was impractical to com-
mute to work every day. Those who got a job outside the community usually
relocated, even if the workplace was as close as Pähoa. A few men stayed out-
side through the week and came back for the weekend. The men who lived in
Kalapana usually combined subsistence farming and fishing for family food
production with part-time work for cash—roadwork for the county and small-
scale selling of vegetables, fish, or pigs. The county road from Kapa‘ahu
through Kaimu and up to Pähoa was a one-lane gravel road. Nearly all the
Kalapana men did roadwork for the county, breaking up rock into gravel. Each
man worked an eight-hour day for $2, four or five days a month. The crew
rotated so that all the members had a chance to work the same amount.
Additional cash could be made by selling extra ‘öpelu to the Chinese store
owners to dry, or to be consumed fresh in Pähoa. Some grew ‘awa in the
uplands that was cut and dried and sold to a buyer from Hilo for export to
Germany. Some husked and dried coconut to sell as copra to Chinese store
owners. Sometimes Chinese drove from Hilo to buy pigs in Kalapana. As
mentioned above, the women sold lauhala, weaving mats to fill orders from
Hilo and Honolulu. They also sold smaller items such as hats and fans to sell
to tourists. Children sometimes sold coconuts to tourists and posed for pic-
tures. In 1918 the movie Bird of Paradise was filmed at Kaimu beach. Grass
huts were built, and the people of Kalapana were paid to wander around in
sarongs.
In the Kalapana village there were three churches — Catholic, Hawaiian
Congregational, and Mormon. As discussed above, Fathers Walsh and Damien
de Veuster established the Catholic Church in Kalapana. Since the time of
Father Damien there has always been a resident priest in Puna and a strong
Catholic congregation. During the term of Father Loots (1881–98) a small
Catholic church was built at Kalapana village. In the twentieth century the

172
puna: a WA H I PA NA sacred to pelehonuamea

main church and rectory were at Pähoa, and the priest went to Kalapana to
hold weekly services.59
In the 1840s the Congregationalists built a church at Kalapana and a meet-
ing house in Kamoamoa. The first minister for the Kalapana under the Hawai-
ian Evangelical Association was Papapa Barenaba, who remained there from
1869 to 1873. After the 1868 earthquake and subsidence the Kalapana church
was rebuilt. By 1905 a third church, called Mauna Kea, had been constructed
on the same location. It was rebuilt again in 1930. From the nineteenth cen-
tury through the present the Hawaiian Congregational churches held periodic
conferences at the island level and the all-island level for discussion of church
business. By the 1880s a feature of the conference was a song competition
between the choirs of the various churches. In 1886 the Kalapana congrega-
tion won the competition and was presented with a silver pitcher and goblet
by Princess Lili‘uokalani herself. The mo‘olelo about this award has been
proudly passed down from one generation to the next.
The Mormon church was built in Kalapana some time before 1910.

Hawai‘i Volcano National Park


In 1932 a new force entered the lives of the Kalapana people. The Hawai‘i
Volcano National Park, urged on by the governor’s office, the Hawai‘i County
Board of Supervisors, and prominent citizens, proposed expanding the park to
include all the land from Apua over to Kaimu Black Sand Beach. The people
in Kalapana strongly opposed the plan. Russell Apple interviewed Edward G.
Wingate, who served as superintendent of the Hawai‘i Volcano National Park
at the time of the proposed acquisition. Wingate said that he supported the
Native Hawaiians in Kalapana and felt it was wrong of the federal government
and the park service to dispossess the Native Hawaiians of their homes, their
land, and their traditional way of life. A compromise was reached. The Hawai‘i
Volcano National Park would expand to include the six ahupua‘a of ‘Apua,
Kahue, Kealakomo, Panaunui, Lae‘apuki, and Kamoamoa; parts of Pülama
and Poupou; and Keauhou in the Ka‘ü district. However, the lands from Kala-
pana over to Kaimu were deleted from the extension proposal.
Wingate was still concerned about negative impacts on the Kalapana
Native Hawaiians’ way of life of the road that was to be built to link the Chain
of Craters road to Kalapana, all the way to Kaimu. He believed that the road
would put pressure on the Native Hawaiians to sell their homes in Kalapana

173
Figure 24 Tourists attracted to the volcano, its rainforest, and natural phenomena such
as the Thurston Lava Tube continued to change the lives and livelihoods of the kua‘âina of
Puna during the Territorial period. 1925. Tai Sing Loo, Bishop Museum.
puna: a WA H I PA NA sacred to pelehonuamea

to developers or others and that their livelihoods, which were still dependent
on the land and sea, would be destroyed. To make it possible for the Kalapana
kua‘äina to continue their way of living, he proposed that home sites be made
available to them in the park extension so that the villagers could move into
the park when they saw the need. In addition a fishing provision was included
that allowed only Kalapana residents and those accompanied by a local guide
to fish within the park extension. No Native Hawaiian was precluded from
fishing in that area provided there was a local guide. This provision, accord-
ing to Wingate, was “to protect the fishing for the people who lived from the
sea and who lived from the land, to have some food source from the sea as
some areas have been fished out.” He also noted that serving as a guide would
provide jobs and a source of a little cash income for the kua‘äina in the dis-
trict. Apple summarized Wingate's thinking as follows:

A new village inside the Kalapana Extension was foreseen. The idea was
a subsistence-type arrangement, with Hawaiians living in a traditional
manner—fishing offshore and along the coast, houses near the shore and

Figure 25 Kua‘âina of Puna established and maintained roads up to and within Volcano
National Park and down to Kaimû Black Sands Beach in Kalapana. 1920s. Theodore Kelsey
Collection, Hawaiian Historical Society.

175
chapter four

agricultural plots inland. Exclusive fishing rights for those still living in
Kalapana and for those living within the Extension were included.60

In 1938 the U.S. Congress passed the Kalapana Extension Act (52 Stat. 781
et seq.), which set an important precedent by including a provision to lease
lands within the extension to Native Hawaiians and to permit fishing in the
area “only by Native Hawaiian residents of said area or of adjacent villages and
by visitors under their guidance.” The special traditional subsistence lifestyle
of the Native Hawaiians in Kalapana was acknowledged by the U.S. Congress,
and measures were passed to protect it.61
Under the New Deal, federal programs created new jobs for the men of
Kalapana. The federal government funded a county project to improve Kala-
pana Park and various road-building projects in Puna. The Civilian Conser-
vation Corps (CCC) established a camp for young single men at the volcano.
They cut trails, built stone walls, and were trained in carpentry skills. As mili-
tary construction expanded in Honolulu in preparation for potential war with
Japan, Honolulu became a boom town, attracting workers from the mainland
and from neighboring islands. Many of the kua‘äina of Kalapana moved there
on the eve of the war.62

World War II and Puna


World War II had a profound effect on Hawai‘i. In Puna, those who remained
behind feared a Japanese invasion by sea. The coastline was watched and
guarded by soldiers stationed in the Kalapana areas. Observation points were
set up at Pänau and at Mokuhulu. The beach at Kaimu and Kalapana was
strung with barbed wire to stave off an enemy landing. Initially the Kalapana
people were not supposed to go through the wire, but eventually the soldiers
let the people crawl through to fish or collect seafood at the beach. There was
a nightly curfew, and blackout curtains were used so that not even a single
glimmer of light could be seen by an enemy observing the area.
The 100 to 150 soldiers stationed in Kalapana were rotated every three
months. Some camped in tents on Kaimu beach and Kalapana beach, some
lived in the school cafeteria, and others lived in the gym and the priest’s
house.
During the war there were still kua‘äina who grew taro in Puna, but many
of them were in their sixties. By the end of the war they were getting too old
to grow taro and make poi. Many younger men had left during the war, and
many of those remaining in Kalapana got jobs on the outside, which left lit-

176
puna: a WA H I PA NA sacred to pelehonuamea

tle time for taro. During the 1930s fewer canoes went out to catch ‘öpelu. The
last canoe which went out from Kaimu was that of Simon Wai‘au Bill. When
he got too old, in the late 1930s, he stopped fishing. Younger men were too
busy going to school or going out to work to learn the technique of catching
‘öpelu. At Kalapana a couple of canoes continued going out even after the
war. Eventually a boat ramp was constructed at Pohoiki, east of Kalapana, and
the canoes were replaced by motorboats.
Other forms of subsistence production continued after the war, such as pole
fishing from shore; gathering limu, opihi, and crab; and raising stock. Wild
pigs were still hunted and remained an important source of meat. Native
plants were gathered for herbal teas and medicine.

Statehood
In 1958, on the eve of statehood, the Puna district began to be parceled out
in nonconforming subdivisions of raw land without any infrastructure. Tropic
Estates bought 12,000 acres of land between Kurtistown and Mountainview
and cut it up into 4,000 lots that were put on the market for $500 to $1,000
each. The project was named Hawaiian Acres.63
Royal Gardens was opened in Kalapana in the early 1960s. One-acre lots
were sold for $995. The brochure for the development read in part:

Along the southern shores of the Big Island, Hawaii, largest of the Hawai-
ian chain lies the historic and legendary lands of Kalapana. This site the
setting for Royal Gardens, a fertile area directly adjacent to the Hawaii
Volcano National Park with its spectacular attractions, yet only walking
distance away from lovely beach and shore areas. Royal Gardens lots are all
one acre in size, making it possible for the owners to have a small orchard
or truck garden, or a magnificent garden, as well as a home and a haven for
retirement.64

By contrast, the Bishop Museum study for the Kalapana Extension in 1959
described the coast nearest to Royal Gardens as follows: “Shoreline of low,
black, lava cliffs, battered continuously by windward waves . . . This coast
bears witness to the great volcanic forces underlying it through numerous
earthquake-opened fissures, and to the violence of tidal waves through huge
blocks of lava which have been ripped from the ocean cliffs and hurled
inland.” 65
Actually, Royal Gardens land was 40 percent ‘a‘ä (rough and broken lava

177
chapter four

rocks in tumbled heaps), 20 percent pahoehoe (solid thick sheets of lava, hard
and smooth surfaced, with no soil covering), and 40 percent ‘opihikao
(extremely rocky muck with pähoehoe underneath). Water was scarce, with
just a few widely scattered waterholes.66
Other nonconforming subdivisions similar to Hawaiian Acres and Royal
Gardens were developed in Puna prior to adoption by the county of a com-
prehensive zoning ordinance. These included Eden Roc, Fern Forest Vaca-
tion Estates, Hawaiian Paradise Park, Hawai‘i Beaches Estates, Aina Loa
Estates, Orchid Land Estates, Leilani Estates, Nänäwale Estates, Vacation
Lands, Kalapana Black Sands Subdivision, Kalapana Gardens, and Kalapana
Sea View Estates. These subdivisions gradually attracted an in-migrant pop-
ulation of retirees, ex-military, and persons seeking an alternative lifestyle to
urban centers in mainland United States.

Puna: A Cultural Kîpuka in the Late Twentieth Century


The landscape of Puna continued to be dominated by the seismic and erup-
tive phases of the Kïlauea volcano. Throughout the late twentieth century the
landscape varied from the rocky shoreline, to barren lava fields, cultivated
orchards, grassy plains, and dense rainforests. It included the Hawai‘i Volca-
noes National Park, large undeveloped nonconforming subdivisions, unset-
tled Hawaiian homelands, forest reserves, and small concentrations of popu-
lation. The district was subject to heavy rainfall and periodically experienced
severe flooding.67
The Puna district of the island of Hawai‘i as a whole, and Lower Puna in
particular, has been a rural area of Native Hawaiian cultural continuity. Of
the 452 Native Hawaiians who lived in the Puna district in 1970, 77 percent
or 350 lived in Lower Puna. In 1980, 1,334 Native Hawaiians lived in Puna,
of whom 75 percent or 1,001 resided in Lower Puna. Between 1980 and 1990
the number of Native Hawaiians in Puna increased by 296 percent, to 3,953.
For the first time, the majority of Native Hawaiians who lived in Puna resided
outside Lower Puna. Only 38 percent lived in Lower Puna, while 62 percent
lived elsewhere. This was due both to migration of Native Hawaiians into
Puna from Hilo and other islands and to the displacement of Native Hawai-
ians from Lower Puna by volcanic flows out of the Kupaianaha lava lake.
Table 6 shows population trends in the district as a whole from 1970 to
2000.
As late as 2000, modern infrastructure for households and farm lots such

178
puna: a WA H I PA NA sacred to pelehonuamea

as electricity, piped water, and sewage was still not available in many parts of
Puna. Puna residents relied on generators, water catchments, centralized
county water stands, and outhouses for their households.
There were four major water systems in the district: Olaa–Mountain View,
Pähoa, Kapoho, and Kalapana. Hawaiian Beaches had a privately owned water
system. Glenwood and Volcano were not serviced by any water system and
depended on roof catchment systems. There were no municipal sewerage sys-
tems in Puna. Most residents used cesspools and individual household aero-
bic treatment units. Aside from the primary routes, the majority of roads in the
Puna district were substandard, and many were only cinder surfaced. Puna had
thousands of nonconforming residential lots that lacked the basic improve-
ments necessary for development or were being kept vacant for future specu-
lation. Construction in the nonconforming subdivisions increased with strip
residential development along the highways. There were three public-school
complexes in the Puna District in the communities of Keaau, Mountain View,
and Pähoa.
Economically, Puna was primarily an agricultural district. Diversified agri-
culture prospered in the form of truck farming of lettuce, flowers, and cab-
bage in the volcano area; papaya groves in Kapoho; and flowers, principally
anthuriums and vanda orchids, in the Mountain View, Pähoa and Kapoho
areas. Factors inhibiting the growth of these industries were a shortage of
labor and housing, processing requirements, and plant disease. Vegetables and
a variety of fruits, primarily oranges and tangerines, were grown throughout
the district. Macadamia nuts were planted on the Hilo side of Kea‘au. With
the closing of the Puna Sugar Company in 1984, former sugar lands were sold
to former workers to farm. They planted papayas, bananas, alfalfa, and trees
for biomass. There were 197,900 acres zoned for agricultural use in Puna in
2000, but fewer than 50,000 acres were actively used for agriculture. The

table 6 ethnic population of puna, 1970–2000


ethnic group 1970 1980 1990 2000

Hawaiian 452 1,334 3,953 9,325


Caucasian 1,237 5,078 9,515 22,010
Other non-Caucasian 3,465 5,339 7,313 NA
Total population 5,154 11,751 20,781 31,335
Sources: State of Hawai‘i, Department of Business and Economic Development & Tourism,
State of Hawai‘i Counties and Districts, 1991; U.S. Census, 2000.

179
chapter four

majority of agriculturally zoned areas were subdivided for large-lot residen-


tial purposes.
The major industrial activity in Puna was a large macadamia processing
plant northeast of the sugar mill. Other industrial activities included a kim
chee factory, quarrying of lava materials, slaughterhouses, bakeries, flower
packaging, papaya processing and packing, and several cottage industries.
These were primarily located around Kea‘au and Pähoa, outside of lower
Puna. There were no major government installations in the district.68
The rocky coastline, made up of sheer cliffs in many sections, was subject
to tsunami inundation and subsidence. Inland areas were vulnerable to volcanic
and seismic activity. These natural phenomena discouraged the development
of major resorts or hotels in the district, although modest bed-and-breakfast
establishments were established in Puna.
Native Hawaiian residents in the district supplemented their incomes from
jobs or public assistance by engaging in subsistence fishing, hunting, and gath-
ering for the households of their ‘ohana. The fishermen, hunters, and gather-
ers utilized and exercised their traditional access to the ocean offshore of the
Puna district and the adjacent mauka (upland) forest lands. The forest afforded
access to middle-elevation plants and resources for Native Hawaiians who
lived in each of the ahupua‘a of the Puna district.
Native Hawaiians of the district utilized the forests of Puna from genera-
tion to generation to gather maile, fern, ‘ie‘ie, ‘öhi‘a, and other such native
plants for adornment, weaving, and decoration. They also gathered plants such
as ko‘oko‘olau, mämaki, and noni for herbal medicine.
Due to the alteration and degradation of low- and middle-elevation forests
in other parts of Hawai‘i Island and the public status of the forests in Puna,
Native Hawaiians from other parts of the island and from O‘ahu also regu-
larly gathered liko lehua, maile, fern, ‘awa, and other native plants for hula and
lä‘au lapa‘au (traditional Hawaiian herbal healing) purposes from this forest.

Puna Kua‘âina in the Twentieth Century


Evidence that subsistence activities continued to be integral to the lives of the
kua‘äina of Puna is provided in three studies conducted between 1971 and
1994. A survey of the role of hunting in the Kalapana-Kaimu Native Hawai-
ian community was conducted in 1971 by the University of Hawai‘i geography
and anthropology departments and by the School of Public Health. The study
revealed that hunting in the Puna Forest Reserve mauka of Kalapana-Kaimu

180
puna: a WA H I PA NA sacred to pelehonuamea

yielded meat that comprised a significant amount of the regular diet of Native
Hawaiian households in the area. Despite the fact that not every household
had a hunter, many households benefited from hunting activities because the
meat was shared among extended family members and friends.69
In 1982 the U.S. Department of Energy commissioned the Puna Hui
‘Ohana, an organization of Native Hawaiian families in Puna, to conduct a
survey of subsistence activities of Native Hawaiian ‘ohana in Puna as part of
a study to determine the social impact of developing geothermal energy in the
district. The Puna Hui ‘Ohana successfully surveyed an impressive 85 percent
of the adult Native Hawaiians in lower Puna (351 out of 413 adult Native
Hawaiians). The study found that 38 percent of those surveyed engaged in
traditional subsistence hunting in the adjacent forests, 48 percent gathered
medicinal plants, and 38 percent gathered maile in the nearby forests for
household use.70
Interviews conducted in 1994 for the “Native Hawaiian Ethnographic
Study for the Hawai‘i Geothermal Project Proposed for Puna and Southeast
Maui” with older and younger Native Hawaiian families in Puna documented
a continuity of subsistence farming, hunting, fishing, and gathering and asso-
ciated cultural customs and beliefs. This study focused on assessing the varied
cultural impacts of geothermal energy development in Puna.71
As part of the study, Native Hawaiian families were asked to indicate the
general location of trails, ancient sites, and areas of subsistence hunting, gath-
ering, and fishing on a topographic map of the district. The map produced as
a result of these 1994 interviews indicated that Puna kua‘äina fished along the
entire coastline of the Puna district and hunted primarily in the mauka
forested areas. Plants were gathered throughout the entire district, both
mauka and makai. Historic Puna trails were still used to travel from coastal
communities up to the forest. The map also showed the cultural sites that the
Puna kua‘äina used and cared for.
The 1994 interviews confirmed that traditional subsistence activities were
still an integral part of the way of life of the Puna kua‘äina in the late twenti-
eth century. Puna families engaged in subsistence when supplies such as fish
and meat ran low. They also fished, hunted, and gathered for special ‘ohana
life-cycle occasions such as birthdays, weddings, graduations, and funerals.
The amounts harvested depended upon family size—that is, the larger the
family, the greater the amount of subsistence resources required. Puna fami-
lies stressed that one must never take more of a resource than what is needed
and can be consumed. Most of the food consumed by Puna ‘ohana still came

181
chapter four

Map 3 In 1994 Puna residents mapped important cultural and natural resource areas and routes
to access these areas. Source: Matsuoka et al., “Native Hawaiian Ethnographic Study,” p. 101.

from some form of subsistence such as taro and sweet potato cultivated in
their gardens; breadfruit gathered from their yards or lowland forest; fish and
seafood harvested from the ocean; and wild pigs, goats, or cattle hunted in the
forest. Selected staples such as rice, coffee, flour, sugar, and cooking oil were,
of course, purchased in Hilo or other nearby towns.
The availability of subsistence resources varied by season. For example, cer-
tain species of fish, such as ‘u‘u and ahi, were more abundant during the sum-
mer months. Maile goes through periods of dormancy during dry months and
regrowth during the rainy season. When a resource in a particular area dwin-
dled because of overuse, a kapu or restriction on harvesting that resource was
observed to allow for regeneration. Puna kua‘äina would also weed an area or
water the plants in the wild to enhance the regeneration of the resources.
Knowledge about where and how to carry out subsistence activities was
passed down to the kua‘äina living in Puna in the late twentieth century from
previous generations. Each ‘ohana respected the boundaries of their respective
gathering and hunting areas. If someone wished to use an area outside of their
own, out of respect they would usually ask permission.
Though most of the kua‘äina identified themselves as Christians, they also
held a set of beliefs that was consistent with traditional Hawaiian spiritual

182
puna: a WA H I PA NA sacred to pelehonuamea

beliefs. They attended church on a regular basis, but they also prayed to
Native Hawaiian deities as part of the regular protocol to succeed in their
subsistence activities. They prayed for good luck before an activity and to
express gratitude for a successful catch. They regularly acknowledged the
presence of deities by asking permission to enter or take resources from their
domain. On special occasions or for particular purposes they offered chants
and ho‘okupu or offerings to pay respect to the deities.
One particular deity that they honored with chant and ho‘okupu was Pele-
honuamea. The Puna families believed that Pelehonuamea protected and
nurtured those who demonstrated respect for her. She could also harm those
who showed her disrespect or acted improperly. The location and direction
taken by some of the lava flows were interpreted by the Puna kua‘äina as Pele-
honuamea’s way of letting the people know that they were not properly caring
for the land. The flows covered over any damage to the land and restored it
to a primal form.
Subsistence activities also helped to perpetuate the knowledge and memory
of ancestors. One of the persons interviewed shared the following experience:

When I pick flowers or medicine, I take the knowledge that my father


taught me. What had to pick with, the whole process of knowing. There’s
a oneness—the whole mind and body has to be centered on the medicine
and how it’s gonna be used. You cannot think about anything else. It opens
the channel, what you give out to that source. Be focused only on one
thing, even making leis is the same concept. The whole time, while picking
flowers, I was thinking about it . . . that’s how my ancestors did it. That
spiritualness is carried on from generation to generation.

Other forms of protocol were also observed. The Puna kua‘äina did not talk
openly about their plans prior to going out on an expedition. They believed
that everything around them in nature had the ability to hear, and if whatever
they were going to hunt or fish became aware of the intended expedition, the
prey would escape or hide. Thus, if they referred to where they were going,
they would use code words such as “holoholo” instead of fishing or hunting.
In fishing, they did not take bananas with them and would give the first fish
caught back to the ocean deities. They would also express gratitude to the
deities after a successful subsistence expedition. The deities are believed to
dislike anyone who is greedy or ungrateful and to have the ability to deprive
such a person of future success in their expeditions if they took too much, did
not share, or did not express gratitude.

183
chapter four

In Puna, subsistence also served as a basis for sharing, gift giving, and trade.
After a successful hunting or fishing expedition, the young men would make
stops at the homes of family and friends, dropping off meat or fish along the
way. By the time they reached home, they usually ended up with just enough
to feed their immediate family. The küpuna were particularly dependent upon
this sharing network. For example, one of the küpuna said that he taught his
children how to hunt and fish and now they supply him with all that he needs.
Historic trails were generally used to access traditional subsistence gather-
ing areas. The trails usually ran from the coastal communities where the kua-
‘äina lived, up into the forest. Four-wheel-drive vehicles replaced horses as
the means of reaching subsistence areas. Vehicles were usually driven on dirt
roads up to the point where the trail narrows and then they hiked the rest of
the way.

Fishing and Ocean Gathering


Virtually every family in Puna engaged in fishing. While each person had his
or her own special or favorite fishing spot, most fished the entire coastline
from Kea‘au to within the national park. The majority fished off of the rocky
coastline or in shallow areas near shore. They threw net, laid net, whipped
(cast), dunked (a kind of stationary fishing), and dove for fish.
Those who had boats fished either for subsistence or commercially one to
two miles from shore for ahi and aku in the spring and summer. They also
bottom fished at a depth of about 600 feet with hand lines. Some of the fish-
ing grounds south of Kaimu were affected by the lava flows. For example, moi
holes were destroyed when the lava altered the contour of the ocean floor.
Visibility was also affected by the flow. Only ulua fishing seemed to remain
unaffected.
In general, the fish caught were eaten raw, mixed with sea salt and limu, or
cooked in a variety of ways — steamed, fried, or grilled. Occasionally, when a
large amount of fish was caught, the fish would be dried for future use. Fish
caught in remote areas, such as in the national park, were quickly cleaned and
salted to dry to avoid spoilage during the long journey home without ice.
Ocean resources such as ‘opihi, wana, hä‘uke‘uke, and limu are also gath-
ered along the rocky shoreline of Puna. This is a hazardous undertaking. As a
rule, Native Hawaiians are taught never to turn their backs to the ocean. The
kua‘äina also throw back their first catch, for example an ‘opihi, as an offering
for protection from the waves.

184
puna: a WA H I PA NA sacred to pelehonuamea

Gathering and Hunting


Puna kua‘äina gathered plants for many purposes — food, medicine, tools,
building materials, art, and adornment. Fruits were gathered in season as they
ripened. Maile was gathered for special occasions such as birthday parties or
graduations. Increasingly, the maile was found only in the forest at higher ele-
vations, owing to commercial harvesting, development of subdivisions, and
continuing volcanic eruptions. Occasionally the Puna families gathered
resources for family members and close friends who moved to another part of
Hawai‘i or another island. Medicinal plants were once gathered throughout
Puna. However, over the years development and volcanic activity have limited
the growth of such plants to the forest reserve areas of the district. Those who
engaged in lä‘au lapa‘au were dependent upon a healthy forest to gather native
plants that still possessed the qualities and potency required for healing.
Puna kua‘äina hunted wild pigs, goats, and cows from within the national
park to Nänäwale. Some might go as often as every day or every other day in
order to provide meat for their ‘ohana. Everything caught was always shared
with the broader ‘ohana and older neighbors. Sharing the meat was tied to the
belief that if one was generous with the catch, then the supply would always
be there. Even if the catch was poor, the hunters still shared with others.
Greed was believed to be punishable by poor hunting or bad luck. When the
hunt was successful, the hunters generally thanked the ‘äina.
The hunting methods used varied from guns to knives and dogs. Larger
animals such as cows were shot and killed. Pigs and goats were usually chased
and cornered by dogs and then stabbed and killed by the hunter. Animals were
cleaned and dressed in the area where they were killed, and the meat was
packed out by the hunters. Most of the time hunters were careful not to kill
more than they could carry out.
The meat was usually smoked and cured. Most hunters had smokehouses
in their back yards and used wood gathered from the forest or lava flows to
smoke the meat.

Cultivation
Many kua‘äina cultivated plants for food and materials around their houses or
in tracts of land that required a long hike or travel by four-wheel drive. They
grew taro, sweet potato, banana, breadfruit, coconuts, kukui (candlenut tree),
papaya, lauhala, noni, ti leaves, and so on. The planting, harvesting, and pre-

185
chapter four

paring of taro was a family effort. ‘Ohana with taro would gather about every
three weeks to harvest taro, replant, and make poi. Individual families would
thus go home with a generous supply of fresh poi to last until the next gath-
ering. When ulu was abundant it was also cooked in the imu and pounded into
poi. Sweet potatoes were also cooked in the imu and sometimes mashed and
mixed with coconut milk for a dessert. Most families in Puna grew ti leaves in
their yards. Traditionally, ti leaves were used as a charm to ward off evil spir-
its at the site where the leaves grew or to protect the person who would wear
the charm. Ti leaves are also used to wrap fish and other food for steaming or
cooking over an open fire or in an imu. Kukui nuts are also easily and custom-
arily grown in yards. The nut is baked and used as a relish with Hawaiian salt
to prepare raw fish. If the land around their house was not suitable for cultiva-
tion, then plots were cultivated in family land with better soil. Some families
in Kalapana, for example, cultivated dry-land taro in Kamaili. Plots were even
cleared in the forest for planting taro, sweet potatoes, and bananas.

Regeneration
At the end of the twentieth century, the Native Hawaiian community of Puna,
particularly the lower part, remained distinct, geographically, culturally, and
socially. A significant part of the population is descended from the first fami-
lies who migrated there and settled in the district. They had a strong tradition
of perseverance in a district that has been constantly changing and evolving.
In addition, young Native Hawaiian families were moving in increasing
numbers into Puna from Hilo, Honolulu, and other neighboring islands.
Beginning in 1958, most moved into the nonstandard subdivisions, which
offered affordable homes for low- and moderate-income families. Yet despite
the increase in the population, the opening of new subdivisions, and contin-
uous eruptions by Pelehonuamea, Puna families still engaged in subsistence
activities.
Pelehonuamea continued to manifest her presence in the Puna district
through an active eruption that began on January 3, 1983, and continued into
the twenty-first century with earthquakes, natural subsidence, and the steady
flow of steam and natural gases out of the earth into the atmosphere.
When geothermal energy development for the generation of electricity
threatened to destroy the Puna Forest Reserve, the kua‘äina and Pele practi-
tioners rallied together to protect the natural and cultural resources of the for-

186
puna: a WA H I PA NA sacred to pelehonuamea

est they and their ancestors had always utilized and protected. This challenge
most clearly demonstrated the regenerative role of Puna as a cultural kïpuka.
Testimony provided by kua‘äina of Puna about their customary use of the for-
est convinced the circuit court judge and the judges of the Hawai‘i State
Supreme Court that Hawaiian cultural and subsistence beliefs, customs, and
practices continued to be actively practiced in the Puna Forest Reserve. This
resulted in a ruling of the Hawai‘i State Supreme Court that more broadly
defined the recognition of Native Hawaiian rights to access undeveloped pri-
vate and public lands for cultural, religious, and subsistence purposes.
The ahupua‘a of Kahauale‘a, owned by the Campbell Estate, was originally
targeted for geothermal energy development. When Pele began, on January
3, 1983, to continuously erupt at Kahauale‘a from mauka to makai the State of
Hawai‘i offered the Puna Forest Reserve for the development project. In 1983
the Pele practitioners formed an organization they called the Pele Defense
Fund. In 1985 they adopted a statement of the inherited beliefs that led them
to oppose geothermal energy.

Pele Perspectives

1 Pele is the heart, the life of the Hawaiian religious beliefs and practices
today.
2 Pele has always been and is today central and indispensable to Hawaiian
traditional religious beliefs and practices.
3 Nowhere in the geographical Pacific except Hawai‘i is there a recognized
volcano-nature God but Pele.
4 Pele is the akua, and ‘aumakua of Hawaiians today. Her blood relation-
ships continue as shared traditions, genealogy and aloha for particular
‘äina and places in Hawai‘i. Pele is küpuna and “tutu” to many Native
Hawaiians.
5 Pele is the inspiration, strength and focus for those who are established
in practices and performances of ancestral tradition and religion.
6 Pele influences daily spiritual and physical life activities, making it essen-
tial that Pele exist in pure form and environment.
7 Pele’s person, her body-spirit, her power-mana, her very existence are the
lands of Hawai‘i. This ‘äina is her, which she replenishes, nourishes, and
protects. She is seen in special-alternate body forms, along with those of
her sisters and brothers, their kino lau: the native fern, the native shrub,
the blossoms of the native trees.

187
chapter four

8 Pele is a living God. She is tangible. She has a home on Hawai‘i. She has
been seen by many living in Hawai‘i. She causes earth quakes, tidal waves
and lands to sink or surface from the ocean.
9 Pele is the magma, the heat, the vapor, the steam, and the cosmic cre-
ation which occur in volcanic eruptions. She is seen in the lava, images
of her standing erect, dancing, and extending her arms with her hair
flowing into the steam and clouds.
10 We know geothermal development will adversely affect and personally
injure the sacred body of the God Pele, and that she would retaliate. We
fear for the loss of our God, for the loss of the spirits of our ancestors,
for the loss of the lives of our children, and for the loss of our places in
Hawai‘i.
11 We believe that geothermal development will unduly burden those who
are the family of Pele, her guardians, her worshippers.
12 Geothermal development will severely impair those who depend on
salient images of Pele, her viability, and her forests which are connections
to the deity.
13 Geothermal development would impinge upon the continuation of all
essential ritual practices and therefore also impacts the ability of training
young persons in traditional religious beliefs and practices, and the ability
to convey these to future generations.
14 Geothermal development will take Pele and diminish and finally delete
her creative force, causing spiritual-religious, cultural, psychological and
sociological injury and damage to the people who worship and live with
Pele.

The Pele Defense Fund filed a suit to stop the exchange of the Puna Forest
Reserve for Kahauale‘a between the state of Hawai‘i and the Campbell Estate
(Pele Defense Fund v. Paty 79 Haw. at 442, 1992). Through the course of
the court case, the kua‘äina of Puna testified about their ongoing access to
the Puna Forest Reserve for the hunting and gathering of resources. They
explained the spiritual protocol followed out of respect for Pele and the mul-
titude of ancestral deities dwelling in the forest.
Though unable to reverse the land exchange, the Pele Defense Fund won
recognition of the rights of Native Hawaiians of Puna to access the Puna
Forest Reserve for traditional and customary practices even under the pri-
vate ownership of the Campbell Estate. The court case set a precedent for all
Native Hawaiian rights of access by ruling that “Native Hawaiian rights pro-

188
Figure 26 The Pele Defense Fund led a broad movement to stop the development of
geothermal energy to protect the Wao Kele o Puna lowland rainforest and protect the
sacred realm of the family of Pele deities. 1990. Franco Salmoiraghi.
chapter four

tected by Article XII. Section 7, may extend beyond the ahupua‘a in which a
Native Hawaiian resides where such rights have been customarily and tradi-
tionally exercised in this manner.” Prior to this ruling, the rights of Native
Hawaiians to access had been limited to the ahupua‘a in which they lived.
The Pele case expanded the recognition of all Native Hawaiian rights and
contributed to the regeneration of Native Hawaiian culture and religion
throughout the Hawaiian Islands into the twenty-first century.72 It also
reaffirmed the continuing existence and belief in Pelehonuamea as the inspi-
ration for new generations of Native Hawaiians from the rising of the sun at
Kumukahi, Puna to its setting at Lehua, beyond Kaua‘i.

190
 five 

Moloka‘i Nui a Hina:


Great Moloka‘i, Child of Hina

Ho‘i a‘e o Wäkea loa‘a Hina Then Wäkea turned around and
found Hina
Loa‘a Hina he wahine moe nä Wäkea Hina was found as a wife for Wäkea
Häpai Hina iä Moloka‘i, he moku Hina conceived Moloka‘i, an island
‘O Moloka‘i a Hina he keiki moku Hina’s Moloka‘i is an island child.
—paku‘i, in fornander, Fornander Collection, vol. 4 (1916–17)

Na Kuluwaiea o Haumea he käne, Kuluwaiea of Haumea as the


husband,
Na Hinanuialana he wahine Of Hinanuiakalana as the wife
Loa‘a Moloka‘i, ke akua, he kahuna Was born Moloka‘i, a God, a priest
He pualena no Nu‘umea The first morning light from
Nu‘umea.
—kahakuikamoana, in fornander, Fornander Collection, vol. 4
(1916–1917)

hants such as these in the epigraph composed by Paku‘i and Kaha-

C kuikamoana, which describe the conception and birth of Moloka‘i by


the Goddess Hina, are sources of the saying Moloka‘i Nui a Hina
(Great Moloka‘i, child of Hina). They convey the image of Moloka‘i as a
child—small and fragile—that needs to be nurtured by the people who live
there. Moloka‘i, smaller than Hawai‘i, Maui, O‘ahu, and Kaua‘i, has finite
resources that must be cultivated and sustained. The kuaäina of Moloka‘i trace
their roots back to antiquity and the traditional responsibility they inherited
to look after and care for the island and its resources.
Another important tradition of Moloka‘i is summed up in the epithet
“Moloka‘i Pule O‘o” (Moloka‘i of the Powerful Prayer). The mo‘olelo of the
‘ohana Kame‘ekua of Moloka‘i, descended from the Kai‘akea family of the

191
chapter five

Mo‘o clan, who they claim dates back to at least 800 bce on Moloka‘i, is pub-
lished as Tales from the Night Rainbow. It recounts how, around 1250 ce, the
high priest Pa‘ao had gone back to his homeland in Tahiti to gather warriors
to take over Hawai‘i. When the Tahitian warriors attempted to invade Molo-
ka‘i, the people of the island stood along the shoreline like a silent army. As the
warriors attempted to beach their canoes, the people of Moloka‘i began to
chant, starting softly until the chant grew into a mighty roar. Spears thrown by
the invading warriors fell short. Men trying to go ashore fell back into the surf
choking, unable to breathe.1 The invasion failed. As the tale of their defeat
spread, the island came to be known as Moloka‘i Pule O‘o.
This saying is also rooted in a tradition that upholds Moloka‘i as the train-
ing center of the most powerful kahuna or priests in sorcery in all of Hawai‘i.
One of the most famous of the powerful Moloka‘i kahuna was the prophet
Lanikaula, who lived in the sixteenth century. He was renowned for his abil-
ity to foretell the future and to give advice. His burial place, in a grove of kukui
trees in East Moloka‘i, was revered as a sacred place from the 16th century
through the early twenty-first century.

Figure 27 This young Hawaiian woman walking along the shoreline of Kapuaiwa Grove in
Kalama‘ula is reminiscent of the saying Moloka‘i Nui a Hina (Great Moloka‘i, Child of Hina).
2004. Richard A. Cooke III.

192
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

The tradition of the Kalaipahoa Gods firmly established Moloka‘i as a cen-


ter of sorcery. According to the mo‘olelo, there were trees at Maunaloa, Molo-
ka‘i, into which Gods had entered. When humans attempted to cut the trees,
they died if the sap or even a small chip of the wood touched their skin. The
Gods instructed Käneiakama how to make offerings, approach the trees, and
chop them in order to carve images of them. Käneiakama managed to chop
three blocks of wood from the trees and carve them into images he called
Kalaipahoa, meaning carved with pähoa axes. The stumps and branches left
over from carving were thrown into the sea to protect humans from coming
into contact with them and dying. The Moloka‘i chiefs and kahuna who pos-
sessed the Kalaipahoa images became very powerful, famous, and feared
throughout the islands. Kahekili secured control of the images when he
became ruler of Moloka‘i. Before he died he gave a little piece of one of the
images to High Chief Kamehameha. After the Battle of Nu‘uanu, Kameha-
meha himself assumed control of these godly images.2
The saying “Moloka‘i, ‘Äina Momona” (Moloka‘i, Land of Plenty) honors
Moloka‘i as the land of “fat fish and kukui nut relish.” The fat fish are raised
in the numerous fishponds on the island. The “kukui nut relish,” used to flavor
the fish, refers to the lush and abundant resources of the island. These abun-
dant resources, along with Moloka‘i’s strategic location between Maui and
O‘ahu, made it an island that the chiefs of Maui and O‘ahu fought over, back
and forth, to control.
“Moloka‘i No Ka Heke” (Moloka‘i is the greatest, the foremost) is a famous
boast about the island of Moloka‘i. It is the traditional rejoinder to Maui’s
boast of “Maui No Ka ‘Oi” (Maui is the best). This saying reflects the pride
that Moloka‘i’s people have in their island home. At the end of the twentieth
century, this pride was reflected in the residents’ praising Moloka‘i as “the last
Hawaiian island.” In an impressive collaborative effort, the communities on
Moloka‘i combined efforts to designate the island a rural enterprise zone
under the U.S. Federal Department of Agriculture in order to attract respon-
sible investment to the island. The vision statement reaffirmed the Hawaiian
values that are embraced by all of the people on Moloka‘i, Hawaiian and non-
Hawaiian alike. It stated in part:

Moloka‘i is the last Hawaiian island. We who live here choose not to be
strangers in our own land. The values of aloha ‘äina and mälama ‘äina ( love
and care for the land) guide our stewardship of Moloka‘i’s natural resources,
which nourish our families both physically and spiritually. We live by our
kupuna’s (elders’) historic legacy of pule ‘o‘o (powerful prayer). We honor

193
chapter five

our island’s Hawaiian cultural heritage, no matter what our ethnicity, and
that culture is practiced in our everyday lives. Our true wealth is measured
by the extent of our generosity.3

The traditions of Moloka‘i are still very much a part of the lives of the peo-
ple who live on Moloka‘i and care for it as their home.

Ruling Chiefs of Moloka‘i


In the mo‘olelo of Moloka‘i, Kamauaua, a descendant of the Nanaulu line was
recognized in the thirteenth century as the first ali‘inui of the island.4 In the
fifteenth century Kahokuohua was one of the principal chiefs of Moloka‘i. He
was conquered by the chief of Hawai‘i Island, Kalaunuiohua, in his drive to
control all of the islands of Hawai‘i. When Kalaunuiohua was defeated on
O‘ahu, Moloka‘i was again ruled by its own chiefly line.
Later in the fifteenth century, Kihaapi‘ilani assumed the chieftainship of
Maui, Moloka‘i, Läna‘i, and Kaho‘olawe. For a period of time he lived in
Waialua and worked with the Moloka‘i chiefs to restore the fishpond walls
and to lay out a road in West Moloka‘i from ‘Ïloli to Mo‘omomi. The road-
way was lined with white shells and became known as Kealaakapüpü (pathway
of the shells).
In the seventeenth century Moloka‘i was ruled by Kalanipehu. His daugh-
ter married a chief of Puna who had moved to Moloka‘i and was closely related
to the ruling chiefs of Hawai‘i. At the end of the seventeenth and in the early
eighteenth centuries, internal conflicts among the Moloka‘i chiefs led differ-
ent sides to seek alliances from Maui and O‘ahu and resulted in the loss of
independent rule over the island.5
At the end of the first quarter of the eighteenth century, the chief Kuali‘i
of O‘ahu allied with the Moloka‘i chiefs of the districts from Kawela to Mo‘o-
momi against the chiefs of the windward valleys. Though victorious, Kuali‘i
left the Moloka‘i chief Paepae and his wife Manau in charge of the island,
subject to his overall rule. Upon Kuali‘i’s death, his son and successor, Kapi‘i-
ohokalani, sought direct control over Moloka‘i and invaded the island with a
large force. The Hawai‘i chief Alapa‘inui, who was on Maui at that time with
his army and fleet of war canoes, went to the aid of the Moloka‘i chiefs and
helped them defeat Kapi‘iohokalani.6 Rather than annexing Moloka‘i, Alapa‘i-
nui instated the Moloka‘i chiefs as rulers over Moloka‘i. He also accompanied
the Moloka‘i chiefs to work out the terms of peace with the O‘ahu chiefs.
On O‘ahu, Kapi‘iohokalani was succeeded by his son Kanahaokalani, who

194
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

lived for only one more year. Upon his death, Peleioholani, a younger son of
Kapi‘iohokalani, assumed the rule over O‘ahu. During his rule, he subjugated
the chiefs of the windward valleys of Moloka‘i through the course of several
expeditions. In the mid-eighteenth century the Moloka‘i chiefs killed a daugh-
ter of Peleioholani. He exacted revenge upon the island’s chiefs and people
and assumed direct rule over the island. Moloka‘i remained under the rule of
the O‘ahu chiefs until the chief Kahekili of Maui conquered O‘ahu in 1785.
Leaving Maui under the rule of his son Kalanikupule, Kahekili moved to
O‘ahu to consolidate his rule.
In 1790, following his victory over Kalanikupule in the Battle of Kepani-
wai in Iao Valley, Kamehameha moved on to Moloka‘i to prepare his invasion
of O‘ahu and battle with Kahekili. After Kamehameha had killed High Chief
Kiwala‘ö in the Battle of Moku‘ohai, the dead chief's mother, High Chiefess
Kalola, his widow and sister High Chiefess Keku‘iapoiwa, and his daughter
High Chiefess Ke‘öpüolani took refuge on Maui with High Chief Kahekili,
the brother of High Chiefess Kalola. When Kamehameha invaded Maui,
Kalola fled to Moloka‘i with her daughter and granddaughter. While making
battle preparations on Moloka‘i, High Chief Kamehameha met with High
Chiefess Kalola and asked to marry and be entrusted with the protection of
the young chiefess Ke‘öpüolani. She agreed to have him care for Ke‘öpüolani
but only after her own death. She died several days later. Rather than invade
O‘ahu at this time, Kamehameha returned to Hawai‘i to stop the abuse of his
people and destruction of his lands in Waipi‘o, Hämäkua, Waimea, and
Kohala by High Chief Keouakü‘ahu‘ula of Ka‘ü. Kamehameha took Liliha
and Ke‘öpüolani with him to Hawai‘i and eventually married Ke‘öpüolani,
who bore his successors, Liholiho Kamehameha II and Kauikeaouli Kameha-
meha III.
A year later, Ka‘eokulani, high chief of Kaua‘i, joined High Chief Kahekili
to pursue and invade High Chief Kamehameha on Hawai‘i at Waipi‘o Valley.
On the way to Hawai‘i, these chiefs and their armies landed on Moloka‘i and
reclaimed the rule of the Maui chiefs over the island.
Upon their defeat in the Battle of Kepüwaha‘ula at Waipi‘o, High Chief
Kahekili returned to O‘ahu and left Chief Ka‘eokulani of Kaua‘i in charge of
Maui. When High Chief Kahekili died in 1794, his son Kalanikupule was des-
ignated as his successor and ruler of Maui, Läna‘i, Moloka‘i, and O‘ahu,
although High Chief Ka‘eokulani remained in charge of Maui. Conflicts
emerged and grew between High Chief Kalanikupule and High Chief Ka‘eo-
kulani. When High Chief Ka‘eokulani decided to return to Kaua‘i, he stopped

195
chapter five

at Moloka‘i to collect tribute, replenish his supplies, and take back the Kaua‘i
men who had been stationed there earlier.
As High Chief Ka‘eokulani proceeded back to Kaua‘i, he prepared to
engage High Chief Kalanikupule in battle when he landed on O‘ahu. The war-
riors of Kalanikupule fought against Ka‘eokulani and his men off of Waima-
nalo, all along the Ko‘olau coast of O‘ahu and over to Waialua. The final bat-
tle was fought at Ponahawale in ‘Ewa. High Chief Ka‘eokulani was killed, and
High Chief Kalanikupule became the sole ruler over Maui, Moloka‘i, Läna‘i,
Kaho‘olawe, and O‘ahu.
In 1795 High Chief Kamehameha, having secured his rule over Hawai‘i,
launched his campaign of conquest over the chiefs of the other islands. He first
invaded and conquered Maui, then moved on to conquer Moloka‘i, and from
there he invaded and conquered O‘ahu. Beginning in 1795 Moloka‘i was ruled
by the central government established by Kamehameha as the first Mö‘ï or
paramount chief and king of the Hawaiian Islands until the overthrow of the
Hawaiian monarchy in 1893.
At contact in 1779, Captain James Cook estimated the population of
Moloka‘i to be 36,000, while his sailing master William Bligh estimated it at
20,000.7 Kenneth Emory calculated the contact population at 10,500.8
Although the traditional chiefs of Maui and O‘ahu valued Moloka‘i for its
bountiful fishponds, verdant fields of taro, and strategic location, Western
trading vessels and whaling ships bypassed Moloka‘i. They considered it a
barren land with a sparse population, lacking adequate protected harbors or
anchorages, fresh water, and provisions.9 The limited freshwater resources
and lack of harbors that made Moloka‘i unsuitable for trade and agribusiness
enterprises contributed to the perpetuation of traditional farming and fishing
subsistence activities on the island throughout the nineteenth century and well
into the twentieth. In light of these circumstances, the pace of cultural change
due to Western influence during the nineteenth century was slower on Molo-
ka‘i than on the major Hawaiian islands. Throughout this period the Moloka‘i
Hawaiians maintained the traditional customs that complemented their tradi-
tional livelihoods.

Moloka‘i in the Nineteenth Century

During the nineteenth century, the major impact of Western contact on


Moloka‘i was the decline of the population due to diseases and emigration. In
1804 the entire population of Hawai‘i was infected by ma‘i oku‘u (either chol-

196
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

era or bubonic plague), and many died. David Malo estimated that through-
out Hawai‘i, one-half of the population succumbed to the disease.10 Given
Moloka‘i’s isolation from the port towns where the disease was introduced,
the kua‘äina of Moloka‘i may not have been as widely exposed. However, the
first missionary stationed on Moloka‘i in 1832 estimated the population to be
8,000, and the Missionary Herald, which provided a more detailed survey, esti-
mated the population at 8,700.11 Assessment of the severity of the impact
upon the Native Hawaiians of Moloka‘i of introduced continental diseases,
and the ma‘i oku‘u epidemic in particular, depends upon which estimate of
the population at contact one uses. If the estimate of Cook or Bligh is used,
then the impact of introduced diseases was extraordinary and tragic. If the
Emory estimate is used, then the decrease in the population may be attributed
more to emigration from Moloka‘i to Maui and O‘ahu, where commercial
activities and opportunities associated with the prosperous whaling industry
attracted many young Native Hawaiians from rural areas.12
In the fall of 1848 a measles epidemic killed one in every ten people on
Moloka‘i.13 Moloka‘i’s population also declined as many were attracted to
the centers of Western commercial activity at Lahaina and Honolulu. The
1849 census counted 3,429 persons on Moloka‘i.14 Those who survived and
remained behind continued to cultivate taro and sweet potatoes and to gather
fish, shellfish, and other aquatic foods from the ocean and mountain streams.
In 1845 King Kamehameha III and the Council of Chiefs announced their
intention to initiate a series of changes, including the introduction of a sys-
tem of private property, the naturalization of foreigners, the appointment of
foreigners to government positions and the imposition of taxes. On Moloka‘i,
1,344 residents signed a petition organized to oppose these changes:

Greetings Honorable King of our ancestors from the time of the Gods (pö)
down to us the descendants, as well as to the Kuhina nui of our Hawaiian
Kingdom and all the ali‘i of you entire nation.
The following is what we desire to request of you, our King, and our
ali‘i under you in the legislature.
1 For the independence of your nation, King [Kamehameha] III, we do
not want the haole you have appointed over the Hawaiian government
to serve as officials.
2 We do not want haole to be made naturalized citizens.
3 We do not want you to sell any portion of your nation to haole.
4 Do not place confusing taxes upon your humble people [huna lepo—bits
of earth].

197
chapter five

May these feelings of ours be shown to you, Your Majesty, and to our ali‘i.
We sign our names.
The following is the total amount of names 1344
Aloha honorable one who has been appointed to the root of King Kameha-
meha II and King Kamehameha I.
Your humble servants, the commoners of your islands,
Given by Keaumaea, July 1845 15

Through the course of the Mähele and the Kuleana Act, 636 awards for a
total of 2,332 acres were awarded to the maka‘äinana of Moloka‘i; ten awards
were granted to six ali‘i, fifteen awards were granted to fifteen konohiki, and
four awards were granted to four foreigners. Three ahupua‘a on Moloka‘i
became Crown land, and thirty-six ahupua‘a were designated as government
land.
Prominent among the ali‘i who received land awards on Moloka‘i was
Kekauonohi, who is described above as a recipient of land awards in Häna and
Puna. She received four times the amount of land distributed to all of the
maka‘äinana combined, a total of 10,341 acres in Kapualei (1,670 acres),
Kumu‘eli (1,607 acres), Moakea (1,092 acres), and Naiwa (5,909 acres). Wil-
liam C. Lunalilo, who also received lands in Häna and Puna, was awarded
1,168 acres in Waialua and 14,787 acres in Kawela. William Pitt Leleiohoku,
who also received land awards in Waipi‘o and Puna, as discussed above, was
awarded 3,921 acres in Kamalö. Julia Alapai Kauwa, who also received land in
Häna, received ahupua‘a part 5 in Honomuni. Victoria Kamämalu, who was
discussed above as having received lands in Puna, received an ahupua‘a in
Hälawa. Enoka Kuakamauna, who was a konohiki for Hoapilikäne and Hoa-
piliwahine on Moloka‘i, received 401 acres in Keopuka Uuku; 168 acres in
Ahaino 2, and 72 acres in Wailau.
According to the 1850 census, 3,540 persons lived on Moloka‘i. In 1845,
1,344 Native Hawaiians had petitioned against the sale of land, yet only 636
awards were granted. The petition indicates that at least 1,344 persons on
Moloka‘i knew of the proposed changes to the land system. Did the majority
of these petitioners (708) decide to boycott the process and not submit an
application? Did those among the petitioners who did not apply succumb to
illness or disease? Were half of the 1,344 persons from the same household so
that only one person would have submitted an application for an award? Per-
haps all of these factors combined to help to account for the low number of
applicants for lands on Moloka‘i. An analysis of this petition, which is pre-

198
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

sented in Appendix III, shows that 268 names on the petition match the
names of persons who received Land Commission awards.
The pattern of land awards on Moloka‘i reflects a concentration of Native
Hawaiian families in Mana‘e or east Moloka‘i, from Kalama‘ula, along the
south and southeast section of the island up to the boundary with Hälawa Val-
ley. Of the Moloka‘i Land Commission awards to maka‘äinana, 69.7 percent
or 443 of the awards were located in east Moloka‘i, which represented 77 per-
cent of the land awarded or 1,791 acres. The 1853 population map for Molo-
ka‘i developed by John Coulter reflects the same pattern of concentration of
the population in the ahupua‘a of East Moloka‘i.
The second area of significant concentration of Native Hawaiians on Molo-
ka‘i was the windward valleys of north Moloka‘i, from Hälawa and over to the
Kalaupapa peninsula. Native Hawaiians living in the windward valleys received
29.7 percent or 189 of the Moloka‘i land awards, which represented 22 per-
cent of the lands awarded or 522 acres.
No awards were given to maka‘äinana in west Moloka‘i, and only four
awards were given to maka‘äinana in central Moloka‘i, for a total of 19.23
acres.
Traditional economic activities were pursued on these kuleana lands pri-
marily at a subsistence level for household consumption and exchange with
extended family members and neighbors. During this period, east Moloka‘i,
or Mana‘e, sustained the bulk of the island’s population with its fertile lands
and numerous fishponds. Kanepu‘u, a Hawaiian writer for the newspaper Ke
Au Oko‘a, toured Moloka‘i in 1867 and praised Mana‘e as having good lands
and providing the people who lived there with a pleasant life. His detailed

Map 4 The Coulter 1853 map shows a population of 3,540 for Moloka‘i, with 2,700 persons
living in Mana‘e. o = 20 persons. Source: Coulter, Population and Utilization of Land and Sea, p. 21.

199
chapter five

account of the east end included descriptions of the rich resources of Mana‘e
Moloka‘i for subsistence economic activities:

Moakea to Honouli—Good place to live and grow sweet potatoes and dry
land taro.
Honoulimaloo—dry . . . only sweet potato and dry land taro grows, not
enough water for wet land taro.
Honouluwai—many taro patches from lowlands to uplands . . . life there
is pleasant.
Kumimi and Moanui—dry and dreary plain cut by ridges. Some taro
patches on side adjoining Honouliwai. Some taro patches at Moanui.
Waialua—taro patches on every side stretching from shore to inland.
Good place to live. Many hala trees along the banks of the streams. A large
stream, but not as large as Moaula.
Poniohua and Puelelu—kula lands that are irregular on the mountain-
ward side. Wet patch taro grows in the upland while dry land taro and
sweet potato thrive in the lowland. At this place begins the patches in
which springs bubble up. Poniohua has a taro patch that had been dug up
at large mounds in the center where bananas, sugar cane, sweet potatoes
. . . and so on had been planted while taro grew below in the water.
Kainalu to Ahaino-iki—whole district fertile . . . plants grow well.
Kailiula, Ahaino nui—uala and kalo . . . fishponds . . . fertile soil.
Kupeke and Pukoo—fertile, but dry and hot.
Punaula and Mapulehu—good land. Mapulehu has a wide plain and
large valley.
Ualapue—a good land, filled with taro patches and a pond.16

At Hälawa and Waialua, surpluses of taro and fish were steadily produced
and sold to markets at Lahaina and Honolulu. George Bates, who visited
Hälawa Valley in 1853, described the beauty of Hälawa and the commercial
production of taro which he observed:

The valley of Hälawa . . . is the finest scene on Molokai. The traveler


stumbles on its brink unawares. At a depth of nearly twenty-five hundred
feet below him, the whole scene is spread out before him . . . scores of taro
beds, and a number of dwellings, and the romantic river are all seen in a
single glance; . . .
The cultivation of taro is carried on here on a large scale. It is raised
chiefly to supply the Lahaina market. I was informed by Mr. Dwight [a

200
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

missionary on the island] at Kalua‘aha, that the entire amount raised for
sale and home consumption was valued at $15,000 to $20,000. The valley
of Hälawa is the richest spot on the island.17

Thus, where it was possible, Moloka‘i Hawaiians had an active interest in


earning cash to acquire a variety of introduced material goods. Moreover,
political and economic changes under the constitutional monarchy required
payment in cash of property, income, and poll taxes. Additional economic
activities to earn cash included the gathering of kukui nuts for oil (through
1858); of pepeiaoakua, a fungus considered a delicacy by the Chinese, for export
to San Francisco and China; and of pulu, the silky fiber of a native fern found
in Moloka‘i forests that was used to stuff pillows and mattresses. Some Molo-
ka‘i Hawaiians also participated in short-lived, small-scale enterprises. During
the 1849 Gold Rush, some Hawaiians cultivated potatoes for export to Cali-
fornia. During the Civil War, cotton was grown and exported from Moloka‘i.
The mission station grew grapes. Coffee was cultivated at Kala‘e. Rice was
grown in the windward valleys. Hawaiians experimented with the cultivation
of tobacco. An attempt to grow sugar on a large scale began in 1872.18
The major enterprise established on the island on a permanent basis was
the ranching of cattle, sheep, and goats on a large scale by Kamehameha V as
well as by small farmers. Dairies were also set up and produced butter and
milk. Ranching was the major “industry” on Moloka‘i until the establishment
of pineapple plantations beginning in 1923.19
Indicative of both the decline of the population and the minimal influence
of foreigners during the nineteenth century is the 1896 census figure, which
showed that there were 2,132 Hawaiians living on Moloka‘i and only 175
non-Hawaiians.20

The Establishment of Kalawao and Kalaupapa


The history of Kalawao County, including Kalaupapa on Moloka‘i, was dis-
tinct from that of the rest of Moloka‘i. In January 1865 the legislature passed
“An Act to Prevent the Spread of Hansen’s disease,” which gave the kingdom’s
board of health the authority to segregate and isolate persons who contracted
Hansen’s disease (leprosy).21 In September 1865 the board of health acquired
over 700 acres of land in Waikolu and Wai‘ale‘ia valleys for the express pur-
pose of isolating victims of Hansen’s disease. Later, in 1865, they also acquired
Kalawao. These lands were selected for their geographic features. Access to

201
chapter five

them by land and by ocean was difficult, and they were self-contained areas
that had previously provided fifteen to twenty families with sufficient resources
for their subsistence. According to the 1866 board of health report:

The tract was extremely well situated for the purpose designed. It is diffi-
cult of access from the sea; has no roads passing through it into other
districts; is supplied with water by two running streams; has a large area
of kalo land; enjoys the advantage of the constant trade wind; has ample
grazing lands; and possesses a soil capable of raising vegetables of all
different kinds adapted to these islands in the greatest abundance.22

Hawaiians who lived in these valleys were given a settlement of $1,800 and
government lands in east Moloka‘i in exchange for their land and homes. Sev-
eral hundred people relocated to Kainalu and Waialua on the southeastern
coast of Moloka‘i.
As the number of persons with Hansen’s disease increased, the board of
health continued to expand the isolation area. In 1866 the board acquired
Makanalua Valley.
Moloka‘i became famous as the “Lonely Isle” because of the Hansen’s dis-
ease settlement at Kalawao-Kalaupapa. However, the settlement operated as a
world unto itself, quite separate from the rest of the Moloka‘i residents. There
was little contact between the settlement and the rest of the island, with the
exception of the windward valleys of Hälawa, Wailau, and Pelekunu, whose
residents provided the settlement with pa‘i‘ai and seasonally gathered salt at
the peninsula.23
The following lament chant conveys the loneliness and grief of those
Hawaiians who contracted Hansen’s disease and were rounded up like animals
and banished to Kalawao and Kalaupapa. It was composed by Ka‘ehu, a
renowned hula master of Kaua‘i who contracted Hansen’s disease and was
sent to Kalawao, where he died.

Lohe ana kauka aupuni Report reached the government doctors


Ho‘ouna ke koa maka‘i Who sent military soldiers
Hopuhopu ‘ia mai kohu moa Caught were we like chickens.
Alaka‘i i ke ala kohu pipi Led along the road like cattle.
Kü ana imua o ka Papa Ola We confronted the Board of Health.
Papa ola ‘ole o nei ma‘i A board promising no cure for this disease.
Ki‘ei wale mai na kauka The doctors just peered at us
Halo ma‘ö ma‘ane‘i Peering this way and that

202
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

Kuhi a‘e nä lima i Lë‘ahi Fingers pointed toward Lë‘ahi


“Hele ‘oe ma Kalawao” “You go to Kalawao”
Lälau nä koa Aupuni Military soldiers seized us
Halihali ia kai kauwapo Fetched us to the wharf
Ho‘ili nä pio a pau All prisoners were sent aboard
Ka luahi ia a ka ma‘i lëpele Victims of leprosy
Hiki ke aloha kaumaha no Great grief and sadness possessed us,
I ka ‘ike ‘ole i ka ‘ohana For we had not seen our families
Ka waimaka ho‘i ka ‘elo‘elo Tears poured like raindrops
Ho‘opulu i ka papalina Wetting the cheeks 24

In 1873, after 600 people had been banished to Kalawao, the government
purchased the Kalaupapa Peninsula, excluding the eastern portion. Despite
the expansion of the settlement, several people continued to live on their
kuleana lands on the peninsula. While they lived there, the kuleana holders
provided hiding places, food, and lodging for the healthy friends and relatives
of the Hansen’s disease exiles.25 In 1895 the Board of Health claimed these
kuleana through condemnation proceedings and evicted the holders for health
reasons.
Although the board of health had expected the Hansen’s disease victims to
fish and farm for their day-to-day sustenance, the board realized within the
first year that the exiles were too ill, demoralized, and debilitated to provide
for their own needs. Beginning in 1866 the legislature appropriated monies
to purchase food and supplies. In 1868 a hospital building, a schoolhouse, and
quarters for the young boys and young girls were constructed by the king-
dom. By 1872 more homes had been built, and the weekly rations included
five pounds of meat and twenty-one pounds of pa‘i‘ai, most of the latter pur-
chased from Hälawa, Wailau, and Pelekunu. Nevertheless, conditions in the
settlement were miserable. The Hansen’s disease victims complained in let-
ters to relatives about the lack of health care, the separation of husbands from
wives, poor and insufficient food, the scanty supply of clothes, the difficulty
they had obtaining rations when they were ill, and complete lawlessness.26 In
response to the complaints of friends and relatives of the victims of Hansen’s
disease, King Lunalilo instituted a number of reforms during his brief one-
year reign in 1874. The same year Father Damien began to live and work
among the suffering lepers of Kalawao.
From 1874 through 1889, through the efforts of Father Damien and his
supporters, conditions gradually improved. In November 1888 the Sisters of

203
chapter five

Charity, led by Mother Marianne, also began to tend and care for the Han-
sen’s disease victims at Kalawao.27 By 1900, when Hawai‘i became a territory,
vast improvements had been completed. By this time most of the exiles lived
on the Kalaupapa Peninsula, where the climate was drier and the dock acces-
sible. Access to clean fresh water, a constant problem, had become available
with the extension of water pipes to Kalaupapa. Life for the Hansen’s disease
patients at the settlement was bittersweet during the early years of the terri-
torial period.

Moloka‘i in the Territorial Period


During the territorial period, the population of Moloka‘i continued to decline
until the Hawaiian Homesteading Program started in 1922 and pineapple
plantations were established in 1923. In 1900, not counting Kalawao the pop-
ulation of Moloka‘i was 1,327. The population of Kalawao was comparable at
1,177. In 1910 the population of Moloka‘i, again excepting Kalawao, hit its
lowest point at 1,006, while the population of Kalawao dropped to 785. The
1920 population figure outside of Kalawao rose by only eleven, to 1,117.28 It
is estimated that up to this point 96 percent of the population of Moloka‘i out-
side of Kalawao was concentrated in the area east of Kamalö and the wind-
ward valleys.29 During this period population declined primarily because of
the emigration of people to other islands, particularly O‘ahu, to seek jobs and
gain access to material goods not available on Moloka‘i.
The Hawaiian Homes Program and the pineapple industry attracted peo-
ple to Moloka‘i beginning in 1922. By the time of the 1930 census, there were
4,427 people living on Moloka‘i. This represented an increase of 148 percent
over the 1920 figure. Of the total population, 1,869, or 42 percent, were
Hawaiian or part Hawaiian. The east end of Moloka‘i continued to be an area
of concentration. Of 971 inhabitants of the southeast coast and the windward
valleys, 566, or 58 percent, were Hawaiian or part Hawaiian. Owing to the
establishment of the Hawaiian Homesteading Program, Pala‘au-Ho‘olehua
became another area where Hawaiians concentrated. Out of its 1,031 inhabi-
tants, 826, or 80 percent, were Hawaiian.30
Cattle ranching was the major industry on Moloka‘i until 1923, when pine-
apple began to be grown on a large scale on land leased from Moloka‘i Ranch.
Although Moloka‘i Ranch was the major operation, there were also smaller
ranches along the southern coast. At the west end, Moloka‘i Ranch developed
a paddock system for cattle, gradually fencing in the open range. They hunted

204
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

down the wild herds of deer and goats; started control measures against guava,
lantana, and pämakani; introduced new grasses; and started a breeding pro-
gram using Devon bulls.
In 1904 the ranch had 5,598 head of cattle, 13,918 sheep, 298 horses, 272
pigs, and 1,614 colonies of bees. Then, in 1917, the ranch decided to gradu-
ally abandon sheep herding in favor of cattle raising because beef earned a bet-
ter return than sheep and because sheep diseases were a perennial problem.
Table 7 compares the raising of sheep and cattle on Moloka‘i from 1900
through 1930.
Moloka‘i Ranch also started apiaries on its land in 1901. In 1904 the first
harvest of honey was sold commercially. Between 1904 and 1909 Moloka‘i
honey was sold to the United States and Australia. From 1909 to World War
I most of the honey was marketed to Germany through H. Hackfeld and
Company. When the German market closed during World War I, Moloka‘i
began to export its honey to California. Moloka‘i was the largest producer of
honey in the world in 1919, when the ranch had 2,250 colonies of bees pro-
ducing 2,946 cases of honey and 80 cases of beeswax. It grossed as much as
$21,000 in one year alone. In 1937 the industry folded when all of the bees
became infected with American foul brood.31
The ranch employed males from ten to twenty years of age, according to
a Native Hawaiian informant, Albert Kahinu. He lived on Moloka‘i until 1902,
then moved to Honolulu, returning to Moloka‘i at age 18 in 1912. To him,
the work was hard, but the cost of living was reasonable. The highest wage
was about $30 a month, and the employees received subsistence supplies twice
a month. Moreover, there was an abundance of fish to supplement the income
from the ranches. According to Kahinu, “Living at that time was very friendly
. . . all the fish you want . . . you can get in the sea, you can kick the fish with
your feet, in those days. Ka i‘a ka wäwae o Hïlia—The fish that can be kicked
at Hïlia.” 32
It was attempted to raise sisal on barren spots on the southeast shore, but

table 7 head of sheep and cattle raised on moloka‘i, 1900–1930


year cattle sheep

1900 6,354 15,800


1910 6,213 7,915
1920 8,140 3,643
1930 7,623 225

205
chapter five

the venture failed owing to high overhead and the low-price sisal from Mex-
ico, a major producer of the fiber. At Puko‘o frogs were raised commercially
for export. The Duvauchelle family of Puko‘o hunted, killed, and processed
sharks for engine oil, which they sold to the Pioneer Mill at Lahaina.33 Puko‘o
was the social center for Mana‘e during the first three decades of the twenti-
eth century. The main post office was located there and had a wharf that ser-
viced the residents of the area.
Other minor industries included boat construction and guitar making. At
Hälawa, Wailau, and Mana‘e, hats woven out of native materials such as lau-
hala, makaloa, and coconut leaves were produced for sale. Lauhala mats were
also woven for household use and for sale.34
Charles Baldwin provided a good overview of commercial economic activ-
ities on Moloka‘i:

The larger part of the island is devoted to cattle raising. Taro is grown in
Pelekunu and Wailau for the leper settlement. Sisal is also grown in places.
Formerly there were a great many fish ponds within the barrier reef
along the southern shore of the island, but many of these ponds are not

Figure 28 Kûpuna Albert and Lani Kahinu were photographed by L. R. Sullivan in 1920 and
interviewed by Mary Kawena Pukui in 1961. Coincidentally, this couple raised a cousin of my
father’s whose mother died in childbirth. 1920. Sullivan Collection, Bishop Museum.

206
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

now, as there is no market for the fish, and the inclosing walls have been
allowed to fall to pieces. Some fish are sent to Lahaina and Honolulu.35

Despite the diversity of commercial enterprises that were pursued at east


Moloka‘i, subsistence farming of taro and sweet potatoes and fishing contin-
ued to be the mainstay of the 900 or so Hawaiians who lived along the south-
ern coast and in the windward valleys of Moloka‘i.
Up through World War II Moloka‘i was a marginal rural area with a small,
predominantly Native Hawaiian population. The population was concentrated
in kuleana lands in the eastern part of the island, the windward valleys, and
the Kalawao Hansen’s disease settlement. The establishment of the Hawaiian
Homes Program and the pineapple industry did not transform life for Hawai-
ians living on the eastern end of Moloka‘i. These economic enterprises did,
however, introduce new forces and influences into the central and western
parts of the island that had previously been sparsely populated and devoted to
ranching. By World War II Moloka‘i was an island with four distinct commu-
nities—the Native Hawaiian subsistence farmers and fishermen of east Molo-
ka‘i and the windward valleys, the Native Hawaiian homesteaders of Kalama-
‘ula and Ho‘olehua, the Hansen’s disease patients at Kalawao-Kalaupapa, and
the immigrant laborers at Kualapu‘u and Mauna Loa. The following sections
describe the distinctive character of each of the communities where Hawai-
ians were predominant.

Mana‘e: East Moloka‘i


East Moloka‘i includes the southern coast east of Kaunakakai as well as the
windward valleys on the north end of the island. However, it is the southeast
coast that is commonly referred to as Mana‘e. Mana‘e consists of a narrow
coastal plain of fertile alluvial lowlands and numerous freshwater springs. Gen-
tle slopes, gulches, and narrow valleys lead inland to the east Moloka‘i moun-
tains, where the peaks are usually shrouded by clouds and waterfalls flow down
sheer cliffs. The mountain rains percolate into the water lense and in some
places emerge along the coast and shore as springs. They also feed intermit-
tent streams that keep the valley recesses lush with native plants.
The coastline of Mana‘e is protected by a barrier reef that extends more
than one mile out to sea. This extensive reef makes the ocean calm along the
shore. The ocean is the backyard of most of the families living in east Molo-
ka‘i. Its proximity, calmness, and shallow depths provide easy access to the
ocean’s resources.

207
chapter five

Of Moloka‘i’s fifty-three fishponds, thirty-five fringe the southeastern


shore. J. N. Cobb listed eighteen fishponds as being in active use for commer-
cial purposes in his report The Commercial Fisheries of the Hawaiian Islands.
The fishponds ranged in size from 1 acre to 54.5 acres and totaled 293 acres
altogether.36 Within these ponds Mana‘e Hawaiians raised mullet, awa, aua,
käkü, äholehole, ‘o‘opu, eels, shrimp, squid, and crab.37 The mullet from the
‘Ualapu‘e, Nï‘aupala, Pipi‘o, Puko‘o, and Küpeke ponds were noted for their
fatness and were especially popular in Lahaina. In an account of Hawaiian
fishing lore by Kahaulelio that ran in Nüpepa Kü‘oko‘a from February through
July 1902, the author said that the Lahaina people ran and leaped and hurried
to get mullet from those particular Moloka‘i ponds, because to delay meant
returning empty-handed. He also provided instructions on how to catch the
fish from these ponds:

In fishing, go just before daylight, with nets having meshes of two finger’s
[sic] width in the hands of Hulu, laumana and others. The breezes blow
strongly and while still shivering with cold, plunge into the icy water. With
one or two drawings of the net, the boat is filled, the sails set with Lahaina
as the goal. At this time it is still dark and light comes when Kekaa is
reached or Lahaina itself. When you open the belly of the mullets of these
ponds, the fat within is like that of a hog and does resemble it in every
way.38

Forty ahupua‘a, averaging two square miles each, make up Mana‘e.39 Kua-
äina living in these ahupua‘a had access to ocean and mountain resources as a
part of their inherited kuleana rights, and they utilized the resources of their
region on a regular basis, hunting wild deer, goats, pigs, turkeys, ducks, and
pheasants and gathering fruits, nuts, vegetables, vines, materials for twine,
native plants for healing, and flowers for special occasions.
From the ocean, shellfish, seaweed, squid, lobsters, crabs, and other reef life
were gathered. The Mana‘e Hawaiians not only harvested the many fishponds
that fringed their shore, they also fished at greater depths by net, diving, and
trolling. In the valley streams they caught o‘opu and ‘öpae.
Mana‘e kuaäina also raised animals and crops on their kuleana land. Most
families raised domestic pigs, one or two head of cattle, and some chickens.
Almost every household owned a horse as its major means of transportation.
They cultivated the traditional Hawaiian staples of taro, sweet potato, and
breadfruit. They also planted onions and watercress and other introduced veg-
etables and fruits in their gardens.

208
Figure 29 A classic kua‘âina and farmer of Moloka‘i. Circa 1912. Ray Jerome Baker,
Hawaiian Historical Society.
chapter five

All these foods, both cultivated and gathered, provided the Mana‘e kua-
‘äina with a healthy daily diet. These foods were usually jointly gathered, cul-
tivated, and shared among the ‘ohana. Throughout Mana‘e several households
commonly lived together on the kuleana land of the ‘ohana. Food obtained
through hunting, gathering, or fishing was also given as gifts or exchanged for
other types of food or services with neighbors. Sometimes it was sold for cash
to buy additional provisions.40
In 1931 Handy surveyed Moloka‘i and wrote up his observations about east
Moloka‘i in volume 1 of The Hawaiian Planter. He described a method of
planting taro in mounds that was unique to east Moloka‘i. His account docu-
ments the extensive cultivation of taro and sweet potatoes by the people of
Mana‘e from 1900 through 1930:

Wet taro was seen at Keawanui, Puko‘o, Kawaikapu, Waialua, Honouliwai,


and Pohakupili. My short visit to Molokai in 1931 did not allow time for
any exploration. At Kamalo I observed and photographed a method of taro
cultivation said to have been common along this coast in ancient times. In
flat swampy ground earth is heaped up into long mounds 3 or 4 feet high
and about 3 feet broad on top, each mound surrounded by water left stand-
ing in the ditches created by digging out and heaping up the earth. The
taro is planted around the lower margins of the mounds near the water;
sweet potatoes are planted on top. This method of swamp-land planting
finds its counterpart in the old style of mounding (termed kipikipi) prac-
ticed in Waiakea, Hawaii.41

The kupuna informant Peter Namakaeha said that lehua, nohu, piko ke‘o-
ke‘o, piko ‘ele‘ele, and pi‘iali‘i taro varieties were planted at Kalama and
‘öhi‘a. Some was pounded into pa‘i‘ai for consumption by the ‘ohana, and
some was sold.42 Handy also wrote about his 1931 tour of south Moloka‘i: “In
1931, potato patches were seen at various places near the road along the south
coast, and Hawaiians said that many parts of the kula land used to be planted
with both sweet potato and dry taro. It is safe to assume that potatoes were
grown all along this coastal plain fringed with fishponds from Waialua to
Punakou.” 43
In 1961 Daniel Pahupu, a kupuna informant, shared his mana‘o about life
in Mana‘e with Mary Kawena Pukui of the Bishop Museum. “Before, loa‘a no
ke kalo, ai no ke i‘a alo la—Just get the taro, the meat is right there in front.
The land was not kapu, therefore we never lacked anything, but now, ea
hemahema—the spirit is lacking.” 44 Namakaeha, a respected fisherman from

210
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

Honouliwai, also summed it up nicely: “I have everything. What more I want?


No island beat Moloka‘i. We have everything . . . Whereas some islands you
don’t get everything. I live one of the best life.” 45 In short, east Moloka‘i pro-
vided Hawaiians with an abundance of natural resources which more than
adequately fulfilled their food needs.

The Windward Valleys


Windward Moloka‘i has seven major ahupua‘a, each averaging six square
miles.46 The valleys are separated from each other by high tablelands that
extend from the east Moloka‘i mountain range out to the coast, where they
drop dramatically down to the ocean. The walls of the valleys are steep and
ascend as high as 1,500 to 2,000 feet. Beaches, rocky and inaccessible by boat
except during the summer months, are found only at the mouths of the val-
leys.47 The uplands are usually cloaked in fog and saturated by rain. Sizable
streams flow year round in each valley. The narrow valley bottoms were set-
tled by early Hawaiians who planted taro in terraces.
In 1916 the Bishop Museum ethnographer Kenneth Emory toured the
windward side of Moloka‘i and observed

cliffs, several thousand feet high; promontories, bold and razor-edged; deep
gorges rushing down to the sea, all in the wildest confusion, and covered
with a beautiful mantle of variegated green, except where the cliffs were
sprayed with salty foam. And one of the added charms of this was the beau-
tiful waterfalls that waved their glossy threads down the face of the cliffs.
Waterfalls, peaceful vales, lagoons hidden under dark caverns, tropical
birds floating above, vines swaying in the wind, every form and color of
beauty lay revealed in the grand precipice above us, filling half the space
between horizon and zenith.48

The windward valleys generously furnished their inhabitants with water to


irrigate taro terraces; edible fruits and root crops; plants for healing; fibers for
cordage; freshwater o‘opu fish, ‘öpae, wï, and hïhïwai; and an ocean teeming
with fish, seaweed, and shellfish. Each of the windward valleys abounded in
native plants that provided for the day-to-day needs of the kua‘äina. ‘Awa usu-
ally grew in the wet uplands. Breadfruit, sugar cane, arrowroot, and bananas
were raised on the borders of the taro patches and the sloping ground. Gourds
for containers grew in warm, damp spots near the sea. Olonä provided fiber
for lines and nets, and the wauke for tapa grew in the uplands. Yams grew in

211
chapter five

the lower forest zone. These cultivated plants were complemented by numer-
ous forest plants that supplemented the diets of the Hawaiians of the wind-
ward valleys, as well as flowers for adornment and decoration.49
Pelekunu, Wailau, and Hälawa are the valleys about which accounts are
available. The descriptions of the lifestyle and livelihoods of the families in
those valleys shed light on the living conditions of kua‘äina along Moloka‘i’s
windward shore during the first three decades of the twentieth century.
According to the Native Hawaiian informant Daniel Napela Naki, who was
born in Wailau, raising taro and fishing were the primary enterprises of the
families who lived in Wailau, Pelekunu, and Hälawa. These activities were
pursued not merely to meet the subsistence needs of the local families. Poi
and fish from these valleys were also sold to the settlement at Kalaupapa and
the urban center of Lahaina. Fred Tollefson, a kupuna of Norwegian descent
who learned to speak fluent Hawaiian growing up on Moloka‘i, recalled how
the pa‘i‘ai, or pounded taro, from Wailau, Pelekunu, and Hälawa was wrapped
in ti leaves and the bundles floated in to the shores of Kalaupapa from off-
shore boats.
James Poaha described the interrelationship between the windward valleys
and the Kona or leeward coastal plain as similar to the mauka-makai pattern
of exchange that was traditionally practiced among the native Hawaiians:

A fishing place . . . that was the livelihood of those who lived here before
[leeward side]. The people who lived on the leeward side, and those who
lived on the wet /water lands would bring pa‘i‘ai and take fish from here.
From Hälawa, Pelekunu, and Wailau.
Poi came from those places—as pa‘i‘ai wrapped in ti leaf. The outer
cord to bind was made of sisal. One bundle was 40 pounds. The monetary
value was only 75 cents. When those bundles were brought over, each one
got out his poi board. When they carried the pa‘i‘ai from Wailau it was the
back which bore the load. The food was carried on the backs. Such great
patience they had . . . Much patience.50

In addition to exchanging fish, the residents on the kona side produced salt
and exchanged it with the people from the windward side.

Pelekunu
In 1916 Kenneth Emory’s group toured windward Moloka‘i and landed in
Pelekunu Valley. Emory provided Mid-Pacific Magazine with a rare description

212
Figure 30 Kûpuna James and Mary Poaha were photographed with their children,
James Jr., Ambrose, Agnes, Eva, and Margaret, by L. R. Sullivan in 1920. The couple was
interviewed by Mary Kawena Pukui in 1961. 1920. Sullivan Collection, Bishop Musuem.
chapter five

of life in that remote windward valley. There were nine rather dilapidated,
old-fashioned, and crudely built houses scattered along the crescent-shaped
bay. Along the front of the village and a little back from the beach was an
ancient stone wall. To the rear of the village and squeezed into the narrow
valley far back were rudely cultivated taro patches that were almost neglected.
There were no roads, stores, or shops—only houses and people with the bare
comforts of civilization, such as kerosene lamps, tin and iron ware, matches,
and so on. Emory was impressed with the hospitality of the people: “Though
these people had just enough to keep them alive, everything of their’s [sic] that
we could use was our’s [sic] while we stayed there. The little school house was
our’s for the night, and they brought us poi, dried fish, and felt hurt when we
most emphatically refused one of their few hogs, the most highly prized of all
their possessions.” 51
Harriet Ne, a kupuna who had spent the first six years of her life, from
1915 to 1921, with the last eight families to inhabit Pelekunu Valley, was
interviewed in 1979. She provided vivid glimpses into the lifestyle and liveli-
hoods of the people there. According to her, everyone usually got up at dawn
to make maximum use of the sunlight. The valley was so narrow that the sun
did not shine very long.52
Poi was the staple food of the Pelekunu families. The production of this
basic necessity, according to Ne, was a collective enterprise of each family in
the valley. The eight families who shared the resources of Pelekunu Valley
were closely bound in their day-to-day lives. For example, every family needed
poi on a daily basis, and the families rotated the responsibility for producing
the weekly supply of poi for the whole valley. According to Ne, the men would
get up early, cook breakfast, and go to the taro patches. The children ate,
cleared the breakfast table, and then followed the fathers to the patch, where
they would tie the pulled taro into bundles, carry the bundles to the stream
to be washed, and then take them to the community shed where the taro was
to be cooked. After cooking, the taro was cooled on a big net, and everyone
gathered around to peel off the skin. The clean taro was placed on the com-
munity poi board, which was seven feet long and about two and a half feet
wide. Men on both ends and sometimes in the middle would pound the taro
into poi. When finished, the poi was given out to each family. They received
enough to last them the whole week, when the next family would take over the
work of pounding poi.53
Pelekunu was rich in natural foods, and gathering plants and shellfish was
a basic feature of life in the valley. Ne explained that the name of the valley

214
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

refers to the smell of freshwater shellfish that dries up and dies on stream
rocks:

Hihiwai is a beautiful shellfish, and it’s very tasty and it lives under the
great big rocks in the fresh water streams. But early in the morning, they
all come up, crawling up to the top of the rock to breath [sic] the sunlight
and it takes them hours to get to the top of the rock. By the time the sun
comes up, they’re stranded up on the top of the rock in the hot sun, so they
all die there, because they can’t get back to their water homes fast enough,
and so when they die on the top of the rock they become smelly and that is
what Pelekunu means, it means smelly and it got it’s [sic] name from the
smelly shellfish that dried up on the rock.
So what the children did, they took their little baskets and stuff, by the
way, the baskets are made from the Makaloa weed (sedge plant) which grew
wild in Pelekunu, and they would gather the Hihiwai while their mothers
would be doing their laundry and the fathers would be preparing their
breakfast.54

In addition to the resources of the valley, the ocean provided a plentiful


supply of food. The major source of protein was fish from the ocean. The only
meat they ate was honu (turtle), which frequented an offshore island and swam
in on a certain tide. When the people in the valley felt a certain wind blow-
ing in from Maui, they knew that the turtles would be swimming toward land.
They would get their spears and go down to the beach to catch the honu.
Whoever caught big honu would share with the others. Pelekunu was also
noted for its blue uhu (fish), which swam off of a certain point. The boys used
to climb on a ridge above the point and dive into the ocean when they spot-
ted the distinctive fish’s blue color.55
During World War I Pelekunu was almost completely depopulated.
According to Daniel Napela Naki, Pelekunu was the first of the valleys to be
abandoned. When Jennie Wilson married Johnnie Wilson, mayor of Hono-
lulu, in 1920, she left the valley to live with her husband. She had been the
schoolteacher until the school was phased out in 1915, and she had also served
as the postmistress for the valley. Everyone else also moved out of the valley
about the same time.56 According to Ne, Jennie Wilson had continued to teach
the children of the valley even after the school had closed. When she left,
many of the families decided to move out of Pelekunu so that their young chil-
dren could attend a regular school. In addition, the windward valley taro farm-
ers lost the contract to provide pa‘i‘ai to Kalaupapa, outbid by Honolulu taro

215
chapter five

growers. This cut the Pelekunu taro farmers off from a major source of
income. Some moved to Honolulu and others to the other side of Moloka‘i,
where they continued to farm. A few applied for and received Hawaiian
homestead lands at Ho‘olehua.57

Wailau
According to Emory, Wailau was very similar to Pelekunu. It had a peaceful
Hawaiian village along the crescent-shaped bay, taro patches behind the vil-
lage, and countless ridges leading up to the precipitous valley walls. Another
traveler to Wailau in 1921 was impressed with the natural resources around
the stream, which provided residents with ample and easily attainable food:
“They hurried into the stream and in no time returned with some lehua-eat-
ing ‘o‘opu and hïhïwai shellfish. A fire was lit, the ti-leaf wrapped fish laid on
it and when cooked we ate lunch. We ate heartily of ho‘io fern leaves, shrimps,
the lehua-eating ‘o‘opu fish of Pi‘ilani, the wï and hïhïwai shell-fish, so numer-
ous in this stream.” 58
Daniel Napela Naki, born in Wailau, recalled the livelihood of the people
there in an interview with Mary Kawena Pukui in 1961. According to him,
fishing and taro were the main work. The valley provided them with most of
their basic needs. However, when they needed provisions such as salt, coffee,
flour, crackers, kerosene, and household or garden tools and supplies, the res-
idents of Wailau would hike out and back in through the ridge above Mapu-
lehu on the leeward side of the island. Access into the valley from the ocean
was possible only during the calm summer months. Otherwise, they would
hike in and out of the valley, a journey of three to four hours if they were bur-
dened with a lot of supplies or two and a half hours if they were traveling
light.59
The native küpuna informants from Wailau boasted of the abundance of
food that the valley provided for the Hawaiians who lived there. The taro was
so plentiful that they did not grow any sweet potatoes there, as was common
in other parts of Moloka‘i. In speaking about the extensive system of taro
patches in the Wailau of his youth, Naki referred to the old Hawaiian proverb
“Aia no i‘a malalo, aia no i‘a maluna” (there is food below and there is food
above).60 There was plenty in the stream—hïhïwai, ‘o‘opu, ‘öpae—and in the
ocean. Even in the taro patches the püpü lo‘i, which is similar to escargot, was
raised. Sugar cane was cultivated on the banks of the taro patches.
Another kupuna from Mana‘e, Amoy Duvauchelle, recalled how she stayed

216
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

in Wailau for one month when she was seven years old. She distinctly remem-
bered the ‘o‘opu, hïhïwai, ‘opihi, mountain apples, and white and yellow gin-
gers of the lush valley.61 Emma Apana, who was born in Wailau in 1892,
recalled how full her net would get whenever she went to catch ‘öpae. She
described how the ocean provided an abundance of ‘opihi, pai‘ea and ‘a‘ama
crabs, squid, ‘ö‘io, moi, and äholehole.62 Mr. Kaopuiki remembered catching
akule, moi, halalü, and all kinds of fish there.63 The problem with getting fish
from the ocean was that boat launchings and landings were safe only during the
summer months. Thus, most of the fish were caught near the shore through
diving and the use of nets. For this reason, the Wailau Hawaiians frequently
exchanged their pa‘i‘ai for fish with the residents living on the kona side of the
island. The Hawaiians also utilized all of the various native plants in the valley
to develop a well-rounded diet. Naki described how even the pöpolo plant and
potato leaves were baked in ti leaves in place of meat and eaten.64
The only basic necessity they lacked was pa‘akai, or Hawaiian salt. Because
the sun shone on the valley floor for only a few hours a day due to the steep
vertical cliffs, there was no place to dry out the ocean water for salt. The res-
idents would gather salt from Kalaupapa during the summer months or buy
it from Kaunakakai when they hiked out for provisions.65
Naki was raised speaking Hawaiian and pidgin English. He recalled that
Hawaiians knew how to heal themselves with native plants and lamented the
reliance on pills and injections due to the disappearance of these plants and
their use. Maile höhono was used as a laxative. The o‘o moa, which grows
sometimes on the hapu‘u, was cooked with potatoes, pounded together, and
made into small balls for a cathartic. Ko‘oko‘olau was the original Hawaiian
tea. Uleulehala was good for expectant mothers, and the ‘ilima flowers were
used as a purgative.66
In summing up life in Wailau, Naki used the phrase “Hala no ia la” (and
so passed the days). Life was pleasant, and the days passed easily. After World
War I there were about four families living in Wailau. The school had closed
by 1920, and the last family left the valley in 1937. The majority of families
who left Wailau moved to Mana‘e.

Hâlawa
The ethnohistoric literature on Moloka‘i singles out Hälawa as the wealthiest
taro-growing valley on Moloka‘i. It has been excavated and studied as a clas-
sic example of irrigated agricultural adaptation in the Hawaiian Islands.67 In

217
chapter five

the early 1900s Chinese moved into Hälawa to grow rice, but they abandoned
the paddies as well as their water buffalo when cheaper California-produced
rice took over the market.68
Through the end of the period under review, Hälawa continued to be occu-
pied and taro continued to be grown for subsistence and sale. Handy’s 1931
survey of Moloka‘i described the extent to which taro was cultivated there:

At the eastern end of Molokai the beautiful valley of Hälawa with its broad
flats is one of the few localities where taro is cultivated intensively by
Hawaiians today. In 1931 only the lower terraces were planted, the taro
being grown partly for subsistence and partly for sale. Since that time
events have transpired which favor a revival of subsistence planting and
perhaps the rehabilitation of the abandoned terraces.69

As late as 1935 Hälawa had a population of ninety people. They continued


to produce taro for markets in Kaunakakai and Honolulu. In fact, it was
not until the tsunami of 1946 that the population of Hälawa significantly
decreased. The tsunami destroyed homes and devastated the two major irri-
gated taro pond field complexes, ending the commercial production of taro at
Hälawa.70
John Akina was born in Hälawa in 1896, and his wife, Edith, was born
there in 1906. In 1961 they were interviewed by Mary Kawena Pukui. The
Akinas explained that everyone in Hälawa, as at Pelekunu and Wailau, worked
together to cook taro in the imu, clean it, and pound it into pa‘i‘ai to sell. The
people of Hälawa used pieces from a cracked pöhue or Hawaiian ‘umeke gourd
to scrape and clean the taro and fed the peelings to their domestic chickens and
pigs.71 Rebecca Uahinui, born in Hälawa in 1891, named some of the varieties
of taro raised in Hälawa: a‘apu, uli, ‘ele‘ele, pi‘iali‘i, lehua, and mana. She also
spoke of the collective labor needed to produce pa‘i‘ai.72
At Hälawa, as at Pelekunu and Wailau, many kinds of marine and aquatic
foods from the fresh mountain stream and the ocean could be caught. Mrs.
Uahinui recalled the abundance of ‘alamihi crabs, ‘opihi, ‘öpelu, ‘äweoweo,
hinalea, and manini caught in the ocean; and ‘o‘opu and ‘öpae caught in the
stream.73 According to John Akina, ‘a‘ama crabs were especially plentiful when
the ocean was calm. He also remembered seeing a lot of ‘ö‘ö and ‘i‘iwi birds
in the valley as a young boy. He blamed the mongoose for killing off the
native birds in the valley.74
Like Pelekunu and Wailau, Hälawa lacked a source of salt. Thus, the resi-
dents got their salt from Kalaupapa. In the summer months when the ocean

218
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

was calm and the salt would gather on the flat beach rocks at Kalaupapa, the
people would venture out by boat to Kalaupapa to gather salt. Kalaupapa was
noted for its clean, high-quality salt.75

Mana‘o Aloha ‘Âina o Nâ Kûpuna


“Hala no ia la” (so passed the days) was the phrase most often repeated by the
kupuna of Moloka‘i in Pukui's interviews of them in 1961 and 1964. These
interviews were conducted in Hawaiian with Moloka‘i Hawaiians who grew
up between 1900 and 1930. In the interviews, the Moloka‘i Hawaiians talked
about the way of life in east Moloka‘i. They revealed a persistence of cultural
values, beliefs, and practices. The phrase “Hala no ia la” was usually spoken
as a final comment in reference to the abundance of natural foods available to
them on Moloka‘i, how pleasant life was, and how easily the days passed from
one to the next.
Sarah Wahineka‘apuni Naoo was native to Honouliwai. She exemplified
the intimacy that Hawaiians maintained with the land. At Pukaulua beach,
Sarah knew where the different varieties of limu, such as lïpa‘akai, limu kohu,
lïpoa, and manuwea, flourished, their use, and how to clean and prepare them
properly. She knew the types of fish, shellfish, and crabs that gathered in the
different sections of the beach, their habits and niches, and when and how
best to catch them.76 Each of the surrounding boulders that formed the dis-
tinctive features of the landscape had descriptive names. For example, a rock
that resembled a person sitting and observing the sea was called Pöhakuloa.
Another rock at Pukaulua was named Pöhaku Puka.
Sarah was also familiar with the life cycle of the freshwater o‘opu which
lived in the Honouliwai stream. According to Sarah, they ran in season during
September. They came down from the upland when there was plenty of water,
but when the water was white they wouldn’t come. The fishermen went at
night to catch the o‘opu. When they caught more than needed for the house-
hold, Sarah’s ‘ohana would sell the o‘opu in the market.
Sarah Naoo had lived in close harmony with the ‘äina for a lifetime, and
the ‘äina provided her with a large part of her food. She assured Mary Kawena
Pukui of the richness of life in east Moloka‘i: “‘A‘ole ‘ike mau i ka hale kü‘ai.
Kahi o Moloka‘i Nei, ‘a‘ole pilikia. Hele hulihuli ma kahakai, ka pïpipi, papa‘i,
‘öhiki” (No constant visits to the store. This part of Moloka‘i, here, no trou-
ble. Go searching at the beach . . . pïpipi, crab, ‘öhiki).77
Sarah Ka‘ai‘a Kalima of Kalua‘aha also shared her mana‘o. She spoke of the

219
chapter five

fragrant limu ‘ele‘ele that grew at Kalua‘aha, where freshwater springs bub-
bled up on the shore. The ‘alamihi and ‘a‘ama crabs were also plentiful.78
The kupuna were steeped in Hawaiian folklore, and it formed an integral
part of their identity as Hawaiians of Moloka‘i. Pukui offered an interesting
comment about the importance of the knowledge about place-names during
one of the interviews:

No ke aloha no, pa‘a ka inoa o ka makani, a me ka ua, pehea aku la. Mina-
mina mäkou ke inoa ‘äina o nalowale, o like auanei me O‘ahu. Ulu mai ko
O‘ahu ka namu ano ‘ë.

[It was out of love, they gave and remembered the names of the wind, the
rain and whatever else. We hate to lose the place-names, lest it become like
O‘ahu. O‘ahu is growing with peculiar foreign speech.]79

Knowledge about the place-names and the traditions behind them was sig-
nificant to the people of Moloka‘i. It informed them about their ancestors,
described their adjustment to the natural environment on Moloka‘i, and
explained their cultural beliefs and practices. In their day-to-day experience,
it was commonplace for them to feel spiritual forces at work behind the nat-
ural phenomena they observed.
Sarah Kalima shared her understanding of the various epithets for Molo-
ka‘i listed at the beginning of this chapter. Sarah affirmed that Moloka‘i Nui a
Hina meant Moloka‘i the Great, Child of Hina, because, according to legend,
Hina gave birth to Moloka‘i. Sarah also described the cave of Hina in the
Kalua‘aha district as having flat stones and maidenhair growing in it. In front
of the entrance grew a kukui tree. According to the ancient Hawaiians, when
Hina went there to bathe, she would pray and the cave would fill with water.
The famous old saying was that no one had really seen Moloka‘i until they had
seen the cave of Hina. In Sarah’s youth, the cave was maintained, and people
would visit it with ho‘okupu or offerings.80
According to Sarah, Moloka‘i Pule O‘o (Moloka‘i of the Powerful Prayer)
referred to the traditional practices of ‘anä‘anä that Moloka‘i was famous for.
As an example of how Moloka‘i acquired that reputation, Sarah told the story
of how Kamehameha‘ailü‘au of Maui conquered Moloka‘i and drove the orig-
inal people inland. The people of Moloka‘i resented the taking of the shores.
When they were commanded to prepare a lü‘au for the subjects of Kameha-
meha‘ailü‘au, the people pounded the ‘auhuhu plant used to stun fish and
mixed it with the sweet potato poi served at the feast. All of Kamehameha‘ai-

220
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

lü‘au’s people died except his steward, who lived to tell the tale and spread the
reputation of Moloka‘i Pule O‘o.81
According to Sarah Kalima, “Moloka‘i I Ka O‘o Lä‘au” was yet another
famous saying. It referred to the poles used to steer and propel canoes through
the shallow waters of the reef that extends over a mile from Mana‘e’s shore.
“Moloka‘i Ku‘i Lä‘au” referred to the ku‘i hula step that was reputed to have
originated on the island.
Regarding the place-names of Moloka‘i, Waldemar Duvauchelle and Zellie
Duvauchelle Sherwood each shared the story of how loulu palms started to
grow on the small island off Wailau. A young chief successfully used the broad
leaves of the loulu to fly from the cliffs above Wailau to the island in order to
win the hand of the ruling chief ’s daughter in marriage. Ever since that time
loulu palms grew on the small island. They also both told the story of Keana-
puhi, the ocean cave at Wailau that is large enough to accommodate a forty-
or fifty-foot sampan. The guardian of the cave fought with and defeated a
Shark God on that side of the island. Anapuhi forced the Shark God to guar-
antee that no human would be attacked by a shark in the waters between Wai-
lau and Hälawa. There is no record of anyone ever having been attacked by a
shark in the ocean there.82
Sarah Kalima explained how the stones for building the ‘Opeahina fish-
pond and the ‘Ili‘ili‘öpae heiau were carried over the mountains from Wailau.
She said that one can hear the music of those who built the fishpond and the
heiau on the night of the Käne moon. The menehune builders play the ‘ükëkë
and the drum on those nights. According to Kalima, the guardian of Hono-
uliwai is the shark ‘aumakua Kauhuhu, who eats only human wrongdoers. She
talked of Ka Pu‘u Ne‘e o Hä‘upu, the hill that was lifted up by a turtle until
it was killed by a supernatural man.83
Zellie Sherwood shared her version of how the prophet Lanikaula was
killed by the jealous kahuna of Läna‘i, who burned and prayed over Lanikau-
la’s excrement. He was buried by his sons in his famous kukui grove above
Hälawa. She remembers seeing the grave when she was young, before the area
was bulldozed.84
Tollefson, Kaopuiki, and Poaha all talked of the kioea sea bird, which fre-
quents the reefs and shore off Mana‘e. They spoke of this bird as “Ka manu
kahea i ka lawai‘a” (the bird that calls to the fishermen). It was believed that
when the birds called out it was time to put out the canoes and go fishing. The
bird was migratory and frequented Mana‘e during the kona wind season,
which coincided with good fishing conditions.85

221
chapter five

Grace Hagerman said that she often heard and saw spirits. As a child she
remembered hearing the huaka‘ipö or night marchers. She shared a story about
going fishing in a pond, not knowing that she had begun to menstruate. She
felt a creeping sensation in one cheek, then felt her hair being yanked, and
suddenly realized that going into the pond in her condition had offended the
spirit guardian of the pond.86
It seemed natural to Mitchell Pau‘ole that they experienced problems with
caterpillars in the area of Pu‘u Pe‘elua or Caterpillar Hill. According to leg-
end, a supernatural caterpillar that could turn into a man was burned alive by
the family of the woman he had married. In the fire he burst into hundreds
of caterpillars, and to this day there are many caterpillars in that area.87
Both Waldemar Duvauchelle of Puko‘o and Daniel Naki of Waialua
recalled that akule stopped frequenting the ocean between Puko‘o and Waia-
lua for nineteen to twenty years prior to the time they were interviewed in
1961. Duvauchelle explained how the akule disappeared after an old man had
used a kü‘ula stone, or Hawaiian Fishing God stone, to pray for the fish to
come into Puko‘o. One day there were so many fish, they couldn’t catch all of
them fast enough. The fish began to die off for lack of oxygen in the water.
The people tried to chase the fish out of the pond, but many still died. Since
that day, no akule returned to Puko‘o, and Duvauchelle had not seen any akule
from the time he was young until 1961. The kü‘ula stone disappeared after the
old man died.88
Naki attributed the disappearance of the fish to problems between the fish-
ermen. According to him, the fishermen kept grumbling among themselves,
wishing each other bad luck, and the fish could hear this. One day the fish
simply disappeared. According to Naki, no fish were seen in the area for nine-
teen years.89
It is evident that the native informants who spent the early years of their life
on Moloka‘i between 1900 and 1930 lived as close to the land and ocean as
their ancestors had prior to the arrival of Cook. They continued to live upon
the lands of their ancestors and to cultivate taro, sweet potatoes, bananas, sugar
cane, and so on. However, even in the remote districts, they had better tools
and equipment available to them, such as iron rather than wooden ‘o‘o, hoes,
shovels, and saws. Also, unlike their pre-contact ancestors, they cultivated the
land as much for commercial enterprise as for household use. They also grew
a variety of introduced cash crops and foreign plants for home consumption
and for sale.
The Moloka‘i Hawaiians continued to fish the fishponds, the mountain

222
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

streams, and the ocean as their ancestors had done in ancient times. In most
cases they drew upon knowledge acquired from their elders about the habits
and habitats of the fish as well as the ocean and its daily and seasonal changes.
Although they used similar methods of fishing, they also used better boats,
tools, and equipment such as motorized boats, nylon nets, snorkels, and gog-
gles.90 In hunting, there were more introduced animals and fowl than had been
available prior to European contact, and methods of hunting had changed
under foreign influence as well.
Despite being punished in school for speaking Hawaiian, children contin-
ued to speak the mother language in households that were composed of the
‘ohana.91 This was the case in Mana‘e as well as the windward valleys. Within
the ‘ohana, many generations lived together, and thus the young continued
to learn the Hawaiian language and traditions from their grandparents and
elders.
Finally, Hawaiians often expressed their love and appreciation for the land
in chants and songs. Songs about places often explained the history and tradi-
tions connected with selected districts, serving to record the significance of
important and sacred places. The song “Nani Hälawa,” written by David Kala-
‘au and sung for Mary Kawena Pukui by Same K. Enos, recounts the signifi-
cance of various districts of Moloka‘i:

Nani Hälawa Waiho mai Beautiful Hälawa lay out there


Me ka wailele o Moa‘ula With the Moa‘ula waterfall
Wai kau mai i ka pali Water suspended there on the cliff
Wai kaulana o ka ‘äina Famous water of the land
Kaulana Waialua ka hela i ka la‘i Famous Waialua spread out calmly,
peacefully
Kehakeha i ka uka o Pakaikai Proudly standing in the upland of Pakaikai
Ilaila ‘ohu ‘oi loua nui There in the thick fog
Home noho a ka lani mehameha Home lived by the chief
Kamehameha‘ailü‘au
Kaulana Kainalu i ka ‘ehu a ke kai Famous Kainalu in the spray of the sea
Me ka wai hu‘i koni o ka wai kapu With the cold tingling water of the sacred
water
Ilaila ho‘i au i ‘ike i iho ai There, indeed, I came to know
I ke ‘ala o ka lïpoa The fragrance of the lïpoa limu 92

Unlike the families of East Moloka‘i, change was the dominant pattern of
life between 1900 and 1930 for the Hansen’s disease patients at Kalawao-

223
chapter five

Kalaupapa and the areas where the Hawaiian Homes Commission sponsored
settlements beginning in 1922.

Kalaupapa and Kalawao


In 1900 there were 1,177 people in the settlement. By 1910 this number had
decreased to 785, of whom 624, or 80 percent, were Native Hawaiian. By 1920
the population at Kalawao had decreased to 667, of whom 589, or 88 percent,
were Native Hawaiian. By 1930 there were only 605 residents at Kalaupapa,
of whom 400, or 66 percent, were Native Hawaiian.
Life in Kalaupapa between 1900 and 1930 was a vast improvement over the
days when Father Damien ministered to the basic needs of the exiles. Board of
Health services and programs steadily expanded during this period with sup-
port from the U.S. federal government. The patients led active lives—socially,
physically, and politically. Yet they still suffered the pain of being separated
from family and loved ones, bore the shame of general social ostracism, and
endured long-term physical deterioration from Hansen’s disease.93 According
to a Hawaiian male patient who had gone to Kalaupapa in 1920: “We are like
people anyplace else. We love, marry, drink, murder, commit suicide, fight.
We have all the human drama. We are everything you are on the outside. Just
like that—life is life.” 94
From 1900 through 1929 the residents of Kalaupapa formed social clubs
and participated in church activities. There were six churches in Kalaupapa
during this period—two Protestant, two Catholic, and two Latter-Day Saints
(Mormon). The Kalaupapa Settlement had assembly halls, a bandstand, a race
track, a baseball field, shooting ranges, athletic clubs, debating societies, two
small brass bands, and glee clubs.
In 1908 a children’s nursery for the healthy infants of afflicted parents was
opened at Puahi at the base of the pali that leads out of Kalaupapa. Parents
were allowed to visit their children on Wednesdays and Sundays if they
received a permit. A Hawaiian who lived at Kalaupapa for sixty-seven years
explained how painful it was to be separated from his children after they were
born to him and his wife:

You know, the babies that were born inside here were not allowed to stay
with their parents. After the babies were born, the law said they had to be
taken away to the baby nursery in Kalaupapa . . . We would try to keep the
babies as long as we could, but most times, we kept them only until morn-

224
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

ing. Then we would carry them to the nursery . . . They allowed the chil-
dren to live one year inside Kalaupapa nursery. There we could see them
only through thick glass, but no can touch! Then after one year, they were
removed. They were either hanai by family members, or “issued” out for
adoption by the Board of Health.
It was so hard to give up your children like that . . . they never took
good care of them, yet they would not let us care for our own children,
even when we knew they were sick. It was hard. You love them, and then
they are taken away, just like we were taken away.95

Nevertheless, such a policy was necessary to protect the health of the child
and was ultimately in the interest of the parents as well. An alternate policy
would have been mandatory sterilization of the patients, which would have
denied the patients their right to have offspring at all.
Residents of Kalaupapa caught fish to supplement their food rations. Under
the sponsorship and supervision of the Board of Health, they raised livestock;
ran a dairy until the drought of 1909; operated a poi factory, a laundry, a saw
mill, a slaughterhouse, and an ice-making plant; and raised crops, including
taro, sorghum, alfalfa, pumpkins, and papayas. Those who worked in these
enterprises were paid wages by the territorial government. In 1901 a resident
opened a private store at Kalaupapa. In 1902 a dry goods store and a grocery
store opened at Kalawao. A bakery and a fish market opened in 1904. In 1905
a Hui Ho‘o‘ikaika Kino (Group to Develop Strong Bodies) contracted with
the Board of Health to cultivate all available land in Puahi in taro. One-third
of the output was to go to the board, and two-thirds could be sold at market
rates.96
In 1930 a committee appointed by the governor, Lawrence M. Judd, to
study Hansen’s disease made a number of recommendations that, when imple-
mented, marked the end of an era for Kalaupapa. The pivotal recommenda-
tion was the second one: “The adoption of a policy whereby there would be
no further involuntary transfers of patients to Kalaupapa.” This ended the
policy of punitive segregation of Hansen’s disease patients and ended Kalau-
papa’s stigma as a penal institution. The committee recommended improve-
ments at Kalaupapa totaling $200,000. It also recommended improvements at
the Kalihi Hospital in order to accommodate a total of 350 patients. Land
acquisition and improvements at Kalihi were to cost $375,000.97
According to the report, there were 760 known cases of Hansen’s disease
in Hawai‘i as of June 30, 1930. Moreover, the number of those afflicted had

225
chapter five

averaged 760 since 1910, with the number of deaths from Hansen’s and the
number of new cases being relatively equal.98
The committee felt strongly that hospitalization was the most effective way
to control the spread of Hansen’s disease but recognized that efforts to insti-
tutionalize the patients at Kalaupapa would “result in failure from lack of
cooperation on the part of the inmates, and would create a hiatus of mental
unrest, something that should be carefully avoided.” For that reason, they
recommended no additional transfers to Kalaupapa and a concentration of
expanded health-care facilities at the Kalihi hospital.99
Nevertheless, an extensive program to rehabilitate and revitalize the facil-
ities at Kalaupapa and to construct new facilities for water and electricity was
initiated in 1931. The legislature appropriated $400,000 in 1931 for public
works projects at the settlement and an additional $200,000 in 1932. Accord-
ing to the Judd Committee report, these improvements were aimed at adding
to the happiness of the patients and their dependents. The policies Judd ini-
tiated marked the beginning of a new era in the Kalaupapa settlement.100
In ending this overview of the conditions in Kalawao from 1900 through
1930, I would like to close with the observations of a blind, disabled Hawaiian
who was among the last to be involuntarily banished to Kalaupapa in 1930.
He reflects the pride and dignity that the victims of Hansen’s disease upheld
in spite of their lifelong suffering and affliction. He also describes the reliance
and love that the patients developed for each other as an ‘ohana:

No, I’m not bitter about this disease. I have been here at Kalaupapa since
I was twelve. Maybe I’m lucky. I never think about it too much. I get three
square meals a day here. I get care from the nuns and the nurses. I have
my talking books and I listen to my big band music . . . Like the other
patients, they caught me at school. It was on the Big Island. I was twelve
then. I cried like the dickens for my mother and for my family . . . They
sent me to Honolulu, the Kalihi Receiving Station, real fast. Then they
sent me to Kalaupapa. That’s where they sent most of us. Most came to
die. So, I stay here . . . Me, I’ve got the heart problem and stomach ulcers.
I’m blind and I can’t walk. Then I get side effects from the sulfone drug
medicine. Still, maybe I’m lucky. In Kalaupapa, we are all in the same boat;
we help one another. We are one family, all the same, with love in our
heart . . . with aloha for each other. Oh, we fight between ourselves, like
in any family, but we are all in it together here. There is no where else for
us to go.101

226
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

In 1941 sulfone drugs were shown to effectively treat and arrest the dis-
ease, and after treatment patients who so chose could reintegrate into society.
For most, however, Kalaupapa was their place and their home. They chose to
live out their lives at the settlement with others who had become ‘ohana to
them.

Hawaiian Homesteading: ‘Âina Ho‘opulapula


In 1921 the Hawaiian Homes Commission was established by the U.S. Con-
gress through the efforts of the Aha Hui Pu‘uhonua o Nä Hawai‘i, the terri-
torial legislature, and Hawai‘i’s delegate to Congress, Prince Jonah Kühiö
Kalaniana‘ole. The purpose of the commission was to manage and adminis-
ter 200,000 acres of the original Crown and government lands of the King-
dom of Hawai‘i for the rehabilitation of Native Hawaiians who were of half
or more Native Hawaiian ancestry. The lands were to be leased out as agri-
cultural homesteads for ninety-nine years for $1 a year.
The first commissioners included the governor of Hawai‘i, Wallace R.
Farrington, and his four appointees—Kalaniana‘ole, Akaiko Akana, Rudolph
Duncan, and George Cooke.102 During the first five years of the program,
operations were to be experimental and limited to Moloka‘i lands and selected
tracts on Hawai‘i island. In 1922, the Hawaiian Homes Program was estab-
lished on Moloka‘i, where it continued for another eight years. The commis-
sion brought together Hawaiians from all of the major islands to start a new
community that would revitalize the Hawaiian people. They selected appli-
cants who were at least half Hawaiian in ancestry, had an aptitude for farm
work, and were of good character.
The homesteaders who settled on Moloka‘i came from Honolulu, rural
O‘ahu, Maui, Hawai‘i, Kaua‘i, and Moloka‘i. The men had worked as steve-
dores, laborers, farmers, carpenters, engineers, mechanics, truck or tractor
drivers, foremen or superintendents, clerks or salesmen, firefighters, and mail
carriers. The settlers also included a dentist, a fisherman, a cowboy, a black-
smith, a welder, a boilermaker, a plasterer, and a music instructor. The women
were primarily homemakers, but there were also a trained nurse and a hotel
cafeteria worker among them. They were Mormon, Catholic, and Protestant.
Some had graduated from the Kamehameha Schools, and some had only sixth-
grade educations. Some were fortunate enough to have moved with their
brothers and sisters to Moloka‘i, but the majority were unrelated to one
another.

227
chapter five

The lands allotted to the Hawaiian Homes Commission on Moloka‘i were


second-class pastoral lands. In 1918 they were valued at $1 per acre, and the
government had rented them out for 5 cents an acre.103 On the positive side,
the Moloka‘i lands consisted of deep-red fertile soil. From September through
February, the wet season, there was enough rain to grow crops without irri-
gation. Thus, Moloka‘i’s primary growing season coincided with the winter
season, when mainland-grown vegetables and fruits were usually scarce in
Hawai‘i markets. The pineapple plantation camps and the Moloka‘i Ranch
provided an accessible local market for fresh produce and animal feed. The
Honolulu market was only sixty miles away by ocean. However, the Moloka‘i
lands also had serious drawbacks. They lacked sufficient water for domestic
use throughout the year and for growing during the dry season from March
through August. Strong winds blew across them throughout the year, and they
were infested with pests and diseases.104
In November 1921 the commissioners toured all the Moloka‘i lands allot-
ted to them with their agricultural experts and engineers. They selected the
coastal flats at Kalama‘ula as the location for the first homesteading settle-
ment and chose plots at Kala‘e, Malehua, and Kalama‘ula for demonstration
farm projects.105
Beginning in January 1922 commission engineers and agriculturalists sur-
veyed the land, constructed roads, and developed water for irrigation and
domestic use from the Waihi‘i spring and other springs in Kalama‘ula. The
Kalama‘ula lands were subdivided into twenty-three lots and a 2,100-acre
community pasture. By the time that homesteaders were ready to move in, a
storage dam and two 10,000-gallon storage tanks were constructed to provide
2.5 million gallons a day from the spring and the well.106
In January 1922, before any homesteaders were on the land, the first
demonstration farm was started at Kala‘e in upper Kalama‘ula. The Commis-
sion planted Guam corn and New Era pigeon peas and raised chickens. The
plot was located on an exposed, windswept hillside. During the first growing
season, rainfall was below average—the season was the driest since the drought
of 1908— and there was no equipment to cultivate the crop after the initial
planting. Nevertheless, the demonstration project succeeded in producing 400
sacks of corn from fifteen acres and a large crop of pigeon peas from two acres.
The project also hatched and raised 500 chickens.107
The demonstration farm at lower Kalama‘ula was located on an old swamp.
The soil packed easily and, without irrigation, was inclined to bake into large,
tough blocks. The commission planted alfalfa, sweet potatoes, beans, corn, and

228
Figure 31 Kapuaiwa Coconut Grove is in the heart of Kalama‘ula, where the first
Hawaiian homesteads were established on Moloka‘i. Circa 1920s. Tai Sing Loo,
Bishop Museum.
chapter five

vegetables.108 As soon as the homesteaders moved onto the land, the demon-
stration project was phased out in favor of having the settlers carry out the trial
work on their own land in consultation with the commission’s experts.109
The third demonstration farm in Malehua experimented in pineapples and
forage crops for fattening of livestock. Initially it was planted in corn, and
then, beginning in 1923, pineapples were grown. Eventually it too was phased
out.110
By February 1923 thirteen homesteaders were on their lots and hard at
work.111 Homesteaders cut down the kiawe trees covering their land and
burned all the brush. The commission assisted them by pulling out the tree
stumps with a tractor. After the land was cleared, the homesteaders plowed
their lots using commission equipment.112 Homesteaders lived in tents and
makeshift shelters for a short period while building permanent homes on their
lands. They worked from daybreak to nightfall to build their homes, clear their
land, and plow and plant it. They called Kalama‘ula’s homesteading area Kala-
niana‘ole Settlement, in honor of Prince Jonah Kühiö Kalaniana‘ole, who had
been instrumental in getting the U.S. Congress to establish the Hawaiian
Homes Commission and who had passed away on January 7, 1922. By the end
of the first year, in 1923, the commission filed a report:

There are present (February 11th) 13 homesteaders on the ground and


actually at work. With the exception of the greater part of the clearing,
they are doing their own work. All are enthusiastic, working hard, and
from present indications, bound to succeed. Men, women and children all
work. For the majority, time, such as eight or ten hour days, matters little.
They lead the regular farmer’s life—sunup to sundown. Pending the final
clearing of the low lands, and the completion of the flume line, many have
cleared portions of their rocky upper lands.113

By summer 1924 the Hawaiian homesteaders were harvesting sweet pota-


toes, watermelons, corn, cucumbers, bananas, tomatoes, and alfalfa. Home-
stead tomatoes controlled the Honolulu market. A net return of nearly $2,000
was made from a field of two and a half acres. The alfalfa yielded one ton per
acre and brought in $30 to $35 a ton when baled. Homesteaders were able
to produce ten alfalfa crops a year and thus earn $300 to $350 per acre, per
year. At the same time they raised 835 hogs and 2,308 chickens. There were
127 head of cattle on the homesteads, 114 horses, 17 mules, and a few goats,
raised on an experimental basis. The project was hailed as the “Moloka‘i
Miracle.” 114

230
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

Expanding upon the successful homesteading experience at Kalama‘ula, the


commission opened the Pala‘au-Ho‘olehua area for homesteading in 1924.
Seventy-five families settled in Ho‘olehua between 1924 and 1926, eight went
in 1928, and forty-eight arrived in 1929.115
The Ho‘olehua families planted large quantities of corn, pumpkins, melons,
tomatoes, cucumbers, sweet potatoes, squash, and peanuts to sell. They also
raised vegetables for home consumption such as beans, onions, and cabbage,
as well as livestock.116 Farming at Ho‘olehua was difficult owing to high winds,
several droughts, and infestation with pests and blights. It required a substan-
tial financial outlay and constant care, including spraying. Marketing of the
crops was also problematic.117 However, the progress they made was impres-
sive.
In September 1926 the Maui County Evangelical Association held its semi-
annual meeting at Kalua‘aha, Moloka‘i. Henry Judd, editor of The Friend,
reported favorably on the progress they observed at Kalama‘ula and Ho‘o-
lehua:

To those who had been acquainted with Moloka‘i as it was a few years ago,
it was also a new experience to witness the transformation that has been
wrought both on the makai lands, where a large area formerly in kiawe
forest is now under active cultivation, and on the mauka lands, where old
pasturage is now being plowed up, houses erected and vigorous efforts
being made to establish the eighty-two homesteaders in a successful ven-
ture in farming.
Away from the deleterious influence of the slums of Honolulu, Hilo and
other crowded places, located on the broad acres of Molokai’s rich fields,
there is every chance for the Hawaiian people to come into their own, to
revive the old traditions of farming practiced on the island of Molokai
itself as well as on other islands of the group.
To one who had not visited Molokai for four years, it was a revelation
to observe the great amount of hard work that has been done in both sec-
tions of this rehabilitation project. It will continue to be an object of much
interest to this observer and will be watched with the confident hope that
success will crown the efforts of the Hawaiian people themselves to be
rehabilitated in body, mind and estate and—most important of all—in
spiritual life and power.118

Gertrude and Mitchell Pau‘ole, who moved to Ho‘olehua from Honolulu


in 1925, were interviewed in 1961 by Mary Kawena Pukui about their pio-

231
chapter five

neering efforts at Ho‘olehua. Gertrude Pau‘ole shared the following experi-


ences about leaving Honolulu and settling at Ho‘olehua:

We came on the ship, Likelike in 1925, January. We left on the 6th. We got
here on the 7th. If we didn’t work hard . . . no more nothing. We don’t see
meat. We don’t see apple. My children eat the tomato and the cucumber as
fruits. We grow the pumpkin and fix poi. I had to put on one stocking, two
stocking and three stocking. Cold this place. No more nothing. Just lantana
and mäniania. And we had a one by twelve.
When we left Honolulu we sold all my things. The money we used to
buy food—palaoa, sugar and all those things. Hard to buy food here . . .
because he wasn’t working. After he found work it was $59 a month. We
had a quarter acre head cabbage, peanuts, watermelon. We lived on the
peanuts. We bagged them up and stored them in the warehouse and when
we want food we go to the store and sell them. $15 a bag . . . that’s plenty.
No irrigation—just depend on the rain. We had forty acres. Just like
you gamble. You plant. If the rain gonna fall then all right. No rain, you
lose all what you had.119

In 1926 Ho‘olehua homesteaders entered into contracts with Libby,


McNeill & Libby to plant pineapples on homesteaders’ land. Planting of pine-
apples began in 1927, and the first crop was harvested in 1929. Initially, the
planting of pineapples proceeded in conjunction with the cultivation of other
crops. In 1927–28 the homesteaders harvested 42 tons of sweet corn, 2 tons of
peanuts, 200 tons of field corn, 22 tons of tomatoes, 22 tons of pumpkins, 13
tons of watermelon, 3 tons of cantaloupes, 6 tons of Irish potatoes, 2 tons of
cucumbers, and 2 tons of sweet potatoes. They collectively grossed $13,670
from that harvest.120 However, the trend of growing pineapples under a plan-
tation system, in conjunction with the major pineapple corporations—Libby,
McNeill & Libby and California Pineapple Company—was a development
that eventually replaced diversified agricultural enterprise on individual plots.
The reasons for turning to large-scale pineapple production were under-
standable and primarily related to the harsh natural conditions, especially the
lack of water from rainfall or irrigation. The spring water from Kamehameha
V Spring at Kalama‘ula had a high salt content. When the water evaporated,
large amounts of salt remained in the soil. By 1930 the soil at Kalama‘ula was
unfit for raising crops. The settlers there asked for lots in Ho‘olehua where
they could have pineapple grown under contract to the pineapple companies,

232
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

and their request was granted in 1931. Also, pests had increasingly become a
major problem in the homesteaders’ crops. These conditions made it difficult
to predict yields and therefore prevented homesteaders from entering into
contracts that could have assured systematic and profitable marketing of their
crops. Prices for produce often fluctuated depending on the supply. At times
the return was barely enough to cover the costs of production.121 These prob-
lems were summarized by a commission employee for Felix Keesing, who
conducted a study of homesteading on Moloka‘i in 1935 for the territorial
legislature:

We cannot tell what produce the homesteaders are going to have in this
climate, hence cannot enter into any contracts in Honolulu. Our inability
to anticipate the yield, found out by bitter experience, is the key to our
marketing troubles. Where homesteaders have quit producing, it has not
been their lack of interest, but rather the uncertainty of getting a return.122

In light of these difficulties, pineapple production provided a welcome


alternative to diversified agriculture for the Hawaiian homesteaders of Molo-
ka‘i. The contract guaranteed $23 per ton of harvested pineapples. In 1929,
fifty-four homesteaders earned $32.30 per ton of harvested pineapple, for a
total of $121,902 or an average of $2,257 per family. In 1930, sixty-two fam-
ilies earned $303,059 or an average of $4,888.123
Most of the homesteaders felt that if it were not for the pineapples, they
would not have been able to meet their financial obligations or earn a living
on their homesteads. The pineapple crop brought in a substantial income for
the homesteaders during the depression years. Prior to the pineapple con-
tracts, homesteaders had begun to seek outside employment as a dependable
source of income while they struggled to keep up with the care of their crops.
The pineapple crops kept the homestead land in active and profitable agricul-
tural production.124
In the initial years, when the pineapple was planted on individual plots of
land on a rotation basis in checkerboard fashion, many of the homesteaders
weeded and cared for the crop on their individual lot. However, beginning in
1931 the block system was used to plant the pineapple in order to reduce labor
costs and infestation by mealy bugs. At that point the majority of homestead-
ers hired Korean and Filipino workers to care for their portion of the crop,
or they opted to pay the plantation work gangs to do it. By 1936 nearly all the
Kalaniana‘ole settlers worked their pineapple lands, but only a few of the

233
chapter five

Ho‘olehua settlers did so. According to Keesing, the pineapple contracts pro-
vided the homesteaders and the companies with mutual benefits. The compa-
nies were given use of good land without having to pay rent or taxes, and the
homesteaders received credit, expert management training, and assured mar-
ket, and an immediate financial return for work done.125
In an interview with Pukui in 1961, Harry Hanakahi explained why the
homesteaders turned to the farming of pineapples:

We built our home. We paid 5 percent interest on $3,000 loan to build a


house. We paid big interest. But we had ho‘omanawanui til the pineapple
came. If no pineapple, we would have quit the land. They ain’t going to live
on that land there. No water. Dry land. They would have quit.
We planted alfalfa. Planted tomatoes. Then nobody buy in Honolulu.
Was not in demand. What we going do. People don’t know how to pick.
Some only half ripe, some too ripe. Then, were not in demand over there.
What we going do? The Commission tell us, your tomatoes are rotten, you
can’t get nothing. We got discouraged. We raised 200 pigs. Everybody have
pig. Then plague came—pneumonia. All wipe us out. Dying everyday.126

Gertrude Pau‘ole explained to Pukui how they spent the money that they
received from the first pineapple crop: “When we had the first pineapple
money we paid all our debts. We didn’t buy radio, we didn’t buy one thing.
Only we looked first was kaukau [food] because hard to get food here. Was
hard, but the children enjoy—always happy.”127 Mitchell Pau‘ole described
why they decided to plant pineapples and what they did with subsequent pine-
apple earnings:

Our life here was very difficult. But we had the hope that some day our
condition would be improved. And during the year 1927 the question of
using our land for pineapple planting was then discussed. And it was during
that time that those homesteaders here went into planting pineapple and in
1929, those who went in planting pineapple got their first earning in 1929.
The earning that we earned then, not like what we earning today. The
earning then ran into the thousands of thousands of dollars. And because
of that, we had our first earning, then the condition was made much better.
Our extension and the furniture were bought with those earnings, in
cash.128

In 1927, after the five-year experimental period the U.S. Congress had
established had passed, the territorial legislature conducted an investigation

234
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

into homesteading at Kalama‘ula and Ho‘olehua. The Honolulu Advertiser sent


a reporter to accompany the Senate committee that inspected the Moloka‘i
homesteads. His article was printed in March:

Molokai homesteaders ask for nothing more than an even chance to make
good on the lands that they have taken up under the terms of the Hawaiian
rehabilitation scheme. They make no exorbitant demands—only a request
for adequate water that they may irrigate their land and grow crops with
which to conduct a profitable farming business. They are happy and con-
tented now despite the fact that the year 1926 brought but 12 to 19 inches
of rainfall on Molokai and consequently, without artificial irrigation, the
way was beset with many difficulties and setback. The project is a success
and will repay the territory a hundred-fold once conditions are created that
will give these ambitious sons and daughters of the soil an even chance to
make good at their new profession.129

The territorial senators were impressed with the hard work and persever-
ance of the Hawaiian homesteaders, who worked cooperatively with each
other for the overall success of the project. Homesteaders frankly recounted
the hardships they had endured during the drought of 1926. For example,
Kenneth Auld, a graduate of Punahou and the University of Hawai‘i who set-
tled his family at Ho‘olehua, explained how he and his family had struggled to
save two acres during the dry spell and manage to break even. Despite feeling
discouraged, he and his family decided that they would be patient: “Everyone
here is happy and hopeful. I am confident that in five years we will be in a per-
manently sound condition.” 130 Although actual progress from a strictly agri-
cultural standpoint had been limited by the unfavorable climate, the territorial
senators were nevertheless encouraged by the sight of whole families working
together, tilling the soil, side by side, for the success of their own farms. Before
the regular session of 1927 adjourned, the territorial legislature passed Joint
Resolution no. 1, which declared the experiment to be a success and requested
the Secretary of the Interior to approve the activities of the commission and
to extend the work of the commission to all of its designated lands on all of
the islands.131
In April 1927 the Secretary of the Interior, Hubert Work, toured Moloka‘i
to assess the progress of the Hawaiian Homes Commission for the U.S. fed-
eral government. Impressed with the progress of the homesteaders, he urged
Congress to approve and expand the commission’s work. Through the joint
efforts of Work; Victor Houston, Hawai‘i’s delegate to Congress; and Gover-

235
chapter five

nor Wallace R. Farrington, Congress voted to extend the work of the commis-
sion and passed amendments to the Hawaiian Homes Commission Act that
were signed by President Calvin Coolidge on March 7, 1928.132
Mitchell Pau‘ole proudly described the welcome given by the homestead-
ers to the Secretary of the Interior:

After the first five years, this homestead project was then declared a success.
The Secretary of Interior, Mr. Work made a trip to Moloka‘i with commis-
sioners who were then on board. We brought what we had planted as a
ho‘okupu. So he saw what we really had done with those five years. Sugar
cane and everything you can think of. Each one of the homesteaders
brought. We presented a ho‘okupu to him at the homestead office. And it
was then that it was declared a success.133

The Homestead Commission’s report to the 1931 legislature pointed to


general social and economic indicators of its success. It noted that from 1910
through 1930 the numbers of Hawaiians increased from 36,000 to 43,000. Sav-
ings deposits of Hawaiians during the same period increased from $232,329 to
$2,339,000. Personal property of Hawaiians also increased from $1,259,000
to $2,669,000.134
Reports about the Hawaiian Homesteading Program on Moloka‘i up
through 1930 continued to be favorable and supportive despite criticisms from
certain planter interests that had opposed the establishment of the program
in the first place. One such report was published in a six-part series in the
Star-Bulletin from October 4, 1930, through October 10, 1930. The articles
were written by Wallace R. Farrington, a former governor and chairman of
the Hawaiian Homes Commission who was publisher of the Star-Bulletin in
1930. Farrington first visited the Kalaniana‘ole Settlement and was impressed
with the perseverance of these pioneer homesteaders:

Here the Hawaiians under the Hawaiian Homes law, demonstrated what
they could do. They did well. A long and interesting story could be told of
their struggles. Some of them are not through struggling, but the good
work of keeping up the homes and the small farms is going on.
These people have the largest number of problems to meet of any that
have gone on the land under the Hawaiian Homes law. Some have proved
that they can make a success of the diversified farming, raising chickens and
doing general truck gardening that is most adaptable to the land on this
lower level. Others have suffered discouragement and gained much experi-

236
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

ence. All, however, are keeping up their places, maintaining their families,
and they look well and happy, despite the fact that they are not in the class
with the larger financiers of the Hoolehua pineapple section.135

At Ho‘olehua Farrington was most impressed with the hard work per-
formed by the women, whom he considered to be the backbone of the home-
steads. According to Farrington, the women not only bore the children and
did the heavy housework, they also went into the fields and worked next to
the men and were not embarassed to wear old clothes. He commented, “I do
not recall a single report of extraordinary expenditures of new wealth, where
the women of the family have been referred to as the ones who are luxuriously
spending much of the money.” 136
In his articles, Farrington addressed rumors and criticisms that he had fre-
quently heard. One such criticism was that the homesteaders worked in out-
side jobs rather than solely on their farms. Farrington defended that practice:

Because some of these people are engaged in other activities than their
farms, an occasional inquirer comes along to say that they are not home-
steaders. It is hard to understand, however, why anyone should complain if,
for instance, one of these homesteaders finds employment as wharfinger at
Kaunakakai, a very important position. Another finds employment as terri-
torial game warden. Another teaches school. The women folk of these
families stay at home and look after the garden and the crop, and the chil-
dren. The men of the family are capable in their position. Someone must
do the work they are doing. Kaunakakai is growing, the pineapple industry
is making a very busy center of the south side of the island. It seems
entirely reasonable and proper that homesteaders who are equipped to fill
positions on the wharf, or elsewhere, or even in the supervision of road
construction, should take this work while the members of their families
are carrying on the small farm at home.137

Farrington challenged those who promoted the notion that homesteaders


did not work. He acknowledged that the Hawaiian homesteaders were not
experts in growing pineapple and that some worked harder than others. How-
ever, he pointed out that in any group of people it was normal for some to
work harder than others. The most important fact was that a high percentage
of women and children, boys and girls, worked in the fields: “Men and women
have the clean, hard hands that can only be made by hard toil.” 138
How homesteaders handled their pineapple profits was another focal point

237
chapter five

of criticism. Charges of wasteful spending on “luxuries” such as automobiles


or trucks and investments into risky businesses were frequently made. In
defense of the homesteaders, Farrington wrote:

Frequent reference is made to the unusual luxuries bought by these people.


I doubt if they buy any more of the unusual things than do the general run
of humanity.
If you have wanted some particular thing all your life, and after years of
hard work you have enough money to buy it, what do you do?
If you have been foolish at some time in your buying career, should you
be too critical when others show about the same percentage of folly as
yourself ? 139

As to investments in uncertain business ventures, Farrington seemed to


think that the homesteaders had been persuaded to invest in the Hawaii Lum-
ber Company by some very experienced salesmen. They had also followed the
unwise leadership of one of the agricultural experts of the commission, who
had invested $2,000 in the company. He felt certain, however, that they had
learned not to invest in new, high-risk businesses:

The gist of this new education is that it usually pays a little fellow, who has
not time to study up on the details of the prospects, to stay out of new
corporations. Usually, a new corporation has to go through a period of
experience, mistakes and troubles that are the result of competition,
adjustment of new personnel, lack of organization and a multitude of
other things that absorb money but do not pay dividends to the stock-
holders. So the prospect seems fair that as some of the homesteaders are
rather anxious to get out of the Hawaii Lumber Co., they will also be slow
in taking up with cooperative enterprises that are experiments.140

Farrington was also impressed with the Pioneer Day lü‘au. The celebration
had been postponed for a month because the homesteaders were too busy har-
vesting their crops. He made special note of the fact that no intoxicating liquor
was consumed and that the party was a wholesome family affair despite the fact
that the pineapple checks had just been distributed on the previous night.
Overall, Farrington’s assessment of Ho‘olehua was summed up in the fifth
article of his series:

If Hoolehua is not a place where real human, spiritual and material values
of great worth exist and are being renewed and further developed, I do not

238
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

know where an American community can be found that will fill such a
qualification.
The critics may complain all they please and the doubtful ones may be
as dubious as they like, the fact remains that these people have prosperous
homes, they are at work, they are learning to be more skilled in agriculture,
they are interested, they are taking care of their children, they are behaving
themselves as well as average humanity and a good sight better than many
of the critics. They are growing in health, happiness and good citizenship.
If that is not a good investment I do not know what is.141

The turning point in the Hawaiian Homesteading Program on Moloka‘i


came in 1931. By that time pineapple had taken firm root on the homestead
lots, the first pineapple checks had been received, and homesteaders were
bound to ten-year contracts. At that point, the block planting program was
initiated to arrest the infestation of the crops by insects and diseases. In 1931,
homesteaders raised public charges that the Hawaiian Homes Commission
had mismanaged their affairs and exposed them to exploitation from sales-
men and creditors. In response to these charges, Governor Lawrence Judd
appointed a special commission to investigate the complaints and criticisms
against the Hawaiian Homes Commission and to make recommendations on
how to improve management of the program.142
The special committee was chaired by Princess Abigail Kawananakoa and
made up of Charles Chillingworth, Lang Akana, John C. Lane, and Oscar
P. Cox. They visited the homesteads, talked with the settlers, held public
meetings to discuss the problems, and studied the records and reports of the
commission. In its final report the committee charged the Hawaiian Homes
Commission with failure to guide the homesteaders and claimed that: (1)
homesteaders were being exploited, (2) no supervision or advice was being
provided on modern methods to protect and conserve the pineapple fields,
(3) the commission lacked interest in the program, (4) there was no spirit of
comradeship and fellowship between the commission and the homesteaders,
(5) one commissioner had not attended meetings for a year, and (6) the com-
mission had spent more money on the community hall than was necessary.
The committee recommended that the rehabilitation program be extended
to those Hawaiian homelands being leased to non-settlers. It also recom-
mended that the Kalaniana‘ole settlers be allowed to take up leaseholds in the
upper pineapple lands while retaining a residence lot at Kalaniana‘ole. In
addition, the committee suggested that homesteaders be allowed more than

239
chapter five

sixty head of cattle in the community pasture, where a paddock system would
be introduced. The committee also recommended that homesteaders be put
on a budget. Finally, the committee sought removal of the executive secretary
and the agriculturalist for not performing their duties.143 The response to and
implementation of the committee’s report was left to the commission, which
adopted some but not all of the recommendations as policy.
By 1936 the pineapple payments were distributed through the commission,
which provided advice to the homesteaders on how to budget their income to
steadily pay off debts over time as well as to maintain enough income for
immediate family expenses.144

Kama‘âina O Moloka‘i
On the homesteads, the holdings were spaced far apart and did not engender
a feeling of community. Nevertheless, the homesteaders shared what little they
had with one another and developed new social networks. Felix Keesing quotes
one of the old-timers as telling him: “When we came there were just a few
houses. There were no autos or pineapples and only a trail to Kaunakakai. We
all lived together like one big family. What was yours was mine and mine yours
with no thought of pay. Now that is lost. When the big money came the
friendship between us broke. Everyone is now for himself.” 145
Mitchell and Gertrude Pau‘ole also recalled the sharing and help given to
one another during the early days of the settlement. Because they helped each
other in their time of need, the homesteaders considered each other sisters and
brothers. Gertrude Pau‘ole said, “We call each other brother or aunty until
today. So that is the happiness of the homestead here.” 146
Homestead life was not all work and struggle. Among the recreational
activities enjoyed by the homesteaders were planting flower gardens, automo-
bile drives, horseback riding, pool, volleyball, baseball, and orchestras. The
whole community gathered together to celebrate May Day, Kühiö Day, Labor
Day, and an annual fair. 147 In 1927 the commission renovated a barn to use
as a community hall for meetings, and community gatherings.148 Most of the
homesteaders participated in churches—Mormon, Hawaiian Protestant,
Roman Catholic, Ke Akua Ola, and Christian Scientist.149
In September 1924 Maui County opened a four-room schoolhouse on
Hawaiian Homes land near the Kamehameha V coconut grove. There were
seventy school-age children on the homesteads in 1925.150 Most homestead
children married children of other homesteaders. Kinship bonds helped to

240
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

gradually cohere the homestead community over the years. As the homestead-
ers became permanently established and their families matured, single-family
households expanded into ‘ohana. Relatives from other islands stayed for
extended periods of time with homestead families. Adult children, single and
married, continued to live with their family on the homestead land.151
Although they developed informal social networks of cooperation and sup-
port, the homesteaders were unable to form a permanent farming coopera-
tive. In 1923 the Kalanianaole settlers formed the First Hawaiian Homes
Cooperative Association. The Ho‘olehua farmers formed a similar association.
Both eventually dissolved. The commission commented on this phenomenon
in its 1925 report to the Hawai‘i legislature:

At present the greatest obstacle to cooperative marketing by the home-


steaders is their failure to realize that they must share their losses as well
as their gains. The homesteader whose individual contribution to a ship-
ment goes through without damage cannot as yet understand why he
should be assessed for a portion of the loss occasioned by a less fortunate
member of the cooperative organization.152

Felix Keesing commented on this issue in his 1936 report. According to


him, certain homesteaders had better crops or considered themselves to have
worked harder and felt penalized when returns where shared. Before long,
homesteaders began to market part or all of their produce outside of the
cooperative and undermined the association. Commercial enterprise replaced
cooperative work.
The homestead communities founded to establish Hawaiians as commercial
farmers were distinct from the subsistence farmers and fishermen of Mana‘e
and the windward valleys. Once their homestead lands were converted to pine-
apple production, the homesteaders were integrated into and dependent upon
a cash economy based on wage earning and marketing of cash crops. Their
primary source of food was the store, rather than the ocean, gardens, valleys,
and mountains. However, as the Hawaiian homesteaders became kama‘äina to
Moloka‘i, they also hunted, fished, and gathered from Moloka‘i reefs, streams
and forests.
Over time, they sank roots into the land as they raised families. Their chil-
dren grew up with, and some even married, lineal descendants of the original
Hawaiian families in the surrounding districts. The homesteaders’ ‘ohana net-
work expanded to incorporate new relatives and friends, and the homestead
communities cohered and evolved into important centers for the Hawaiian

241
chapter five

community with new cultural and social patterns of interaction. As they expe-
rienced the natural phenomena of Moloka‘i, they began to learn the ways and
traditions of the Moloka‘i people. Thus in 1961, when Mary Kawena Pukui
interviewed homesteaders and their descendants, most were able to share a
deep knowledge of Moloka‘i’s place names, mele, myths, and legends. Mitch-
ell Pau‘ole, one of the most prominent of the homesteaders, who gained a
reputation as the honorary mayor of Moloka‘i, probably spoke for most of the
pioneer homesteaders when he told Pukui: “And so the homestead life has
been a wonderful life for each one of us, regardless of the hardship but we
endured all of that by having that spirit of ho‘omanawanui and be with it. We
took up homesteading with the spirit that we were coming to make the best
of it and we stuck with it until this day. We made our home here and we will
be buried here.” 153 Thus, the Hawaiian homesteaders established themselves
on the ‘äina of Moloka‘i and opened a new chapter in the history of Hawai-
ians on the island. Their achievements, however, were important in establish-
ing a Hawaiian Home Lands Program for Hawaiians on every island. A beau-
tiful and popular song written to praise the beauty of Kalama‘ula, Moloka‘i,
and to honor it as the location of the first Hawaiian Homes Settlement pro-
vides a fitting end to this section:

A he sure maoli no, ea Sure indeed


Me ke onaona, ehe With the soft fragrance
Me ka nani—Kalama‘ula With the beautiful—Kalama‘ula
‘Äina ua kaulana Land made famous
I ka ho‘opulapula Because of the homesteading or rehabilitation
Me ka nani—Kalama‘ula With the beautiful—Kalama‘ula.

Moloka‘i at the End of the Twentieth Century

As in the other districts previously discussed, life on Moloka‘i was affected by


World War II and the 1946 tidal wave. During the war, 3,000 men were sta-
tioned on Moloka‘i to defend the island from any attack. Young men left
Moloka‘i to fight in the war and to work at the lucrative jobs at Pearl Harbor.154
The tidal wave hit Kalaupapa and the area from Hälawa to Puko‘o with the
greatest force. Waves on the north shore measured fifty-eight feet. As at Wai-
pi‘o, the destruction of many homes, taro patches, and gardens led the fami-
lies of Hälawa to move to other sections of Moloka‘i.
Throughout the postwar period the economy of Moloka‘i was sustained by

242
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

two major pineapple companies, Dole and Del Monte, and by ranching. Stud-
ies of the economy of Moloka‘i from 1970 through 1994 documented the con-
tinuing importance of subsistence fishing, hunting, gathering, and farming as
part of the livelihoods and culture of the Moloka‘i kua‘äina. Moloka‘i per-
sisted as a model of a community-based island economy rooted in subsistence
where the kua‘äina pursued traditional livelihoods.
In 1970 the University of Hawai‘i Departments of Anthropology and
Geography and the School of Public Health sponsored research in human
ecology on Moloka‘i. The report, published as Moloka‘i Studies: Preliminary
Research in Human Ecology, noted that the exchange of wild food in east Molo-
ka‘i was based on the abundance of natural food resources of the area and the
frequency of interaction of the residents.155 The report noted that if tourist
activities were expanded, they would encroach on traditional gathering spots.
A resulting decline in supply of wild foods, coupled with a lessened interest in
gathering due to competing forms of entertainment and increased demands
on time, would effect a decrease in wild food exchange.
Maui county developed the Moloka‘i Community Plan in 1981 to guide
future decisions about development on the island of Moloka‘i. Included in the
plan was the East End Policy Statement, which determined that East Molo-
ka‘i should retain its rural character. It encouraged development of aquacul-
ture and restoration of the many fishponds on that part of the island.
Also in 1981, the Urban and Regional Planning Program of the University
of Hawai‘i conducted a study that examined the major values of the commu-
nity so that policy decisions about alternative energy developments could be
grounded in the residents’ preferred way of life. It was published as the Molo-
ka‘i Data Book: Community Values and Energy Development.156 The study indi-
cated that the “preferred way of life on Moloka‘i” was closely associated with
rural living, Hawaiian culture, slow pace, everybody knowing everybody, fam-
ily togetherness, and living off the land. Tourism, development, and higher
prices were inconsistent with the preferred way of life on the island.
A study of traditional Hawaiian land use in 1982 investigated the feasibil-
ity of locating a small traditional Hawaiian community on conservation lands
in the remote Pelekunu Valley on Moloka‘i.157 The elements of the project
included self-sufficiency; use of low-technology, low-energy-consumption,
labor-intensive, self-built structures; diversified subsistence farming and fish-
ing; and the maintenance of Hawaiian culture. “Residential subsistence,”
emphasizing residential use of homestead parcels while at the same time
encouraging and integrating “backyard” agriculture, was suggested in a 1983

243
chapter five

study for the Kalama‘ula Hawaiian homestead.158 A report by the University


of Hawai‘i’s Department of Urban and Regional Planning, Contemporary Sub-
sistence Lifestyles in Hawai‘i: Implication for State Policy, recommended in 1985
that the state legislature fund a more detailed study of subsistence on Molo-
ka‘i as a case study for other rural Hawaiian communities. That same year
another study, An Economic Development Strategy and Implementation Program
for Moloka‘i, recommended support for developing fisheries on Moloka‘i. It
concluded:

Fishing on Moloka‘i continues to play a strong part in the Hawaiian cul-


ture; it suits the rural lifestyle of the Island; and it accepted as an appropri-
ate type of resource use for economic development. For generations, the
Native Hawaiian population survived largely on food harvested from the
sea. The wide reef that fringes the southern shore of Moloka‘i supported
extensive subsistence fisheries, and shoreline fishponds were used to age
and fatten several species.159

The study pointed out that a lack of jobs and ready cash for groceries forces
many Moloka‘i families to depend on the ocean for subsistence. It also stated
that many Moloka‘i residents own fishing gear or boats, and the majority of
families have ready access to fresh seafood through family members, relatives,
and friends.
In 1987 the last pineapple company closed its operations. That same year a
tuberculosis epidemic led to the decision to eradicate all the cattle on Molo-
ka‘i. Moloka‘i General Hospital geared down its operations, limiting births to
delivery by midwives. At 20 percent, Moloka‘i’s unemployment rate was three
times the state’s average. Many small businesses shut down. In response to
this economic crisis, the state opened the Moloka‘i office of the Department
of Business and Economic Development and Tourism (DBEDT) in 1987 and
set up the Moloka‘i Interdepartmental Task Force. The task force noted that
“increased consideration should be given to alternate approaches supportive
of subsistence activity as an integral, preferred way of life for many Moloka‘i
residents.” A special loan program was set up to stimulate small businesses,
particularly in the areas of agriculture, fisheries, and culture. The Moloka‘i
DBEDT tried to enhance subsistence activity on Moloka‘i while introducing
mainstream economic development programs such as industrial parks, a
slaughterhouse, and an ice-making plant.
To stimulate the fishpond industry, a model project was initiated in 1989 at
‘Ualapu‘e fishpond in Mana‘e by Maui county and the state of Hawai‘i. In 1990

244
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

the Moloka‘i Office of the Department of Business and Economic Develop-


ment and Tourism sponsored development of the “Master Plan for ‘Ualapu‘e
Ahupua‘a: Blending Traditions & Technology.” The plan was designed to
protect the vital water resources of the fishpond by proposing a management
plan for the resources of the entire ahupua‘a. The nonprofit Moloka‘i-based
group Hui o Kuapä developed a plan for establishing a finfish hatchery in
December 1991.
The University of Arizona started a demonstration project in 1991 for the
cultivation of seaweed in a fishpond managed by a Native Hawaiian land trust.
Based upon its success, the program has now expanded into commercial pro-
duction and is providing training to interested community members.
In 1993 the Governor’s Task Force on Moloka‘i Fishpond Restoration
identified 74 fishponds ranging in size from less than an acre to 73 acres, with
the majority located on the southeast side. Moreover, it found that the island
of Moloka‘i, with a protected reef extending over 14,000 acres, is blessed with
a very high percentage of restorable fishponds. The task force recommended
that the state provide the money needed to establish the fish hatchery as
proposed by Hui o Kuapä. The task force also recommended that the state
appropriate additional funds for the repair of ten fishpond walls, community
training and research, and a Moloka‘i Fishpond Commission to implement a
long-term plan for the restoration and revitalization of the fishponds on
Moloka‘i. In the same year, a fishermen’s cooperative made up of forty fisher-
men who produce an estimated 60–70 percent of the island’s commercial fish
landings opened an ice house.
The Governor’s Moloka‘i Subsistence Task Force was set up, in response
to concerns raised by a group of hunters, to review all subsistence activities on
Moloka‘i, including hunting, and to make recommendations for policies to
protect and enhance subsistence on the island. Over the years, a number of
activities contributed to the degradation of the natural environment of Molo-
ka‘i. Offshore reefs and oceans were impacted by pollution, erosion, and soil
runoff from resorts, residential development, and ranching. Sand from the
west end of Moloka‘i was mined and shipped to O‘ahu to make cement to build
the freeways and hotels and to replace lost sand at Waikiki Beach. Gravel and
rocks from east Moloka‘i were used in freeway construction on O‘ahu. Ranch-
ing on the east end contributed to deforestation, erosion, and runoff. Once-
productive fishponds were allowed to fill with silt, and the walls fell into disre-
pair following tsunamis and storms. Overharvesting of marine resources relied
upon for subsistence was a growing problem. Traditional resources such as the

245
chapter five

turtle could not be used for subsistence under new federal regulations. Wild-
life such as deer, goats, pigs, and birds were abundant on privately owned lands
but were too scarce to be hunted on public lands.
Nevertheless, the study, completed in 1994, concluded that many families
on Moloka‘i, particularly Hawaiian families, continued to rely upon subsis-
tence fishing, hunting, gathering, or cultivation for a significant portion of
their food. A random sample survey of the families on Moloka‘i revealed that
28 percent of their food was acquired through subsistence activities. Among
Native Hawaiian families the survey found that 38 percent of their food was
derived from subsistence activities. The families reported receiving food
through subsistence activities at least once a week. Virtually every person sur-
veyed believed that subsistence was important to the lifestyle of Moloka‘i.160
Availability of the natural resources needed for subsistence was essential to
Moloka‘i households, where the unemployment rate was consistently higher
than on other islands and a significant portion of the population depended
upon public assistance. In March 1993, the unemployment rate of 8.1 percent
on Moloka‘i was higher than the statewide rate of 4.7 percent. With regard
to public assistance, in 1990, 24.4 percent of the Moloka‘i population received
food stamps, 12 percent received Aid to Families with Dependent Children,
and 32.5 percent received Medicaid. According to the U.S. census for 1990,
21 percent of the families on Moloka‘i had incomes that fell below the pov-
erty level of $12,674 for a family of four. The ability to supplement meager
incomes through subsistence was very important to maintaining the quality of
life of families on the island through 1994.
Subsistence was also critical to the persistence of traditional Hawaiian cul-
tural values, customs, and practices. Cultural knowledge, such as about place-
names; fishing ko‘a; methods of fishing and gathering; or the reproductive
cycles of marine and land resources were passed down from one generation to
the next through training in subsistence skills. The sharing of foods gathered
through subsistence activities continued to reinforce good relations among
members of extended families and with neighbors.
Without subsistence as a major means for providing food, Moloka‘i fami-
lies would have been in a dire situation when agribusiness phased out. Subsis-
tence provided families with the essential resources that compensated for low
incomes and a means of obtaining food items that were prohibitively costly.
Food items such as fish, lobster, crab, limu, and deer meat that are normally
obtained through subsistence were generally unavailable or were very costly

246
MOLOK A ‘ I NUI A H I NA : great moloka ‘ i

in stores. In this respect, subsistence not only provided food, it also ensured a
healthy diet critical to the prevention of illness.
A primary reason for the persistence of subsistence practices on Moloka‘i
through 1994 was the continued availability of renewable natural resources.
In turn, while years of macroeconomic strategies wreaked havoc on Hawai‘i’s
natural environment and endemic species of flora and fauna in urban areas
and on plantations, subsistence practices allowed the natural resources in rural
communities like on Moloka‘i to persist.
On Moloka‘i, subsistence was a viable sector of the economy that continued
to function alongside the sugar and pineapple plantations and the ranches.
Hawaiian extended families commonly supplemented their incomes with sub-
sistence fishing and hunting. Unfortunately subsistence was generally not rec-
ognized as a bona fide economic sector by Western economists. In the face of
economic decline in Hawai‘i, such as with the phasing out of agribusiness,
decisions were generally made that promoted new economic development
based on linear progress toward capital accumulation. This usually came in the
form of tourism.
Moloka‘i provides a rare example of how residents adapted to changing
economic circumstances without massive external intervention. Historical

Map 5 In 1993 subsistence practitioners on Moloka‘i mapped the locations of their various
subsistence activities. Source: Matsuoka, McGregor, and Minerbi, “Governor’s Moloka‘i Subsis-
tence Task Force,” pp. 3, 77.

247
chapter five

accounts indicate that when agribusiness closed on Moloka‘i, subsistence


became a more vital aspect of the economy.161 Through community-based
efforts, residents organized to successfully stave off tourism development
while promoting values related to community and family integrity. Subsistence
and other community-based endeavors were considered the forces that bound
the social elements necessary for cultural perpetuation together. Subsistence
was not a replacement economy but a tradition that survived after macroeco-
nomic strategies (i.e., plantations and ranches) failed.
Indicative of the continuing significance of subsistence to the people of
Moloka‘i at the end of the twentieth century was a rural empowerment grant
application which succeeded in attracting $2.5 million to the island over a
ten-year period. Designed by a broad cross section of the island, the vision
statement quoted at the opening of this chapter was adopted for the island’s
economic development in the twenty-first century. It continued with the fol-
lowing insights:

• We envision strong ‘ohana (families) who steadfastly preserve, protect


and perpetuate these core Hawaiian values.
• We envision a wise and caring community that takes pride in its
resourcefulness, self-sufficiency and resiliency, and is firmly in charge of
Moloka‘i’s resources and destiny.
• We envision a Moloka‘i that leaves for its children a visible legacy: an
island momona (abundant) with natural and cultural resources, people
who kökua (help) and look after one another, and a community that
strives to build an even better future on the pa‘a (firm) foundation left
to us by those whose iwi (bones) guard our land.162

The plan outlined community-based economic projects that would make


sustainable use of the islands’ resources. These included watershed protec-
tion, a native plant nursery, taro production, aquaculture development, and a
community land trust.163
The vision statement and economic development plan reflected the ongo-
ing significance of Native Hawaiian cultural values and subsistence practices
to the people of the “Last Hawaiian Island” of Moloka‘i as a major cultural
kïpuka of modern Hawai‘i. The central role of the Moloka‘i kua‘äina in regen-
erating Native Hawaiian culture during the Hawaiian renaissance will be dis-
cussed in chapter 6.

248
 six 

Kaho‘olawe: Rebirth of the Sacred

I have my thoughts, you have your thoughts, simple for me, difficult for
you. Simply . . . the reason is . . . I am a Hawaiian and I’ve inherited the
soul of my küpuna. It is my moral responsibility to attempt an ending to
this desecration of our sacred ‘äina, Kohe Mälamalama O Kanaloa, for
each bomb dropped adds further injury to an already wounded soul.
—george helm, “Reasons for Fourth Occupation of Kaho‘olawe,”
January 30, 1977

n the 1970s the island of Kaho‘olawe stirred the ancestral memory of

I Native Hawaiians and inspired the first cultural renaissance in Hawai‘i


since the islands came under American control in 1898. Throughout the
twentieth century, the United States colonized Hawai‘i through political,
social, economic, and military institutions. World War II transformed the way
of life in Hawai‘i and led to the development of new political and economic
alliances.1 In 1959 U.S. colonial policy culminated with statehood for Hawai‘i.
When Hawai‘i became a state, tourism grew to be the main industry for
the islands’ economies. American progress seemed to be overdeveloping the
islands and replacing the Native Hawaiian and local way of life. However, in
an extraordinary convergence of events, the island of Kaho‘olawe became a
focal point of a major political movement challenging American control of
Hawai‘i. Surprisingly, rather than leading to fuller assimilation of Native
Hawaiians into American society, statehood sparked a reassertion of Native
Hawaiian rights and a revitalization of Native Hawaiian language, culture, and
spirituality. The island of Kaho‘olawe served as the dynamic catalyst for this
unexpected native rights movement.
George Helm was one of the Native Hawaiian leaders whose vision shaped
the movement. He was born and raised on Moloka‘i as a kua‘äina. Together
with other young men and women from the island they roused the Moloka‘i

249
chapter six

küpuna to guide a movement to stop the bombing of Kaho‘olawe. From Molo-


ka‘i, these young men and women also reached out to kua‘äina and küpuna
from Häna and the rest of Maui, Hawai‘i, Läna‘i, and Kaua‘i to join in the
effort. Their outreach extended to Native Hawaiians at the University of
Hawai‘i and political activists on O‘ahu. However, it was the kua‘äina from
Moloka‘i and Hawai‘i’s rural communities who started the movement for
Kaho‘olawe and nurtured it into a uniquely Hawaiian movement. They suc-
cessfully involved multiple generations—from küpuna, or elders, to mäkua, or
the adult parent generation, as well as ‘öpi‘o from the university and high
schools. Kaho‘olawe became a primary site of cultural regeneration through
the aloha ‘äina projects initiated by the kua‘äina from Hawai‘i’s cultural
kïpuka. Indeed, it is through Kaho‘olawe that we can best understand and
appreciate the significant interrelationship between the knowledge and expe-
rience of the kua‘äina from the various cultural kïpuka of the islands and the
contemporary revival of Native Hawaiian culture and religion.
In March 1977 Helm and Kimo Mitchell, a kua‘äina of Häna, were mys-
teriously lost in the ocean off of Kaho‘olawe during a protest of the bombing
of Kaho‘olawe. The movement for which they had become martyrs grew into
an islands-wide movement that not only succeeded in stopping military use

Figure 32 A view of Kaho‘olawe looking over the top of the island from the southern cliffs
reveals one of the quiet moods of this sacred island. 1994. Franco Salmoiraghi.

250
k aho ‘ ol awe: rebirth of the sacred

of Kaho‘olawe in 1990, but also sparked the revitalization and impressive ren-
aissance of Hawaiian culture, music, navigation, arts, agriculture, and aqua-
culture.

Conventional History of Kaho‘olawe


Kaho‘olawe, the smallest of Hawai‘i’s major islands, consists of forty-five
square miles of volcanic craters, hills, valleys, rugged cliffs, and sandy shore-
lines, with nearshore reefs and deep ocean channels. It lies eight miles south
of the second largest island of Maui.
In the late eighteenth century goats were introduced to the island and
allowed to go feral. In the nineteenth century the Hawaiian monarchy used
the island as a penal colony to which they banished political rivals, adulterers,
and Catholic converts. In 1858 the island was leased for sheep ranching. Over-
grazing by goats and sheep destroyed the natural flora and left the island bare
and vulnerable to the erosive forces of wind and rain. Reforestation efforts in
the early twentieth century introduced noxious alien species of trees, grasses,
and shrubs to sustain cattle ranching, which further degraded the island’s
ecosystem.2
At the dawn of World War II the island was taken over by the U.S. Navy
for live-fire ordnance exercises and combat training. Kaho‘olawe came to be
called the “most shot-at” island in the Pacific. By September 1945 150 Navy
pilots, the crews of 532 major ships, and 350 Navy, Marine, and Army shore
fire control officers had trained at Kaho‘olawe. Another 730 service members
had trained in joint signal operations on the island. During the Korean War
Navy carrier planes used Kaho‘olawe to practice airfield attacks and strafing
runs on vehicle convoys and other mock North Korean targets. In 1965, dur-
ing the cold war, a one-kiloton nuclear explosion was simulated on the island
when the U.S. Navy detonated 500 tons of TNT. During the Vietnam era,
Navy and Marine Corps planes practiced attacks on surface-to-air missile sites,
airfields, and radar stations. By the time of the Gulf War, live fire training on
the island was reduced as the Navy shifted its primary training to other state-
of-the-art electronic target ranges.3

The Cultural Significance of Kaho‘olawe Is Revealed


In January 1976 Native Hawaiians staged an occupation of Kaho‘olawe as a
means of drawing national attention to the desperate conditions of Native

251
chapter six

Hawaiians.4 A bill to grant Native Hawaiians monetary reparations for the


illegal overthrow of the Hawaiian monarchy by U.S. naval forces was pending
in Congress. Two young men from Moloka‘i —Noa Emmett Aluli, M.D., and
Walter Ritte —were in the one boat that made it past the Coast Guard block-
ade and actually landed on the island. While the Navy arrested the protesters,
Ritte and Aluli remained hidden. Having stayed behind, they roamed the
island for two days before being discovered and arrested. Not only did they
witness vast destruction around the island, they also felt the presence of a deep
spiritual force. Kaho‘olawe revealed to them that it was not just a barren tar-
get island.
Seeking an explanation of their spiritual epiphany on Kaho‘olawe, Aluli and
Ritte sought out Native Hawaiian küpuna to share their memories of Kaho‘o-
lawe. They recruited Helm and drew upon the kua‘äina of Moloka‘i, Häna,
and Hawai‘i.
Moloka‘i küpuna such as Aunty Mary Lee, Aunty Clara Ku, Aunty Lani
Kapuni and Aunty Harriet Ne revealed that Kaho‘olawe had served as a refuge
for Native Hawaiian spiritual customs and practices and that it was a center

Figure 33 Moloka‘i kîpuna revealed that Kaho‘olawe was a sacred place dedicated to Kanaloa,
the Hawaiian god of the ocean. Shown here are Aunty Barbara Hanchett, Aunty Lani Kapuni,
Aunty Clara Ku, Aunty Mary Lee, and Aunty Rose Wainui (front). 1974. Franco Salmoiraghi.

252
k aho ‘ ol awe: rebirth of the sacred

for training in the arts of non-instrument navigation involving the sighting of


heavenly bodies. Uncle Harry Künihi Mitchell of Ke‘anae, Maui, shared the
chants and mo‘olelo about Kaho‘olawe he had learned from his küpuna. He
interpreted the meaning of the island’s place-names. Aunty Alice Kuloloio
and her ‘ohana spoke of their tradition of fishing, gathering limu, and subsist-
ing upon the abundant marine resources of the ocean surrounding the island.
Aunty Edith Kanaka‘ole of Hilo, Hawai‘i, advised the young men to organize
their work in stopping the military use of Kaho‘olawe in a Hawaiian manner:
as an ‘ohana for the island rather than as an association.
Through the course of this spiritual journey, an entirely new image of
Kaho‘olawe as a sacred island emerged. The Native Hawaiian küpuna revealed
that the island had originally been named Kanaloa, for the Native Hawaiian
god of the ocean. Hawaiian ancestors respected the island as a physical man-
ifestation of Kanaloa. It is the only island in the Pacific named for a major
god. Kaho‘olawe was also named Kohemälamalama o Kanaloa, which can be
translated as “the shining birth canal of Kanaloa” or “the southern beacon of
Kanaloa.” Both names link the island to its role as a traditional center in the
training of way-finding between Hawai‘i and Tahiti. Mastery of the art of nav-
igation required a complete spiritual and intellectual immersion in the natu-
ral elements of ocean, wind, currents, stars, moon, and sun. One must come
to know the nature and characteristics of the Hawaiian god of these elements,
Kanaloa, to become a master way-finder and travel across his magnificent
body form, the vast ocean. Kaho‘olawe provided the ideal location to immerse
oneself within the natural elements related to sailing and navigation that
Native Hawaiians honored as Kanaloa.

Mo‘olelo o Kanaloa
Native Hawaiians first settled Kanaloa some time around 1000 ce. They built
their homes in its valleys, fished its waters, and farmed its slopes. At Pu‘u
Moiwi they hewed stone adzes out of basalt veins at the second largest such
quarry in the Hawaiian Islands. They also crafted basaltic glass cutting tools,
carved petroglyphs, and built fishing shrines and temples to sacred deities.
Early Native Hawaiian settlers constructed sixty-nine coastal fishing shrines
around the island to mark separate fishing grounds for distinct varieties of fish
that thrive in the ocean offshore. In addition, there are numerous inland
shrines which also appear to have a connection to fishing. The ocean sur-
rounding the island has continued to be accessed by fishermen from Maui for

253
chapter six

fish, seaweed, limpets, and other forms of marine life for subsistence and
medicinal uses. The coral reefs surrounding Kanaloa are in a pristine condi-
tion relative to the reefs off heavily populated areas of our islands. There is
still a wide variety and abundance of fish and marine life in the reefs and
ocean around Kanaloa. Seabirds live in cliffs and rocky islets on the leeward
side of the island.
The following chant, “Oli Kühohonu o Kaho‘olawe Mai No Küpuna Mai”
(Deep Chant of Kaho‘olawe from Our Ancestors) was composed by early
Native Hawaiians for Kanaloa. Harry Künihi Mitchell learned it from his
küpuna. Mitchell’s ancestors settled at Honua‘ula on Maui and fished and
gathered marine life in the ocean and along the shorelines of both Honua‘ula
and Kanaloa. The family moved to Ke‘anae, Maui after a lava flow in Honua-
‘ula displaced them. Nevertheless, they continued to visit Kanaloa, gather
marine life from its shoreline, and fish in the surrounding ocean. The chant
connects Kanaloa to navigators returning from a transpacific voyage.

Wehewehe mai nei kahi ao Dawn is breaking.


Kü mai nä wa‘a kaulua Two double-hulled canoes are sighted.
Püë ke kanaka mai ka wa‘a mai The men cheer from the canoe.
Kükulu ka iwi o ka ‘äina Land is sighted.
‘Ailana Kohemälamalama To your left it is like heaven all lit up
Ho‘ohiki këia moku iä Kanaloa We dedicate this island to Kanaloa.
Akua o ka moana ‘ili, moana uli God of the shallow and deep ocean
Ke holo nei me ke au kähili We are running in an erratic current.
‘Öhaehae mai ka makani The wind is blowing from all directions.
‘Alalä keiki pua ali‘i The chief ’s child is crying [Alalä is also the
name of the channel between Kaho‘olawe
and Maui].
Ka piko hole pelu o Kanaloa The island of Molokini is shaped like the
navel of Kanaloa.
Kahua pae ‘ili kïhonua ähua The channel between Molokini-Kanaloa
and Maui Kahiki Nui is shallow.
Puehu ka lepo o Moa‘ula Dust is spreading over Mount Moa‘ula.
Pu‘uhonua mo‘okahuna kilo pae Gathering place of the kahuna classes to
honua study astronomy.
Pöhaku ahu ‘aiküpele käpili o Stone of deep magic of Keaweiki
Keaweiki
Kau lï lua ka makani ke hae nei The wind is chilly
Käwele hele nei ‘o Hineli‘i Light rain is falling

254
k aho ‘ ol awe: rebirth of the sacred

Näpo‘o ka lä i Kahikimoe The sun is setting toward Kahiki.


Nue mai ke ao Lanikau The glow after the sunset is like the colors
of the rainbow
Kapu mai ka honua küpa‘a loa The world seems to be standing still.
Pau ka luhi ‘ana o ka moana We shall no more labor on the ocean.
Mana‘o hälana pü I ke Akua My thoughts are enlightened toward God.
He aloha pili ka‘u no këia ‘äina My love for this land will always be deep
within my heart.
Aloha nö ka mana o nä küpuna I love the knowledge and power of my
ancestors.

The chant reveals four ancient names for Kanaloa: Kohemälamalama,


meaning “to your left and lit up like heaven”; Hineli‘i, “light rain”; Kahiki
Moe, “the sun sets in Kahiki”; and Kanaloa, name of the Hawaiian and Poly-
nesian god of the ocean, ocean currents, and navigation. A fifth name com-
bines two of these into Kohemalamalama o Kanaloa, “the southern beacon of
Kanaloa.” The more recent name, Kaho‘olawe, can be translated as “to take
and to embrace.”
According to Hawaiian tradition, the name Kanaloa singles out the entire
island as a wahi pana since it is the only island in the Pacific named after and
dedicated to a major Polynesian god. The name Kohemälamalama o Kanaloa
can also be interpreted as meaning “the shining birth canal of Kanaloa.” This
identifies the island as a traditional pu‘uhonua.
Creation myths for Kanaloa also reinforce its significance as a wahi pana.
The island, like Hawai‘i, Maui, Kaua‘i, Ni‘ihau, and O‘ahu, was born of Papa
and Wäkea. Two chants by composers of the time of Kamehameha give sim-
ilar accounts of the birth of the island by Papa.
The chant by Kaleikuahulu gives the following version:

Papa was weakened at the birth of the island Kanaloa.


It was born beautiful as a birdling and a nai‘a [porpoise],
It was the child born of Papa.
Papa forsook her husband and returned to Kahiki;
Returned to Kahiki she lived at Kapakapakaua.5

The chant by Pakui records the birth as follows:

Kaahea Papa iä Kanaloa, he moku, Papa was prostrated with an island,


Kanaloa,
I Hänauia he pünua he nai‘a, Who was born as a birdling; as a porpoise;

255
chapter six

He keiki ia nä Papa i Hänau, A child that Papa gave birth to,


Ha‘alele Papa ho‘i i Tahiti, Then Papa left and went back to Tahiti,
Ho‘i a Tahiti Kapakapakaua Went back to Tahiti at Kapakapakaua.6

David Malo provides a creation chant that attributes the birth of Kanaloa
to Papa and Wäkea and describes the island as being red, a traditionally sacred
color. The part of the chant referring to the birth of the island reads as
follows:

Lili-‘öpü-punalua o Papa The womb of Papa became jealous at


iä Ho‘o-hoku-ka-lani. its partnership with Ho‘ohokulani
Ho‘i hou o Papa noho iä Wäkea. Papa returned to live with Wäkea.
Hänau, o O‘ahu, he moku, Born was O‘ahu, an island,
Hänau o Kaua‘i, he moku, Born was Kaua‘i, an island,
Hänau o Ni‘ihau, he moku Born was Ni‘ihau, an island
He ‘ula ‘ä o Kaho‘olawe. Glowing fiery red was Kaho‘olawe.7

Chants of Pele, Hawaiian god of the volcano, and her family of deities rein-
force the significance of Kanaloa as a wahi pana and pu‘uhonua. Pele was born
in Kapakuela. Her husband, Wahieloa, was enticed away from her by Pele-
kumuhonua. Pele traveled in search of him. With her came the sea, which
poured from her head over the land of Kanaloa. This is said to be the first time
that the sea is brought to Kanaloa. Her brothers chant at this phenomenon:

A sea! a sea!
Forth bursts the sea,
Bursts forth over Kanaloa,
The sea rises to the hills.8

According to the rest of the chant, the sea floods the land three times, then
recedes. The floodings are called the sea of Kahinali‘i, the mother of Pele.
Kepelino provides the following account of how Pele brought the sea to
Hawai‘i at Kanaloa:

It is said that in ancient times the sea was not known here. There was not
even fresh water, but with the coming of Pe-le the sea came also. It was
thus that Hawaii got the sea. Her parents gave it to her and she brought it
in her canoes to the land of Pa-ku-e-la and thence to the land of Ka-na-loa,
and at this place she poured the sea out from her head. That is how Hawaii
got its sea. But when the sea burst forth her brothers chanted:

256
k aho ‘ ol awe: rebirth of the sacred

A sea! a sea!
The sea bursts forth,
The sea bursts forth on Ka-na-lo-a
The borders of the sea reach to the hills,
Gone is the restless sea,
Twice it breaks forth
Thrice it breaks forth,
The sea borne on the back of Pe-le.9

The brother of Pele who navigates for the family in their voyage through
the Hawaiian chain of islands is Kamohoali‘i, the principal male shark god.
There are two sites on Kanaloa that are associated with Kamohoali‘i. The first
is Lua o Kamohoali‘i, or the abyss of Kamohoali‘i. This is one of four pu‘u-
honua for Kamohoali‘i in Hawai‘i.10 It is located in a deep cave that opens
onto the ocean on the northeast side of the island. No one has explored it in
modern times. In the story of Lauka‘ie‘ie, Kamohoali‘i, and his shark people
are said to be living at Kanaloa. This is likely to have been at the site identified
on maps as Kahua Hale o Kamohoali‘i, or the house foundation of Kamoho-
ali‘i in the central portion of the island.11 Shrines to Kamohoali‘i have been
rediscovered on the cliffs above Kanapou Bay, which is a breeding ground for
sharks.
The Fornander account of Pu‘uoinaina, the legendary mo‘owahine who
lived on Kanaloa, refers to the island as a sacred land. According to the myth,
“This daughter of theirs was placed on Kahoolawe; the name of Kahoolawe
at that time, however, was Kohemalamalama; it was a very sacred land at that
time, no chiefs or common people went there.” 12
Kanaloa’s status as a wahi pana is also related to its role as a training cen-
ter in the art and science of navigating transpacific voyages using the stars for
way-finding.

Kanaloa as Training a Center for Way-Finding


Kanaloa, both the god and the island, figured prominently in the long voy-
ages between Hawai‘i and Tahiti. Moa‘ulaiki at the piko or central part of the
island was the location of a traditional training school for navigators. Moa‘ula
is a place-name associated with a place in Tahiti. There are other places in
Hawai‘i named Moa‘ula—the waterfall in Hälawa Valley on Moloka‘i; the falls,
stream, ridge, and heiau in Waikolu on Moloka‘i; a heiau in Waipi‘o Valley on

257
chapter six

Hawai‘i Island; a gulch in Ka‘ü on Hawai‘i Island; and a heiau in Kïpapa Gulch
on O‘ahu. Moa‘ula was one of the powerful kahuna priests associated with
Kanaloa. An important feature of this site is a bell stone that was broken in half
and carried to this point in two parts and put back together. The split in the
rock is oriented north to south. The ancient name of the rock is Pöhaku Ahu
‘Aiküpele Käpili o Keaweiki” (the put-together rock that kneads the knowl-
edge of the mo‘okahuna priest Keaweiki). The kahuna, Keaweiki, was associ-
ated with the school for training in astronomy and navigation at Moa‘ulaiki.
At Moa‘ulaiki are found the foundations of a platform used for the naviga-
tional school and of a house for the kahuna who instructed the students in nav-
igation. Moa‘ulaiki affords a panoramic view of the islands of Läna‘i, O‘ahu,
Moloka‘i, Maui, and Hawai‘i, all the interconnecting channels, and the cur-
rents that run through them. It was and remains an ideal site for astronomi-
cal observation in relation to the surrounding islands and channels.
Oral traditions identify Lae o Kealaikahiki as the major point of departure
for Hawaiians leaving for Tahiti in the thirteenth century. The name trans-
lates as Point of the Pathway to Tahiti. The Hawaiians probably waited here
for the ideal moon, wind, and other signs to launch their voyages to Tahiti in
the strong southerly Kealaikahiki Channel and current. Members of the voy-
aging canoe Höküle‘a estimate that they could have saved five days sailing if
they had left from here rather than from the Big Island.13 While in the Keala-
ikahiki Channel, the crew also noted that if the stern of Höküle‘a is aligned
with Pu‘u o Hökü, Moloka‘i, on the northern horizon and the north star above
it in the heavens, then the bow of the canoe is aimed directly at the point
between two stars that marks the location of Tahiti.
The Höküle‘a crew also noted that Kanaloa lies one mile north of 20 30' 0

latitude, making it the closest land mass to the latitudinal center of the main
Hawaiian Islands. Memorizing the image of the Southern Cross from sea
level near this latitudinal center is critical for navigators learning how to find
Hawai‘i on their voyage home. Thus, Kanaloa and Lae o Kelaikahiki itself
were ideal locations for the training of navigators. A platform has been
recently constructed at Lae o Kealaikahiki for the training of new generations
in non-instrument navigation.
Just above the high-water mark, inland from Lae o Kealaikahiki, is a tradi-
tional compass site made up of four large boulders. The lines formed by the
placement of the stones mark true north, south, east, and west, as has been
verified by placing a compass in the center of the stones. Jutting out from the
shoals just south of Lae o Kealaikahiki is another key navigational marker,

258
k aho ‘ ol awe: rebirth of the sacred

both traditional and contemporary. On the charts it is identified as Black Rock.


The traditional name for it is Pöhaku Kuhi Ke‘e I Kahiki (“the rock that points
the way to Tahiti”). The rock was an important marker for boats sailing along
the western side of Kanaloa because it indicated how far the shoals extended
into the channel. It mysteriously disappeared in 1984.
The legend of Mo‘ikeha, chief of Kaua‘i, who sent his son Kiha to bring
his other son, La‘amaikahiki, back to Hawai‘i places Kanaloa as the center-
piece in navigation between Hawai‘i and Tahiti. Fornander offers the follow-
ing translation of the La‘amaikahiki account:

As the place [Kahikinui, Maui] was too windy, Laamaikahiki left it and
sailed for the west coast of the island of Kahoolawe, where he lived until
he finally left for Tahiti. It is said that because Laamaikahiki lived on
Kahoolawe, and set sail from that island, was the reason why the ocean to
the west of Kahoolawe is called “the road to Tahiti.”
After Laamaikahiki had lived on Kahoolawe for a time, his priests
became dissatisfied with the place, so Laamaikahiki left Kahoolawe and
returned to Kauai. Upon the death of Moikeha [his father] the land
descended to Kila, and Laamaikahiki returned to Tahiti.14

The tradition of the migratory chief Tahitinui also refers to Kealaikahiki


as central to the voyage between Hawai‘i and Tahiti. Fornander writes:

After Hawaii Loa was dead and gone, in the time of Ku Nui Akea, came
Tahiti-nui from Tahiti and landed at Ka-lae-i-Kahiki (the southwest point
of Kahoolawe, a cape often made by people coming from or going to
Tahiti). Tahiti-nui was moopuna of Ki, Hawaii Loa’s brother, and he
settled on East Maui and died there.15

In the account of the voyage of Wahanui, a chief of O‘ahu, who went to


Kahiki, they sailed from O‘ahu to Haleolono on Moloka‘i, then on to Kau-
nolü on Läna‘i, and finally left for Kahiki by way of Kealaikahiki. According
to Samuel Kamakau:

Wahanui was a chief of O‘ahu who went to Kahiki. With him were Kilohi
the kilo, who knew the stars, Moopuaiki the kahuna, and the crewmen.
They sailed from O‘ahu and landed at Haleolono on Moloka‘i. Early in the
morning they sailed by Kaholo on Lana‘i and by broad daylight were pass-
ing Kaunolu Cape. A little to the southeast of there is Apua Cape, where
lived a man called Kaneapua . . . After repeated attempts to sail, Kaneapua

259
chapter six

was given a place on the canoe, and they sailed for Kahiki by way of
Ke-ala-i-kahiki at Kaho‘olawe.16

Peter Buck concluded through his research that Kealaikahiki was the pri-
mary departure point for voyages to Tahiti:

The point of departure for the south was the passage between Kahoolawe
and Maui which was named Ke Ala i Kahiki (The Course to Tahiti). In a
translation from Kamakau, Alexander (1891b) refers to the southern sailing
directions. Hokupaa, the North Star, was left directly astern; and when
Hokupaa sank below the northern horizon on reaching the Piko o Wakea
(the Equator), Newe became the guiding star to the south. No sailing
directions were given for the return voyage to the north.17

Two known accounts also place Kealaikahiki as a point of landing in


Hawai‘i after the long journey from Kahiki. Placing Kealaikahiki as a point of
arrival would coincide with the oral tradition related in the chant from Harry
Künihi Mitchell, “Oli Kühohonu o Kaho‘olawe Mai Nä Küpuna Mai.”
Samuel Kamakau, in Ka Nüpepa Kü‘oko‘a of January 12, 1867, wrote about the
coming of the gods. A recently published translation of the article provides
an account of the arrival of Kanaloa to Hawai‘i by way of Kealaikahiki on
Kanaloa:

According to the mo‘olelo of Kane and Kanaloa, they were perhaps the
first who kept gods (‘o laua paha na kahu akua mua) to come to Hawai‘i
nei, and because of their mana they were called gods. Kaho‘olawe was first
named Kanaloa for his having first come there by way of Ke-ala-i-kahiki.
From Kaho‘olawe the two went to Kahikinui, Maui, where they opened up
the fishpond of Kanaloa at Lua-la‘i-lua, and from them came the water of
Kou at Kaupo.18

Kamakau’s account of the legend of Kükanaloa again identifies Kealaika-


hiki on Kanaloa as a point of arrival to Hawai‘i from Kahiki:

In the wanana and pule and mele of Ka po‘e kahiko, it is said that Kukana-
loa came during the time of Kaka‘alaneo . . . It is said that Kukanaloa ma
landed in Waihe‘e from Ke-ala-i-kahiki; Kiwi was the spot where they came
ashore, and Kahahawai the place where they panted and stammered.19

The fishing resources of Kanaloa also elevate the island as a traditional


place of significance to the fishermen of the surrounding islands of Läna‘i,
Moloka‘i, and Maui.

260
k aho ‘ ol awe: rebirth of the sacred

In the Ocean of Kanaloa


The primary evidence of the rich and varied fishing resources of the waters
surrounding Kanaloa is the location of the sixty-nine fishing ko‘a, or shrines,
around the island. The first settlers may have been attracted to Kanaloa from
Maui by the fishing resources and decided to make a home there. Ko‘a were
used by fishermen to mark and develop their fishing grounds. The first fish
caught were given as offerings on the ko‘a upon the men’s returning from a
day of fishing as gratitude for the guidance of the shrine. The ko‘a serve as
land markers for ocean fishing grounds. In some cases the fish were fed at cer-
tain grounds to assure that they would be plentiful in those designated areas,
and the ko‘a serves as a land marker.
Kü‘ula, the patron of fishing, is honored at the fishing ko‘a. He is repre-
sented on the shrine as an upright stone. Pieces of coral or ko‘a are usually
placed on the shrine to represent Hina, the wife of Kü‘ula. The practices hon-
oring Kü‘ula were introduced in Hawai‘i by his son A‘ia‘i. Beckwith offers an
explanation of the Kü‘ula custom:

The god lived as a man on earth on East Maui in the land called Alea-mai
at a place called Leho-ula (Red-cowry) on the side of the hill Ka-iwi-o-Pele
(the bones of Pele). There he built the first fishpond; and when he died he
gave to his son Aiai the four magic objects with which he controlled the
fish and taught him how to address the gods in prayer and how to set up
fish altars. The objects were a decoy stick called Pahiaku-kahuoi (kahuai),
a cowry called Leho-ula, a hook called Manai-a-ka-lani, and a stone called
Kuula which, if dropped into a pool, had the power to draw the fish thither.
His son Aiai, following his instructions, traveled about the islands establish-
ing fishing stations (ko‘a) at fishing grounds (ko‘a aina) where fish were
accustomed to feed and setting up altars (kuula) upon which to lay, as
offerings to the fishing gods, two fish from the first catch.20

One of the early shrines built by A‘ia‘i in Hawai‘i was on Kanaloa at Haki-
oawa. It is described as a square-walled Kü‘ula like a heiau, set on a bluff look-
ing out to the sea.21 The following is an account of how A‘ia‘i constructed the
shrine on Kanaloa:

Thus was performed the good work of Aiai in establishing ku-ula stations
and fish stones continued all around the island of Maui. It is also said that
he visited Kahoolawe and established a ku-ula at Hakioawa, though it differs
from the others, being built on a high bluff overlooking the sea, somewhat

261
chapter six

like a temple, by placing stones in the form of a square, in the middle of


which was left a space wherein the fishermen of that island laid their first
fish caught, as a thank offering. Awa and kapa were also placed there as
offerings to the fish deities.22

In 1902 the Hawaiian-language Nüpepa Kü‘oko‘a published a series of arti-


cles titled “He Mau Kuhikuhi No Ka Lawai‘a Ana” (Fishing Lore) by A. D.
Kahaulelio, who had been trained as a fisherman by his parents and grand-
parents. Kahaulelio’s grandparents were born in Keone‘öi‘o, Honua‘ula, Maui,
and then moved to Lahaina. Kahaulelio fished inshore and in the deep sea for
a living. He and his grandparents were very familiar with the ocean around
Kanaloa. His articles in Nüpepa Kü‘oko‘a provide extensive information about
the fishing resources around Kanaloa. He described the boundary of the more
than 100 fishing grounds that he frequented between Läna‘i, Kanaloa, Uku-
mehame, and Lahaina:

From the cape of Hawea at Kaanapali running directly to the cape of


Hema on Lanai, close to Maunalei; then to the cape of Kamaiki on Lanai;
thence directly to the cape of Paki (the same as the cape of Kelaikahiki) on
Kahoolawe; thence to the cape of Kukui on Kahoolawe then straight to the
cape of Papawai are the places that are well known and have been fished in
by your writer, in sunshine, in rain and in the winds that rage and blow
into a terrific gale.23

Kahaulelio described in detail the methods used for catching some of the
various species of fish caught off Kanaloa — mälolo, weke, ulua, and uhu. He also
gathered he‘e and ‘opihi from the reefs and coast of the island. According to
Kahaulelio, mälolo were numerous at Kanaloa and sold for $20 a canoe-load
at Lahaina. He fished for weke with a net and with fishhooks. On dark nights
he did ku‘iku‘i fishing for ulua with his father along the hilly and rocky coast
of the island. They used paka eel for bait and a stout wooden pole and three-
ply olonä cord. To catch uhu near the beaches of Kanaloa, Kahaulelio used a
kind of bamboo pole that was also used to catch aku, with hä‘uke‘uke, wana,
and ina sea urchins as bait. According to Kahaulelio, the fishing ground on the
seaward side of Kealaikahiki called Laepaki was one of the most productive of
the three deep-sea fishing grounds of Kanaloa.24 Kahaulelio wrote at length
about the big ‘opihi makaiauli of Kanapou Bay:

It is at that large stream facing Honuaula. The opihi are as large as the
bowls found in shops, not large ones, but the smaller ones. Goat meat could

262
k aho ‘ ol awe: rebirth of the sacred

be boiled in opihi shells and the twenty-five cents worth of beef bought in
Lahaina could be cooked entirely in the opihi shells of that locality, not the
opihi dived for, but that which clung to the sea cliffs. Your writer was there
for a week without vegetable food, living only on water, fish, opihi and goat
meat. That is how I discovered that that was the place of large opihis.25

In his articles, Kahaulelio related the legend of Pu‘ui‘aiki, who left Kohala
on a small canoe and was swamped in the middle of the ‘Alenuihähä Channel.
Feeling sorry for Pu‘ui‘aiki, the prophet Moa‘ula, who lived at Moa‘ula on
Kanaloa, sent the opihi makaiauli to rescue him. After grasping the ‘opihi,
Pu‘ui‘aiki was swallowed whole by a shark. Pu‘ui‘aiki used the ‘opihi to scrape
away at the flesh of the shark, and after three nights and days the shark landed
at Kanapou Bay and died. Pu‘ui‘aiki came out of the shark, his head now bald
and shiny from having been inside the shark, and rested on the beach. He sur-
vived an attempt by some fisherman of the island to stone him and was rescued
by the prophet Moa‘ula. The story provides a vivid description of the marine
life, the native plant vegetation, and the freshwater sources around Kanapou
Bay and reinforces Kahaulelio’s own experience of being able to live on the
island for a week relying upon those same resources.26

The Sea and the Land


There are other place-names on the island that identify additional marine
resources utilized by Hawaiians who lived on the island. Honokanai‘a is the
traditional name for what is called Smuggler’s Bay. It means “the dolphin har-
bor.” Dolphins are frequently observed playing in the offshore waters of this
bay. A bay between Honokanai‘a and Kealaikahiki near what is called Smug-
gler’s Bay is named Honukanaenae, meaning “tired turtle.” This was where the
turtles came to rest and to lay their eggs. It is not currently used as a nesting
spot by turtles, probably because of the military encampment located in the
vicinity.27
Ko‘ele Bay is said to refer to a variety of large tough ‘opihi.28 The eastern-
most and westernmost points of Kuhe‘eia Bay are the sites of fishing ko‘a and
bear names referring to the white hilu fish and the red hilu fish, respectively.
The eastern point is Laehilukea, “white hilu point.” The western point is Lae-
hilu‘ula, “red hilu point.” Kuhe‘eia itself means “squid grounds.” 29
Laepuhi means “eel point.” Laeokuaka‘iwa means “the point where the
frigate bird roosts.” Nälaekoholä, the name of the two outermost points of
Ahupü Bay, means “humpback whale points.” 30

263
chapter six

An islet off of the southern coast of Kanaloa is called Pu‘u Koa‘e, meaning
“hill of the tropical bird.” Seabirds that feed daily off deep-sea fish guide fish-
ermen to schools of fish in the open ocean and can be followed back to land-
falls.
Between 1400 and 1600 ce, Native Hawaiians opened agricultural plots
inland and planted sweet potatoes and dry-land crops. By the middle of the
seventeenth century, the northeast portion of the island at Hakioawa emerged
as the largest settlement on the island. The people built many house sites, two
major heiau, and several ko‘a there. The population began to decline in the
eighteenth century due to inter-island warfare and the introduced diseases that
had affected the other Hawaiian Islands. In particular, High Chief Kalani-
opu‘u of Hawai‘i invaded Kanaloa in one of his battles against High Chief
Kahekili for control over Maui, Läna‘i, and Kanaloa and many of the resi-
dents perished. In 1793 Captain George Vancouver gave Kahekili a gift of
goats, which he sent to Kanaloa to graze and multiply. The goats ultimately
reproduced into thousands of animals that roamed the island, denuded it of
its vegetation, and caused severe erosion and destruction of the island’s natu-
ral resources.

Kanaloa and Aloha ‘Âina


The contemporary rediscovery of Kaho‘olawe as a sacred island dedicated to
Kanaloa led to a revival of the traditional Hawaiian value of aloha ‘äina, or
love and respect for the land. Ancestral memories of the küpuna focused upon
aloha ‘äina as the Hawaiian value at the core of traditional spiritual belief and
custom.
According to the küpuna, Native Hawaiians respect, treasure, praise, and
worship the land and all natural elements as deities and the source of univer-
sal life. At one level, family genealogies link contemporary Hawaiians to
astronomers, navigators, planters, fishermen, engineers, healers, and artisans.
These ancestors settled Hawai‘i and constructed great walled fishponds, irri-
gated taro terraces, dry-land agricultural systems, heiau, and family settle-
ments. At a deeper level, beyond these human forebears, ancestral chants trace
Hawaiian origins to such great gods as Papa Hänaumoku, the earth mother
and birth mother of the Hawaiian Islands; Wäkea, the sky father; Käne, the
springs and streams; Kanaloa, the ocean; and Pele, the volcano. Hawaiians are
genealogical descendants of the earth, sea, sky, and natural life forces.
Acknowledgment of such ancestry thus places a responsibility upon con-

264
k aho ‘ ol awe: rebirth of the sacred

temporary Native Hawaiians to protect the land and all of its resources in
one’s lifetime and for the lifetimes of future generations. This is the moral
responsibility that George Helm spoke of in the epigraph to this chapter. In
addition, aloha ‘äina embodied several layers of responsibility. At one level, it
meant protecting the physical sustainability of Hawaiian lands and natural
resources. At another level, it meant organizing and rallying for Hawaiian
native rights and sovereignty to achieve the political standing necessary to
protect the ‘äina. At the deepest level, it meant a spiritual dedication to honor
and worship the gods who were the spiritual life of these forces of nature.
Gradually, the movement that developed to protect Kanaloa from contin-
ued military bombing and combat exercises grew to be more pro-Hawaiian
than anti-American. It embraced the environmental, political, and spiritual
meaning and practice of aloha ‘äina.
For fourteen years, from 1976 through 1990, the Protect Kaho‘olawe
‘Ohana led Native Hawaiians and the general public in protests to end the des-
ecration of Kaho‘olawe/ Kanaloa. A series of illegal occupations of the island
led to arrests and lengthy and expensive court defenses. Members were sen-
tenced to imprisonment or were barred from ever returning to the island. In
some instances ‘Ohana members were ostracized by family, friends, and the
broader community for their activism. The hardest loss was the tragic disap-
pearance of George Helm and Kimo Mitchell, in March 1977, in the waters
surrounding the island during their protest of the bombing.
In 1980, as the result of a civil suit filed in 1976 by Helm, the Protect
Kaho‘olawe ‘Ohana reached an out-of-court settlement with the U.S. Navy,
called a consent decree. The Navy was mandated to conform to the National
Historic Preservation Act and to survey and develop a plan to protect historic
sites, complexes, and features on the island. Under the Environmental Pro-
tection Act, the Navy was mandated to stop bombing the island for ten days
of each month, to limit their bombing and shelling to the central third of the
island, to clear two-thirds of the island of surface ordnance, to eradicate the
feral goats, and to begin soil conservation and revegetation programs. In com-
pliance with the American Indian Religious Freedom Act of 1978, the Protect
Kaho‘olawe ‘Ohana was acknowledged to be Ke Kahu O Ka ‘äina or Steward
of the Land and allowed access to the island for religious, cultural, and educa-
tional activities for four days in ten months of each year. This served as a crit-
ical turning point in the struggle to restore Kanaloa to the people of Hawai‘i.
Several members of the ‘Ohana criticized the consent decree as an unaccept-
able compromise because it meant joint use of the island with the U.S. Navy

265
chapter six

and therefore an acknowledgement of their presence and use of the island. In


protest, they resigned from the organization. Those who remained in the
‘Ohana viewed the consent decree as an interim measure to relieve the island
of full-scale, year-round bombing. Isolating the bombardment to a third of
the island meant that two-thirds of the island could begin the healing process,
being cleared of surface ordnance and replanted with grasses and trees.
Under the consent decree, the ‘Ohana began to resettle and heal Kanaloa.
An average of sixty participants were taken to the island each month to work
with the ‘Ohana on erosion control and revegetation projects. The ‘Ohana
established a permanent base camp on the northeast side of the Island at
Hakioawa as well as three temporary camps along the north side at Kühe‘eia
and Ahupü and on the west side at Keanakeiki. Hiking trails were cleared,
water catchments installed, and soil conservation and revegetation projects
initiated. Ancestral shrines and temples were rededicated, and new cultural
sites, such as a traditional meeting house, a hula platform, and a memorial for
küpuna who had passed on, were established. Beginning in 1982 the ‘Ohana
revived the annual celebration of the Makahiki or harvest ceremonies in honor
of Lono, the Hawaiian god of agriculture. Strategically, the consent decree
established the ‘Ohana as the steward of Kanaloa and significantly expanded
the network of support for permanently ending all military use of the island.
Within the framework of this mo‘olelo, the relevance of these develop-
ments is that these various projects and the manner in which the Protect
Kaho‘olawe ‘Ohana conducted its work were designed and guided by kua‘äina
from Maui, Läna‘i, Kaua‘i, Moloka‘i, and Hawai‘i. Harry Mitchell, the Küka-
hiko ‘ohana, and the Lind ‘ohana of Maui shared their knowledge of the bays,
points, currents, and channels surrounding Kanaloa and how to safely navi-
gate them. They also guided the ‘Ohana in how to anchor in the bays and land
upon the shores of the island that lack moorings, piers, or docks. Their knowl-
edge of the fishing grounds and marine resources was utilized in securing food
for those who ventured to Kanaloa to help stabilize its natural and cultural
resources. The Maui kua‘äina provided guidance on how to prepare to enter
the ocean surrounding Kanaloa and how to pack to survive on the island.
Everything brought to the island is floated through the ocean and landed by
swimming it to participants who form a line from the shore out into the waves.
Those who go to Kanaloa bring sufficient supplies of fresh water and cook
their food on propane burners and in traditional imu. It is an amazing experi-
ence in living with the ‘äina, one that transforms the lives of those who make
the open ocean crossing. Harry Mitchell, who endured the ultimate sacrifice

266
k aho ‘ ol awe: rebirth of the sacred

of losing his son Kimo for the life of Kanaloa, walked the island with ‘ohana
members, pointed out each significant cultural site, explained its purpose and
function, interpreted the name given to it by our ancestors, and shared the
mo‘olelo associated with it. These were recorded in video and produced by
the ‘ohana to educate the wider public and new generations of students about
the historical cultural importance of Kanaloa.
Maui families also shared their knowledge of the historic connection of
Haleakalä on Maui to Kanaloa. Harry Mitchell spoke of an underground lava
tube that connects Kanaloa to Haleakalä in the same way an umbilical cord
connects a fetus to its mother and noted that the island is shaped like a fetus.
The küpuna described a long cloud, Keaoloa, that formed daily above the for-
ests of Ulupalakua on the slopes of Haleakalä and extended across the Alalä-
keiki Channel to Moa‘ulanui on Kanaloa, across the island and the Kealaika-
hiki Channel to Läna‘i, and across the Kaiokalohi Channel to Ho‘olehua on
Moloka‘i. This cloud was the principal source of rain for the island of Kana-
loa throughout most of the year, when the northeast trade winds prevailed.
When the forests on Haleakalä were cleared for pasture, the cloud stopped
forming over ‘Ulupalakua, and the rain stopped falling on Kanaloa for the
greater part of the year.

Figure 34 The annual Makahiki ceremony on Kaho‘olawe includes a procession to the top of
the island to open the ceremonies in November and a procession across the island to close the
ceremonies in January. 1987. Franco Salmoiraghi.

267
chapter six

‘Ohana of Läna‘i provided guidance in how to cross the Kealaikahiki Chan-


nel to the western shores of Kanaloa from their island. They shared their expe-
rience and knowledge about restoring the dry-land forests and soils of Läna‘i,
which of those of all the islands are the closest in type and condition to those
of Kanaloa. The dry-land forest at Kanepu‘u on Läna‘i remains a model for
the restoration of Kanaloa’s flora and natural resources.
‘Ohana of Kaua‘i shared their traditions and knowledge of fishing and gath-
ering and the transmigrations throughout the island chain that connected
Kaho‘olawe to the families of all of the islands.
The kua‘äina of Moloka‘i organized the groups who visited to work as an
‘ohana in practicing the values of aloha kekähi i kekähi, or love and respect for
each other, and laulima, or cooperative work. They organized the work proj-
ects in fulfillment of aloha ‘äina and mälama ‘äina, or care for the land. Hui
Ala Loa, which was the predecessor and parent organization of the Protect
Kaho‘olawe ‘Ohana, served as a model in how to organize multiple genera-
tions — both the mäkua and the ‘öpi‘o — under the leadership of the küpuna.
Through the leadership of Noa Emmett Aluli and Colette Machado of
Moloka‘i, the ‘Ohana remained rooted in the founding principles of aloha
‘äina, küpuna leadership, ‘ohana or collective decision making and action,
accountability to the grassroots kua‘äina, and a clear focus on healing and
restoring the life of Kanaloa itself. The ‘Ohana navigated the treacherous
political times and steered clear of centering the organization in Honolulu,
where opportunistic Native Hawaiian leaders desired to use Kaho‘olawe to
advance their own political agendas and the abstract goal of political sover-
eignty and independence. The ‘Ohana continued to follow the philosophy of
George Helm —“follow your na‘au, but do your homework”—which made it a
committed organization of political action guided by an informed and sophis-
ticated strategy. Moreover, the Moloka‘i kua‘äina kept alive the memory and
vision of George Helm for Kaho‘olawe to be regreened and restored as a pu‘u-
honua for the Hawaiian culture, so that the sacrifice of Helm’s life would not
be in vain.
Before George Helm died he and Aluli founded the Kohemälamalama o
Kanaloa/Protect Kaho‘olawe Fund. Together, Aluli and Machado linked the
Protect Kaho‘olawe Fund (PKF) to national funding groups supportive of
Native American culture, land rights, and sovereignty, among them the Sev-
enth Generation Fund, the Tides Foundation, and the Gerbode Foundation.
The Protect Kaho‘olawe Fund board included representatives of grassroots

268
k aho ‘ ol awe: rebirth of the sacred

communities on every island and therefore did more than raise money for the
work of stopping the bombing of Kanaloa and restoring its cultural and nat-
ural resources. The fund also helped establish and fund grassroots organiza-
tions engaged in aloha ‘äina land struggles on Kaua‘i, Maui, Läna‘i, Moloka‘i,
and Hawai‘i, continuing to expand the network of support for Kanaloa.
In the 1980s the PKF helped the Häna Pöhaku on Maui raise funds for self-
sufficiency projects in taro cultivation and fishing, research into protecting
their land from federal condemnation for a national park, and protection of
their water rights from diversion for hotels. The Hui Ala Nui o Makena on
Maui was assisted by the PKF in the research and legal work to keep access to
the Makena coastline open for fishing and gathering by local people.
On Kaua‘i, the fund assisted the Niumalu-Nawiliwili Tenants Association
to develop an alternative land use plan that included their new homes. For
Läna‘i, monies were raised for research of kuleana lands and water concerns.
Hui Ala Loa on Moloka‘i received assistance for its litigation and organizing

Figure 35 On October 22, 1990, President George H. W. Bush stopped the bombing and all
ordnance delivery exercises on Kaho‘olawe. The amphibious exercises shown here did not
involve ordnance and were the last military exercises conducted on the island in 1993. 1993.
Franco Salmoiraghi.

269
chapter six

work to protect that island’s cultural, natural, and agricultural resources from
overdevelopment.
On the island of Hawai‘i, the PKF assisted the Mälama Ka ‘Äina Häna Ka
‘Äina community organization to get monies to develop a plan to settle Hawai-
ian Home Lands at King’s Landing by families desiring to pursue traditional
Hawaiian subsistence livelihoods instead of building standard residential
houses on lots. Ka ‘Ohana O Kalae at South Point worked with the PKF to
receive monies for a community curatorship program to protect the historic
sites and rich natural resources of the Ka‘ü district from industrial develop-
ment. Efforts of Pele practitioners to protect her and the Kïlauea volcano and
rainforest from geothermal development were initially funded by grants to
the Protect Kaho‘olawe Fund until the Pele Defense Fund branched out from
the PKF.
In summary, the Protect Kaho‘olawe Fund, under the leadership of Aluli
and Machado, served as a launching pad for much of the practical work of
aloha ‘äina in the 1980s and the early 1990s. The work to protect Hawaiian
ancestral lands from development; to perpetuate traditional Hawaiian spiri-
tual relationships to the land, including religious ceremonies; and to continue
traditional subsistence economic activities was established and persisted and
grew through the efforts of the various community-based organizations affil-
iated with and supported by the Protect Kaho‘olawe ‘Ohana and Fund.
Kua‘äina from Hawai‘i organized the building of a hale halawai or meeting
house, bringing in ‘öhi‘a logs and grass for thatching from the forests and pas-
tures of their island. Fishermen from Hawai‘i brought their skills to provide
subsistence for people while they were staying on Kanaloa. The Kanaka‘ole
‘ohana of Hawai‘i provided the chants and the ceremonies to celebrate the
Makahiki and to reopen cultural sites, as well as to build new sites and dedi-
cate them.
The reestablishment of the Makahiki and other Native Hawaiian cultural
and religious ceremonies and practices on Kanaloa was the most significant
outcome of the movement to stop the bombing of Kanaloa. These ceremo-
nies and practices reconnected a generation of Native Hawaiians with their
ancestors and their soul as a people. The revival of these religious ceremonies
deserves special attention. It was inspired and guided by Edith Kanakaole, her
daughters Nalani and Pualani, her son-in-law Edward Kanahele, her son Par-
ley Kanaka‘ole, and the Edith Kanakaole Foundation of the island of Hawai‘i
that was founded in her honor.

270
k aho ‘ ol awe: rebirth of the sacred

Rebirth of the Sacred


From the outset, Helm and Aluli followed the guidance of the küpuna who
counseled them by acknowledging and including the ancestral spirits of Kana-
loa in the effort to stop the bombing and heal the island. Kahuna Sam Lono
and Emma DeFries of O‘ahu conducted a ceremony in 1976 at Hakioawa to
ask permission of the ancestral spirits of the land to open the religious sites
on the island to receive ho‘okupu or offerings. In 1979, John ‘Änuenue Ka‘imi-
kaua of O‘ahu and Moloka‘i and his hälau hula or school of hula conducted a
ceremony to give life to the land by burying offerings of food in the ground
and dances of hula kahiko. Papa Paul Elia of Moloka‘i offered a prayer to the
ancestral gods for strength, organization, and protection of the land. At that
time Aunty Emma DeFries did a ho‘uwë‘uwë or lamentation chant for the ‘äina
acknowledging the neglect of the island which caused its devastation. Other
küpuna who committed their mana for the respect and return of the ancestral
spirits and the Native Hawaiian gods of nature to Kanaloa included Aunty

Figure 36 George Helm helped found the movement to stop the bombing of Kaho‘olawe, for
which he gave his life a few months after this photo was taken. The movement to stop the
bombing of Kaho‘olawe sparked the Hawaiian renaissance and the movement for Native Hawai-
ian sovereignty. 1977. Bishop Museum.

271
chapter six

‘Iolani Luahine, Uncle Sam Hart, Aunty Luka Naluai, Uncle Henry Lindsey,
and Aunty Gardie Perkins.31 In 1981 the ‘Ohana asked Aunty Edith Kanaka-
‘ole and her daughter Nalani Kanaka‘ole of Hälau o Kekuhi to train them in
how to conduct a Makahiki ceremony. The ‘Ohana wanted to place the heal-
ing and regreening of the island under the care of Lono, Hawaiian god of
agriculture and productivity.
Hälau o Kekuhui and the Edith Kanaka‘ole Foundation are the most influ-
ential force in the revitalization of sacred ceremonies and rituals on Kanaloa
and in contemporary Hawai‘i as a whole. Aunty Edith Kanaka‘ole was trained
in the hula by her mother, Mary Ahi‘ena Kanaele Fujii, who was born in the
1880s and raised in the hula kapu or sacred hula in the Puna district of
Hawai‘i.32 Hälau o Kekuhi, according to its Web site, “is celebrated for its
mastery of the ‘Aiha‘a style of hula and chant . . . a low postured, vigorous,
bombastic style of hula which springs from the eruptive volcano persona, Pele
and Hi‘iaka.” 33 It is “a traditional classical dance company” rooted in “seven
generations of family practitioners” and leaders in hula and oli. The Edith
Kanka‘ole Foundation was founded in the summer of 1990. Its purpose is to
heighten indigenous Hawaiian cultural awareness and participation through
educational programs that maintain and perpetuate the teachings, beliefs,
practices, philosophies, and traditions of Edith and Luka Kanaka‘ole and their
ancestors, including Edith’s mother, Ahi‘ena, her great-uncle Lonokapu, and
Luka’s father and mother, Ioana Kanaka‘ole and Haleaka Kaleopa‘a.
In January 1982 the Protect Kaho‘olawe ‘Ohana conducted what may have
been the first public Makahiki ceremonies in honor of the Akua Lono since
High Chief Kekuaokalani conducted the Makahiki ceremonies before going
into battle in defense of the Hawaiian religion in 1819, in the year of the ‘Ai
Noa or Abolition of the Kapu. The purpose of the ceremonies was to attract
the akua, Lono, to Kanaloa in the form of rain clouds to soften the earth and
ready it to receive young plants to revegetate the island. Every year since 1982
the ‘Ohana has opened the Makahiki season in November after the appear-
ance of the Makali‘i or Pleiades constellation on the horizon at sunset and has
closed the Makahiki season in January or late February. Edith Kanaka‘ole and
Nalani Kanaka‘ole prescribed the ho‘okupu, ten offerings including various
kino lau or body forms of the akua. They advised the ‘Ohana in the crafting
of an akua loa or image of Lono. Nalani Kanaka‘ole composed the chants of
prayer to Lono. The Edith Kanaka‘ole Foundation described the central
chant:

272
k aho ‘ ol awe: rebirth of the sacred

This Lono chant concentrates on the kinolau or body forms of Lono which
are the manifestations that encourage growth. The prayer is a formula used
in many traditional chants that is; recognizing and addressing the great
Gods of the elements, followed by an account of their creations, then an
enumeration of offerings, a statement of the body forms of the deities and
finally the reason for the prayer. The need in this case is to ensure vege-
tation and growth on the island. The very last line releases the formal
communication with the God. This is the FIRST formal prayer chant
composed for a formal modern day Makahiki ceremony.34

In May 1986 Pualani Kanaka‘ole Kanahele and Edward Kanahele were


asked by the Protect Kaho‘olawe ‘Ohana to design a ceremony for the Akua
Kanaloa. The biennial RIMPAC naval exercises were scheduled to culminate
with the joint ship-to-shore shelling of the island by U.S. and Canadian naval
forces after forty-five days of joint naval maneuvers from California to
Hawai‘i. After years of protest, the ‘Ohana resolved to enlist Kanaloa himself
in the effort to protect his kino lau from the bombing and in the effort to
restore the island to the people of Hawai‘i. The ceremony was designed to be
small and private. The central chant asked Kanaloa to give strength and skill
to those united in the goal of protecting and giving life to the island. The
ho‘okupu of he‘e (octopus), a kino lau of Kanaloa, could not be eaten by those
involved in the ceremony. This ceremony provided focus and inspiration to
those involved in the ongoing work to stop the bombing and restore the life
of the island.
In 1982 Kumu Hula Hokulani Holt Padilla of Maui, a member of the Pro-
tect Kaho‘olawe ‘Ohana, decided to build a pä hula or hula platform at Hakio-
awa so there would be a formal arena for the hula practices on Kanaloa. Over
the course of five years, many people who came to Kanaloa on an access visit
with the Protect Kaho‘olawe ‘Ohana contributed their aloha and hard work
to build up the platform. During the opening of the Makahiki in November
1987, the pä hula was dedicated to Laka and named Ka‘ie‘ie in a ceremony led
by Kumu Hula Hokulani Holt Padilla with the participation of Kumu Hula
Pualani Kanaka‘ole Kanahele and Kumu Hula Keali‘i Reichel.
A chant composed by Pualani Kanahele for the healing ceremony described
below, “He Ko‘ihonua No Kanaloa He Moku,” chronicled the origin and his-
tory of the island and provides a poetic summary of the process of reviving
religious ceremonies on the island:

273
chapter six

Ua ala Hawai‘i mai ka moehewa nui The Hawaiian woke from the
nightmare
Ho‘omaopopo i ke keiki i‘a a Papa Remembered was the child of Papa
‘O Kanaloa O Kanaloa
Ke moku hei Haumea The sacred land of Haumea
‘O Kohemälamalama O Kohemälamalama
Ke Kino o Kamohoali‘i The body form of Kamohoali‘i
E ho‘öla käkou iä Kaho‘olawe Save Kaho‘olawe
Ola i ka lani a Käne To live in the heavens of Kane
Ola i ke kai a Kanaloa To live in the sea of Kanaloa
Ua Kahea‘ia ‘o Lono i ka makahiki hou Lono was summoned for a new year
Ma ka Hale Mua o Lono i kahea ‘ia ai At the Hale Mua of Lono, he was
called,
Ua Kanaloa ‘o Kanaloa i Kanaloa was reconfirmed to
Kohemälamalama Kohemälamalama
Puka hou a‘e ka mana o Kanaloa The energy of Kanaloa was revitalized
Ua kani ka leo pahu i ka Mälama o Hoku The voice of the drum sounded in the
care of Hoku
Kuwawä i ka houpo a Laka Resounding in the bosom of Laka
Ua ala ‘o Laka ma Ka‘ie‘ie i Kanaloa Laka awoke at Ka‘ie‘ia at Kanaloa.

The revival of sacred ceremonies on Kanaloa is what distinguished the Pro-


tect Kaho‘olawe ‘Ohana as a pro-Hawaiian and cultural organization. As the
‘Ohana and its political struggle and practical restoration efforts evolved, cul-
tural protocol and religious prayer proved to be an essential element in shap-
ing and defining their efforts. Ultimately, perseverance and this holistic
approach proved successful.

The Bombing Stops


Finally, on October 22, 1990, after a decade of persistent, dedicated, and
focused work for Kanaloa under the consent decree, the bombing of Kaho‘o-
lawe stopped. The Democratic senator from Hawai‘i, Sparky Matsunaga,
passed away in office in April 1990. The National Republican Party and Pres-
ident George Bush himself urged Congresswoman Patricia Saiki to run for the
deceased senator’s seat as part of the national campaign to win a Republican
majority in the U.S. Senate. According to Saiki’s Native Hawaiian campaign
manager, Andy Anderson, Bush asked the congresswoman what it would take
for her to get elected. She said it would take a miracle for her as a Republican

274
k aho ‘ ol awe: rebirth of the sacred

to get elected from the Democratic state of Hawai‘i. Asked what such a mira-
cle might be, Anderson and Saiki’s campaign staff suggested that stopping the
bombing of Kaho‘olawe would win her the support of the general public in
her bid for the U.S. Senate.35 In an interview on November 11, 2003, at the
time the U.S. Navy transferred control of access to the island, Saiki described
her conversation with the president:

He said, “What can I do for you to give you a hand here, to help with the
state and get your election looked at positively?” I said, “No. 1, you’ve got
to stop the bombing of Kahoolawe. It is an island that has been devastated
by the impact exercises. Although the exercises are very worthy, it is an
assault and an insult to the Hawaiian people.” 36

Bush directed the Secretary of Defense to immediately discontinue use of


the island for bombing and target practice a day before he set out from Wash-
ington to campaign for the congresswoman in Hawai‘i.37 Jet bombers sched-
uled to make a bombing raid on Kaho‘olawe that morning were grounded.
In order to keep his competitive edge in the race for the U.S. Senate seat,
Saiki’s challenger, Democratic Congressman Daniel Akaka, worked with Dem-
ocratic Senator Daniel Inouye to do more for Kaho‘olawe/ Kanaloa than Saiki
and Bush had done. In November 1993 the U.S. Congress passed and Presi-
dent Bill Clinton signed an act that recognized Kaho‘olawe/Kanaloa as a
national cultural treasure and stopped the use of Kanaloa for any military
training for two years and 120 days. It also established the Kaho‘olawe Island
Conveyance Commission to make recommendations for the future use of the
island. This was a more significant and permanent measure. A presidential
memorandum could be rescinded or overridden at any time in the future,
unlike a Senate bill that had been passed into law. This shifted the balance of
support in favor of Congressman Akaka, who was first elected to the U.S. Sen-
ate in November 1990. The Conveyance Commission recommended that
title to the island be turned over to the state of Hawai‘i and that the Congress
appropriate $400 million to conduct a ten-year omnibus ordnance clean-up of
the island. On May 9, 1994, the U.S. Navy formally returned the island to the
state of Hawai‘i in ceremonies at Palauea, Maui, across the channel from
Kanaloa. Under a specific Hawaiian law, Hawai‘i Revised Statutes Chapter
6K, the island will eventually be turned over to a sovereign Hawaiian entity.
It states in part, “The resources and waters of Kanaloa shall be held in trust
as part of the public land trust; provided that the State shall transfer manage-
ment and control of the island and its waters to the sovereign native Hawaiian

275
chapter six

entity upon its recognition by the United States and the State of Hawai‘i.” 38
This measure set a precedent for Native Hawaiian sovereignty in that the
state of Hawai‘i acknowledges that there will be a sovereign Native Hawaiian
entity and that repatriated federal lands can be part of the land base of the
sovereign entity.
A unique feature of the approach taken by the Kaho‘olawe Island Con-
veyance Commission to arrive at its recommendations and complete its final
report was a special cultural ceremony held in August 1992 at Hakiowa. “E
Kaho‘olawe, E Ho‘omau Ana Hou I Ka Mauli Ola,” a healing ceremony for
the land, ocean, and people of Kaho‘olawe, was the name and purpose of the
event. It was organized by the Edith Kanaka‘ole Foundation in coordination
with the Protect Kaho‘olawe ‘Ohana and the Conveyance Commission. The
central feature was the construction of a mua or memorial platform to honor
the küpuna who had dedicated their lives to the healing of the island. Govern-
ment leaders from the federal, state, and county governments and from the
Office of Hawaiian Affairs were invited to sit on the mua with küpuna from
each island and the leaders of the Protect Kaho‘olawe ‘Ohana. Each was served
the ceremonial drink ‘awa by Parley Kanaka‘ole, the kahu or leader of the cer-
emony, and in receiving the drink each was asked to make a decision and com-
mit to doing whatever was in their power to heal the island of Kaho‘olawe.
The Edith Kanaka‘ole Foundation composed special chants to open the cer-
emony at dawn and to acknowledge the genealogy of the decision makers and
küpuna upon their stepping onto the mua, a genealogical history of the island,
and chants in honor of George Helm and Kimo Mitchell. The opening dawn
chant, called “E Ala E,” has become a popular chant throughout the islands
as a protocol for starting the day in a Hawaiian frame of mind.
This cultural ceremony affirmed support for the recommendation of the
commission to Congress to permanently end military use of the island, turn
title over to the state of Hawai‘i, and appropriate $400 million to clear the
island of ordnance and begin the healing and restoration of the island’s cul-
tural and natural resources.

The Native Hawaiian Cultural Renaissance


Kanaloa also sparked a broader cultural revival, one that reverberated
throughout the Hawaiian Islands from 1976 through the 1990s and chal-
lenged other institutions of American colonization.

276
k aho ‘ ol awe: rebirth of the sacred

A general cultural renaissance developed as the number of hälau hula or


schools that teach traditional Hawaiian dance and chant expanded and the
number of dancers of both Hawaiian and non-Hawaiian ancestry increased.
Hawaiian music gained new popularity and new songs, styles, and rhythms
were created. Lä‘au lapa‘au, traditional herbal and spiritual healing practices,
were recognized as valid and significant. Traditional Hawaiian healers began
to train a new generation in the Hawaiian healing arts.39
Of international significance, traditional navigational skills were revived
through transpacific noninstrument navigation in traditional Hawaiian sailing
canoes such as the Höküle‘a, the Hawai‘i Loa, and the Makali‘i were accom-
plished. These voyages spread the Hawaiian cultural renaissance throughout
the Pacific—from the Marquesas to Tahiti and the Cook Islands, New Zea-
land, Australia, and Rapa Nui (Easter Island).
As Native Hawaiians from all of the islands continued to visit Kanaloa and
to become involved in the struggle to stop military use of the island, they also
carried back to their home islands the message of aloha ‘äina. Where resort or
industrial developments threatened to expand into the rural districts that had
served as the last strongholds of Native Hawaiian custom, belief, and practice,
communities began to organize to protect their lands and natural resources
and Hawaiian way of life, many with the support of the Protect Kaho‘olawe
Fund.
Native Hawaiians on Moloka‘i continued to organize as Hui Ala Loa to
stop resort development that threatened to divert limited water resources away
from community-based economic development and to destroy subsistence
resources.
Native Hawaiians in Ka‘ü organized against plans to launch rockets from
Hawaiian Home Lands at South Point and to develop a spaceport at Kahili-
pali and Palina Point. They argued that these massive projects would destroy
Native Hawaiian cultural sites in the district, bring in newcomers, and trans-
form the Native Hawaiian way of life.
As discussed in chapter 4, practitioners of the Hawaiian volcano goddess
Pele on the island of Hawai‘i organized against geothermal energy develop-
ment in the Wao Kele o Puna rainforest of Puna. Industrialization of the vol-
cano threatened to destroy the largest expanse of lowland tropical rainforest
in the United States. The Pele practitioners asserted that geothermal energy
would desecrate and destroy the life force and manifestations of their deity.
Native Hawaiians of Maui organized against sprawling resort development

277
chapter six

that blocked access to the beaches of Makena for subsistence fishing and gath-
ering of marine resources. Hawaiians on all islands organized to protect their
traditional burial grounds from destruction by various forms of development,
a movement sparked by the controversy at Honokahua in Maui, where over
1,000 graves were dug up and relocated to build a Ritz-Carlton Hotel.
In 1983 a group of University of Hawai‘i professors and native speakers of
Hawaiian from Ni‘ihau formed Pünana Leo, to develop Hawaiian language
immersion preschools. In 1984 the first Pünana Leo school opened, and by
1995 the nine Pünana Leo Hawaiian Immersion preschools had a total enroll-
ment of 181 students. In addition, Pünana Leo began to work toward rescind-
ing the law that mandated that English be the only medium of instruction in
public schools. In 1987, after an absence of 100 years, the Hawaiian language
began again to be a medium language of instruction in the public schools. By
the 1999–2000 school year Ka Papahana-Kaiapuni Hawai‘i (public education
Hawaiian Language Immersion Program) had approximately 1,750 students
enrolled in eighteen schools. The Hawaiian language has been rescued from
extinction and continues to grow and expand as the language of choice for
Native Hawaiians.
When the last sugar plantation in leeward O‘ahu shut down in the 1990s,
taro farmers on windward O‘ahu petitioned the Hawai‘i State Water Commis-
sion to stop diverting the waters of the Wai‘ähole and Waikäne streams to the
‘Ewa plains. Thanks to their efforts, only half of the water is now diverted,
and the other half is allowed to flow into the Wai‘ähole stream. Native stream
life returned, marine life in Kane‘ohe Bay became more abundant, and farm-
ers reopened taro terraces that had lain dry and overgrown with brush for
decades. A new generation of Hawaiian and local youth from windward O‘ahu
began to pursue livelihoods involving the cultivation of taro, as their grand-
parents had done.
Throughout these struggles it became increasingly apparent that Native
Hawaiians lacked official legal standing to adequately protect Native Hawai-
ian lands and resources. Moreover, developments that occurred on state or
federal government lands exploited the original Crown and government lands
of the Hawaiian Kingdom to which Native Hawaiians maintained vested
rights of inheritance that were never surrendered. Thus, Native Hawaiians
began to seek political solutions to effectively protect the lands and resources
essential for the perpetuation of spiritual and cultural customs, beliefs, and
practices.

278
k aho ‘ ol awe: rebirth of the sacred

Native Hawaiian Sovereignty and Recognition


Under the U.S. Constitution, indigenous Native American tribes are recog-
nized as domestic dependent nations, with inherent powers of self-governance
and self-determination, for which the U.S. federal government sustains a trust
responsibility.40 This status has been extended to Eskimos, Aleuts, and Native
Alaskans under the Alaskan Native Claims Settlement Act. However, other
ethnic and racial minorities within the fifty U.S. states do not enjoy the status
of nationhood, they do nor do they have the right of self-governance and self-
determination, and the federal government does not have a trust responsibil-
ity for them. The status of Native Hawaiians as an indigenous people within
the currently defined boundaries of the United States has been recognized
through various policies.
From 1906 through 1998 the U.S. Congress, in effect, recognized a trust
relationship with the native people of Hawai‘i through the enactment of 183
federal laws that explicitly included Native Hawaiians in the class of Native
Americans.41 Some of the laws extended federal programs set up for Native
Americans to Native Hawaiians, while others represented recognition by the
U.S. Congress that the United States bore a special responsibility to protect
Native Hawaiian interests.42 Although the operational policy of the U.S. Con-
gress has been to exercise a trust responsibility with Native Hawaiians similar
to that which Congress has with Native Americans, the laws passed extended
an implicit rather than explicit and formal recognition that Native Hawaiians
are a sovereign people with the right of self-governance and self-determina-
tion. Without such an explicit law, Native Hawaiians may not be eligible to
receive the special benefits, entitlements, and protection the U.S. Congress
extended to them beginning in 1906.
The U.S. Supreme Court, in the case of Rice v. Cayetano, ruled on Feb-
ruary 23, 2000, that elections for the trustees of the State of Hawai‘i Office
of Hawaiian Affairs (OHA), in which only Native Hawaiians were allowed to
vote, used unconstitutional race-based qualifications.43 The majority of the
members of the court ruled that the Native Hawaiian OHA election violated
the Fifteenth Amendment of the U.S. Constitution, which states that the
right to vote cannot be denied on account of race or color.44 Subsequently, in
the November 2000 election for OHA trustees, all registered voters, regard-
less of Native Hawaiian ancestry, were allowed to cast votes and to run for
these offices.

279
chapter six

The U.S. Supreme Court, in its ruling, stated that Native Hawaiians have
a shared purpose in the Islands with the general public and that the Constitu-
tion of the United States has become the heritage of all the citizens of Hawaii,
including Native Hawaiians. In addition, the court raised questions about
whether Native Hawaiians are in fact a distinct and unique indigenous people
with the right of self-governance and self-determination under U.S. law or
are instead an ethnic or racial minority not eligible for such rights.
In the ruling, a majority of the Supreme Court justices also raised, but did
not resolve, four fundamental questions regarding the status of Native Hawai-
ians. May Congress treat the Native Hawaiians as it does the Indian tribes?
Has Congress in fact determined that Native Hawaiians have a status like that
of Indians in organized tribes? May Congress delegate to the State of Hawai‘i
the authority to preserve that status? Has Congress delegated to the State of
Hawai‘i the authority to preserve that status? 45 A negative answer to any of
these questions could result in a determination that Native Hawaiians do not
qualify under U.S. law for the rights and protection afforded other indigenous
peoples within the fifty states. The majority of the Supreme Court Justices also
seemed to open the door to future legal challenges on the status of Native
Hawaiians when it stated:

It is a matter of some dispute, for instance, whether Congress may treat


the native Hawaiians as it does the Indian tribes. Compare Van Dyke, The
Political Status of the Hawaiian People, 17 Yale L. & Pol’y Rev. 95 (1998),
with Benjamin, Equal Protection and the Special Relationship: The Case
of Native Hawaiians, 106 Yale L.J. 537 (1996). We can stay far off that
difficult terrain, however.46

Suddenly, the status, rights, and entitlements Native Hawaiians had enjoyed
throughout the twentieth century could be legally challenged out of existence.
Moreover, the Supreme Court ruling seemed to contradict the policy of the
U.S. Congress toward Native Hawaiians.
Hawai‘i’s congressional delegation, led by Senators Daniel Akaka and Dan-
iel Inouye, drafted and introduced legislation (called the Akaka Bill) to explic-
itly and unambiguously clarify the trust relationship between Native Hawai-
ians and the United States. Although the bill failed to pass from 2000 to 2006,
Hawai‘i’s congressional delegation will continue to introduce similar bills
until one passes. Such a bill would formally and directly extend the federal pol-
icy of self-determination and self-governance to Native Hawaiians as Hawai‘i’s
indigenous native people. The legislation would provide a process for recog-

280
k aho ‘ ol awe: rebirth of the sacred

nition by the United States, under the Secretary of the Department of Inte-
rior, of a Native Hawaiian governing entity.47

Lessons of Stewardship from Kanaloa


Kanaloa is important as an example of the role played by kua‘äina from cul-
tural kïpuka in the perpetuation and revival of Native Hawaiian culture.
Through Kanaloa, kua‘äina shared their understanding and experience related
to traditional concepts of wahi pana, aloha ‘äina, and lökähi, popularized
as core practices in the stewardship of Hawai‘i’s land, ocean, and natural
resources.
Upon receiving title to the island of Kaho‘olawe, the State of Hawai‘i set
up the Kaho‘olawe Island Reserve Commission (KIRC) to manage the island
and its surrounding waters, out to two miles, as a Native Hawaiian cultural
reserve. The commission adopted a vision statement that reads in part:

The kino (body) of Kanaloa is restored . . . Na po‘e Hawai‘i (the people


of Hawai‘i) care for the land in a manner which recognizes the island and
ocean of Kanaloa as a living spiritual entity. Kanaloa is a pu‘uhonua (spiri-
tual refuge) and wahi pana (sacred place) where Native Hawaiian cultural
practices flourish.
The piko (navel) of Kanaloa is the crossroads of past and future gen-
erations from which the Native Hawaiian lifestyle spreads throughout the
islands.48

This vision statement encompasses the physical and spiritual restoration of


Kanaloa, both the island and the god. It projects activities on the island that
revolve around restoration. The isolation of the island provides a historic
opportunity to revegetate it with native plants. It also envisions a protected
marine sanctuary that can serve as a pool for restocking marine life for the
ocean in and around Maui nui. The statement presents Kanaloa as a cultural
learning center where traditional cultural and spiritual customs, beliefs, and
practices of Native Hawaiians such as way-finding, fishing, and healing can
flourish and spread out to all the islands. Native Hawaiian culture exists
nowhere else in the world, and Kanaloa will play a role its perpetuation.
In order to implement this vision, the KIRC worked with the Protect
Kaho‘olawe ‘Ohana and the Edith Kanaka‘ole Foundation to develop tradi-
tional kua‘äina stewardship principles for guiding the development of a land
use plan for the island.49 First among these principles is that the ahupua‘a is the

281
chapter six

basic unit of Hawaiian natural and cultural resource management. An ahu-


pua‘a runs from the sea to the mountains and contains a sea fishery and beach,
a stretch of kula or open cultivable land, and, higher up, the forest. The court
of the Hawaiian kingdom described the ahupua‘a principle of land use in the
case In Re Boundaries of Pülehunui, 4 Haw. 239, 241 (1879) as follows:

A principle very largely obtaining in these divisions of territory [ahupua‘a]


was that a land should run from the sea to the mountains, thus affording to
the chief and his people a fishery residence at the warm seaside, together
with products of the high lands, such as fuel, canoe timber, mountain birds,
and the right of way to the same, and all the varied products of the inter-
mediate land as might be suitable to the soil and climate of the different
altitudes from sea soil to mountainside or top.

The entire island was on ahupua‘a of the moku or district of Honua‘ula.


The island was divided into twelve watersheds or ‘ili. Restoration of the island
will start at the central point of the island and proceed down to the ocean, ‘ili
by ‘ili, recognizing the integral relationship between soil disturbance, water
flows, wind, erosion, and runoff.
A second important principle is that the natural elements—land, air, fresh
water, ocean—are interconnected and interdependent. From the ocean rise
clouds that drench the land with rain, which recharges the island’s water table
or flows across the landscape in streams or rivers. The rivers and springs ulti-
mately flow back to the beaches and into the ocean. Cultural land use man-
agement must take all aspects of the natural environment into account. The
atmosphere where clouds form is an integral link in the cycle of life. Ho‘ai-
lona or prophetic natural signs and omens appear in the atmosphere and help
guide and validate Hawaiian practices.
Hawaiians consider the land and ocean to be integrally united and that
these ahupua‘a also include the shoreline as well as inshore and offshore ocean
areas such as fishponds, reefs, channels, and deep-sea fishing grounds. The
sixty-nine fishing ko‘a that had been constructed on the island were also mark-
ers for offshore fishing grounds. These fishing grounds are also part of ‘ili and
must be considered in restoration activities on the island.50
A third important principle of Native Hawaiian stewardship is that of all
the natural elements, fresh water is the most important for life and needs to be
considered in every aspect of land use and planning. The Hawaiian word for
water is wai; the Hawaiian word for wealth is waiwai. In Hawai‘i, water is the

282
k aho ‘ ol awe: rebirth of the sacred

source of well-being and wealth, and the wealth of the land is based upon the
amount of fresh water available upon it.
A fourth important principle is the acknowledgment that Hawaiian ances-
tors studied the land and the natural elements and became very familiar with
the land’s features and assets. Ancestral knowledge of the land was recorded
and passed down through place-names, chants, and legends that name the
winds, rains, and features of a particular district. Hawaiians applied their
expert knowledge of the natural environment in constructing their homes,
temples, cultivation complexes, and irrigation networks. Hawaiian place-
names, chants, and legends inform Hawaiians and others who know the tra-
ditions of the cultural and natural resources of a particular district. Insights
into the natural and cultural resources inform those who use the land how to
locate and construct structures and infrastructure so as to have the least nega-
tive impact upon the land. In planning for the land, ancestral knowledge about
the land and its natural resources should be gathered in order to allow for sus-
tainable use of its resources.
A fifth principle recognizes that an inherent aspect of Hawaiian stewardship
and use of cultural and natural resources is the practices of aloha ‘äina and
mälama ‘äina, or respect and conservation of the land to ensure the sustain-
ability of natural resources for present and future generations. These rules of
behavior are tied to cultural beliefs and values regarding respect of the ‘äina,
the virtue of sharing and not taking too much, and a holistic perspective on
organisms and ecosystems that emphasizes balance and coexistence. The
Hawaiian outlook that shapes these customs and practices is lökähi, or main-
taining spiritual, cultural, and natural balance with the elemental life forces of
nature.
Taken together, these principles, learned through the collaboration of kua-
‘äina, scholars, Native Hawaiian activists, and planners in projecting future
uses of Kanaloa, provide an excellent foundation for the stewardship of the
Hawaiian Islands as a whole.
In recognition of the important lessons of Kanaloa for the stewardship of
our Hawaiian Islands, the Bishop Museum partnered with Community Devel-
opment Pacific and the Protect Kaho‘olawe ‘Ohana to develop an exhibit,
Kaho‘olawe: Ke Aloha Küpa‘a I Ka ‘äina, Steadfast Love of the Land. Photo-
graphs featured in the exhibit were also published in a book, Kaho‘olawe, Nä
Leo O Kanaloa: Chants and Stories of Kaho‘olawe. Through photo murals, framed
photos, cultural and military artifacts, videos, interactive displays, and com-

283
chapter six

puter simulations, the story of Native Hawaiians’ love and care for the land
was told through the mo‘olelo of Kanaloa. The exhibit opened at the Bishop
Museum in January 1996 and eventually traveled to all the major islands—
Maui, Läna‘i, Hawai‘i, Kaua‘i, and Moloka‘i. Nearly 187,000 people, espe-
cially students from various public and private schools, learned of the impor-
tance of caring for the natural and cultural resources of the Hawaiian islands
through the exhibit.
In 2001 the exhibit was reorganized for a national exhibit at the Arts and
Industries Building of the Smithsonian Institution at the National Mall in
Washington, D.C. Renamed “Kaho‘olawe: Rebirth of a Sacred Hawaiian
Island,” it opened on June 5, 2002. By the time the exhibit closed on Septem-
ber 2, 2002, 304,037 visitors had viewed the it. The national exhibit took the
visitor behind the tourist and Hollywood images of Hawai‘i and related the
trials and tribulations of living on islands with fragile ecosystems. Kanaloa
provided a model of how a community can successfully combine efforts with
biologists and technical experts to restore premium natural resources and
respect the cultural beliefs, customs, and practices of an indigenous people.
The exhibit also depicted a chronology of the historical relationship of Native
Hawaiians to the U.S. government up through the congressional legislation
that would have recognized the sovereignty of Native Hawaiians. To help
interpret the exhibit to a national audience, fifty-five persons who had connec-
tions to Hawai‘i and were living in the Washington, D.C., area were trained
as docents. Public programs on celestial way-finding, Native Hawaiian story-
telling, and songs of Native Hawaiian political resistance were organized to
expand upon the information presented in the exhibit.
The exhibit reaffirmed that Kanaloa has been reborn in the hearts and
minds of Native Hawaiians as a sacred island. It connects Native Hawaiians
to their ancestral spirits, as described in the passage by the Hawaiian scholar
Edward Kanahele quoted in chapter 1.
Kanaloa is such a place for all Native Hawaiians. Moreover, it is a place
where non-Hawaiians too can experience the Native Hawaiian culture that is
at the core of the Hawaiian Islands.

Kanaloa and the People of Hawai‘i


Throughout the years of the struggle to stop the bombing and reclaim Kana-
loa as sacred Hawaiian land, the broader multiethnic local community also
rallied in support of the effort as their own. The Protect Kaho‘olawe ‘Ohana

284
k aho ‘ ol awe: rebirth of the sacred

is not exclusively Native Hawaiian and includes distinguished leaders who are
of local Asian and Caucasian ancestry. Because Native Hawaiians comprise
only 20 percent of the Hawai‘i population, the success of the movement to
protect Kanaloa was made possible only through the support of the broader
local community. Hawaiian and non-Hawaiian political leaders alike, such as
Senator Daniel K. Inouye, Senator Daniel Akaka, Congressman Neil Aber-
crombie, Congresswoman Patsy Mink, former Congresswoman Patricia Saiki,
and Governor John Waihe‘e, played important roles in ending the bombing
and the authorization of $400 million for the clean-up and restoration. Among
the thousands who have visited the island with the Protect Kaho‘olawe ‘Ohana
since the 1980 consent decree made it possible have been Native Hawaiians
and non-Hawaiians alike.
Everyone who wants to embrace and experience Native Hawaiian culture is
welcome to go to Kanaloa and work on projects to heal her natural and cul-
tural resources. Kanaloa reconnected a generation of Hawaiians to their ances-
tral soul. Throughout the years of access to the island, it has been shown that
Kanaloa can also connect non-Hawaiians for whom Hawai‘i is home to Native
Hawaiian culture. Through the experience on Kanaloa, Native Hawaiian val-
ues and way of life are being spread throughout the Islands. The final verse of
the chant composed by Pualani Kanaka‘ole Kanahele on the history of the
island seems a fitting note on which to end this chapter:

Ua hö‘ea ka lä ho‘iho‘i ea The day for sovereignty is at hand


Ka lä ho‘iho‘i moku The day to return the island
Ka lä ho‘iho‘i mana küpuna The day to return the ancestral
influence
Aia i ka Mua Ha‘i Küpuna e Hänau nei It is at the Mua Ha‘i Küpuna where
it is born
E kanaloa ‘ia ana i ka piko o kapae ‘äina To be established in the navel of the
islands
He ‘äina küpa‘a no nä Hawai‘i A steadfast land for the Hawaiians
E ola ka Mua Ha‘i Küpuna Give life to the Mua Ha‘i Küpuna
A mau loa i ka lani a Käne Forever in the heavens of Käne
A mau loa i ke kai a Kanaloa Forever in the sea of Kanaloa.51

285
 seven 

Ha‘ina Ia Mai: Tell the Story

opened this mo‘olelo with a personal journey in an attempt to land on

I the island of Kanaloa with Uncle Harry Mitchell. I finally crossed the
channel and landed on Kanaloa in November 1984. Through Kanaloa and
Mitchell I was introduced to the kua‘äina of our islands and led back to my
ancestral soul as a kanaka ‘öiwi.1 Through Kanaloa I have participated in the
annual Makahiki ceremonies to Akua Lono, beginning in 1986, as well as cer-
emonies in honor of Akua Kanaloa and Akua Kane.
My involvement with Kanaloa led me to focus my scholarly research on
the cultural kïpuka of our islands among the kua‘äina who protected the
sacred nature of nä kanaka ‘öiwi and of the islands of our lives. This research
endeavor gave the responsibility to me as a kanaka ‘öiwi to first place this work
at the service of those who shared their knowledge with me. In Hawaiian we
have a saying: “Aloha mai no, aloha aku” (When aloha is given, aloha should
be returned). The kua‘äina in the various cultural kïpuka described in this
book shared their knowledge and experience with me with aloha, and I have
strived to return their aloha by applying the information I gathered to our
efforts to protect the natural and cultural resources and subsistence lifestyles
of their communities. This has to a large degree postponed the publication of
this work, but I believe it was necessary to first make sure that the traditional
knowledge of the kua‘äina was given back to them to protect their well-being
and quality of life. My own purpose in publishing this work is to promote the
critical significance of these cultural kïpuka for the perpetuation of kanaka
‘öiwi culture and, hopefully, to inspire public efforts and the formation of
public policies that will protect these cultural kïpuka.
Sustainable use of Hawai‘i’s natural and cultural resources is a core concept
for planning the future of Hawai‘i, and Native Hawaiian stewardship princi-
ples can play a significant role in achieving sustainability for the Islands. These,
in combination with the traditional principles associated with the subsistence

286
H A ‘ INA I A M A I : tell the story

practices of kua‘äina in cultural kïpuka described in this mo‘olelo, are impor-


tant in protecting Hawai‘i’s precious cultural and natural resources.
The subsistence practices described in this book’s chapters on Waipi‘o,
Häna, Puna, Moloka‘i, and Kanaloa can serve as a foundation for the design of
sustainable use of Hawai‘i’s resources. As we move into the twenty-first cen-
tury it is critical to acknowledge the importance of protecting these cultural
kïpuka and the natural and cultural resources, which are critical to the subsis-
tence practices of the kua‘äina of these districts — not only for the kua‘äina and
their descendants, but also for Hawai‘i’s multicultural society as a whole.

Waipi‘o
Of all of the cultural kïpuka covered in this book, Waipi‘o is the only place
where I did not conduct oral histories or engage in community-based work.
Here I relied upon the unpublished field notes of Stella Jones from her ethno-
graphic work there, sponsored by the Bishop Museum in 1931. I also mined
the comprehensive oral history interviews of the Ethnic Studies Oral History
Project staff, collected in Waipi‘o Mano Wai: An Oral History Collection, vols. 1
and 2. Most recently my colleague Luciano Minerbi, a professor in the Uni-
versity of Hawai‘i at Mänoa’s Department of Urban and Regional Planning,
conducted a planning practicum with the taro farmers, landowners, educators,
and residents of Waipi‘o Valley in the fall of 1999. The technical report pro-
vided me with an update on the production of taro in the valley and the cul-
tural education projects centered in the valley.
Waipi‘o Valley flourishes as an important Native Hawaiian center that
sparks and nourishes our cultural imagination. The valley made news in 1994
when the sacred ka‘ai or sennit burial caskets believed to contain the bones of
High Chief Lïloa and High Chief Lonoikamakahiki were taken from the
Bishop Museum and apparently returned to a burial cave in Waipi‘o Valley. As
discussed in the chapter on Waipi‘o, these ka‘ai were removed in 1829 from
the Hale o Lïloa by Kuhina Nui Ka‘ahumanu after her conversion to Chris-
tianity. The Bishop Museum had planned to place the ka‘ai in the Kaläkaua
crypt at Mauna Ala in Honolulu, while Native Hawaiians from Hawai‘i Island
had asked permission to instead inter the ka‘ai in Waipi‘o Valley. Before any
formal action was taken, the ka‘ai were removed, and it is believed that they
were taken back to their place of origin—Waipi‘o Valley.
In March 2004, the State Water Commission ruled that the Lalakea Ditch,
which had diverted the Lalakea and Hakalaoa Streams to irrigate Hämäkua

287
chapter seven

coast plantations for eighty years, could be dismantled.2 On June 29, 2004,
the stream flow was fully restored and once again there were two Hi‘ilawe
falls instead of only one. The stream restoration is being studied by Bishop
Museum scientists in collaboration with students from Hawai‘i Island middle
schools.3 The restoration of these Waipi‘o streams will set a precedent. The
only other case of restoration of stream waters was on windward O‘ahu in the
case of the Waiähole, Waianu, and Waikäne streams, but these streams have
only been partially restored.
Waipi‘o Valley is one of the most gorgeous and picturesque ahupua‘a in our
islands. Its cultivated taro pond fields conjure memories of traditional Hawai‘i,
and it is one of the few places that now flourishes as a Native Hawaiian cul-
tural and educational center, with programs offered by the Edith Kanaka‘ole
Foundation and Kanu O Ka ‘äina Charter School. The Bishop Museum may
one day return its land to a Native Hawaiian land trust in support of the cul-
tural and educational programs in the valley. The entire valley is protected
within the coastal zone wherein any development requires a special manage-
ment area permit, including an assessment of impacts upon Native Hawaiian
subsistence, cultural, and religious practices. Waipi‘o Valley is an important
cultural icon in the hearts and minds of Native Hawaiians and local people as
a traditional and contemporary center of Native Hawaiian culture and taro
production. All of these factors combined will contribute to its perpetuation
as a cultural kïpuka.

Hâna
The rugged and luxuriant Häna district of Maui, with its deep, verdant val-
leys, glistening waterfalls cascading down to the sea, and magnificent moun-
tain slopes disappearing into billowing clouds, is alluring and enchanting.
However, until my first visit to Ke‘anae-Wailuanui with Harry Mitchell, as
described in the opening of this mo‘olelo, I never really experienced its spe-
cial qualities, nor did I learn the mo‘olelo of the Hawaiian ‘ohana who pro-
vided stewardship of this land from generation to generation. In subsequent
visits with Mitchell he often reminisced and shared stories of the land and of
his youth as we drove along the coast, from valley to valley, past stream after
stream, to his taro lands in Ke‘anae-Wailuanui. One weekend, in April 1988,
in order to assist me with the Häna chapter for my dissertation, Mitchell drove
me from Kahului to Ke‘anae-Wailuanui and on to Häna, through Kïpahulu,
Kaupö, Nu‘u, Kahikinui, Kanaio, and ‘Ulupalakua, down Haleakalä and back

288
H A ‘ INA I A M A I : tell the story

to Kahului to share the mo‘olelo of his ‘äina and his own connection to each
place. The landscape, for me, was no longer just a beautiful place to look at,
swim in, hike through, and enjoy. Mitchell's mo‘olelo brought the ‘äina to life
with the spirit of our ancestors, the kua‘äina who lived on the land for gener-
ations, and their descendants who continue to care for it today. The land was
alive with resources that had fed the generations of ‘ohana who nurtured the
land, embraced and cared for it as a part of their ‘ohana, and relied upon it for
subsistence. Mitchell and his ancestral knowledge had led me onto the path of
the kua‘äina who held sacred the life of the land and the resources of the land.
In addition to the interviews with Mitchell, I also discovered interviews
conducted by Mary Kawena Pukui, in Hawaiian and English, with küpuna of
Häna, Kïpahulu, Kaupö, and Nu‘u in May 1960. These are reposited in the
Audio-Recording Collection of the Anthropology Department of Bishop
Museum. Some of these interviews were transcribed and translated, and some
were not. The purpose of the interviews was to record and document the
original Hawaiian names of places on each of our islands. The collection is an
invaluable treasure, and I was honored and privileged to be the first scholar
allowed access to them in 1987. For Häna, there were fifty-five half-hour taped
interviews with thirty küpuna. A translation by Pukui of Thomas K. Maunu-
pau’s article about his trip to Kaupö, published in the Nüpepa Kü‘oko‘a of June
15, 1922, was another valuable source of information for Kaupö.
In 1994 I had a chance to contribute to the protection of Ke‘anae-Wailua-
nui and give something back to Mitchell’s community. I worked on a study
with Group 70 and Cultural Surveys Hawai‘i that resulted in a report titled
“Kalo Kanu O Ka ‘Äina: A Cultural Landscape Study of Ke‘anae and Wailua-
nui, Island of Maui.” Mitchell had passed away in 1990, but his spirit guided
my work. I conducted interviews with thirteen kua‘äina of Ke‘anae-Wailuanui
who ranged in age from twenty-seven to ninety-four. I felt so privileged to be
able to interview küpuna Enos Akina, who was ninety-four; küpuna Helen
Nakanelua, eighty-three; küpuna Mary Kaauamo, eighty-two; küpuna James
Hueu, eighty; küpuna Maggie Alu, seventy-nine; and küpuna Apolonia Day,
seventy-two. The knowledge they and the younger mäkua of Ke‘anae-Wailu-
anui shared with me was invaluable.
The purpose of the study was to develop a policy for Maui County and the
State of Hawai‘i to protect historical cultural landscapes in Hawai‘i, such as
Ke‘anae-Wailuanui. In January 1993 the Hawaii‘i State Legislature established
a task force on cultural landscapes within the Department of Land and Nat-
ural Resources. In 1994 the task force completed its work by providing

289
chapter seven

descriptions and typologies of cultural landscapes in Hawai‘i: (1) abandoned


villages or agricultural systems; (2) taro-producing areas; (3) sugar lands;
(4) ranches; (5) fishing areas; (6) religious and legendary sites; (7) fishponds;
(8) traditional gathering areas; and (9) entire islands. The task force recom-
mended that a model project focusing on the Ke‘anae-Wailuanui community
be funded to develop specific guidelines on how to protect historical cultural
landscapes in Hawai‘i.4 The Maui County General Plan and the Häna Com-
munity Plan recommended that Maui county retain the cultural identity of
the Häna region and promote only those economic activities that are environ-
mentally benign and compatible with the cultural sensitivities of the area. Fed-
eral funds were available to Maui county to fulfill the purposes of the National
Historic Preservation Act to designate protection of national historic cultural
landscapes, and Maui county decided to use these federal preservation monies
for a study to recommend measures to protect the cultural landscape of Ke‘a-
nae-Wailuanui.
From the oral history interviews that I conducted, I was able not only to
document the rich cultural heritage of the ‘ohana who live in Ke‘anae-Wailua-
nui, but also to generalize patterns of traditional subsistence custom and usage
among the kua‘äina of the ahupua‘a. These patterns have served as a template
for comparison in other cultural kïpuka. I have been able to build and expand
upon this model to document and protect subsistence customs and practices
and the natural resources upon which kua‘äina throughout the cultural kïpuka
of our islands rely. In subsequent studies to document and map Native Hawai-
ian cultural and natural resources used for cultural, religious, and subsistence
purposes, I incorporated the cultural information from the Ke‘anae-Wailua-
nui study.

Boundaries
The information provided to me by the kua‘äina of Ke‘anae-Wailuanui helped
me understand that there are actually two intersecting areas that make up a
cultural landscape—the core area and the broader area of traditional cultural
practices. The core area includes land used for residence, areas of taro cultiva-
tion, irrigation networks, and associated settlement and circulation systems.
The broader traditional cultural practices area usually coincides with the
traditional ahupua‘a and moku. It includes all of the zones needed for the peo-
ple to gather, hunt, and fish for subsistence, cultural, and religious purposes.
In many cases, the areas utilized by ‘ohana for gathering, hunting, and fishing
may have extended beyond the ahupua‘a into other areas of the moku or dis-

290
H A ‘ INA I A M A I : tell the story

trict or another part of the island. One must rely upon the ‘ohana of the area
who are subsistence practitioners to describe the boundaries of the traditional
cultural practices area.
Often environmental impact studies focus only on impacts on natural and
cultural resources within the core cultural practices area. However, it is impor-
tant to broaden the scope of such studies to include the expanded cultural use
area of the entire ahupua‘a and moku.

Landscape Components
Through the Ke‘anae-Wailuanui study, I learned that the cultural landscape
is composed of physical elements that manifest the technological and cultural
basis of human use of the land through time.
The components of a Hawaiian cultural landscape include (1) areas of taro
cultivation; (2) other areas of cultivation; (3) circulation networks; (4) build-
ings, structures, nonstructural facilities, and objects; (5) clusters (defined as
groupings of buildings or features that result from function, social tradition,
climate, and other influences); (6) internal boundaries; (7) an irrigation ditch
system, including roads and tunnels; (8) archaeological and historic sites; (9)
open areas; (10) small-scale elements; (11) viewing points; and (11) cultural
resources and use areas. Of these components, I focused on the cultural
resources and use areas in the Ke‘anae-Wailuanui district and in subsequent
studies.
First, there are wahi pana such as heiau, shrines, burial caves and graves,
and geographic features associated with deities and significant natural, cul-
tural, spiritual, or historical phenomena or events. Edward Kanahele offered
the following description of wahi pana: “The gods and their disciples specified
places that were sacred. The inventory of sacred places in Hawai‘i includes the
dwelling places of the gods, the dwelling places of venerable disciples, temples,
and shrines, as well as selected observation points, cliffs, mounds, mountains,
weather phenomena, forests, and volcanoes.” 5
Secondly, streams and springs are important as habitats for native species
of marine life, for taro cultivation, and for domestic uses.
Third, shorelines, reefs, and nearshore and offshore ocean resources are
important for gathering of foods and medicines and for conducting cultural
and spiritual customs.
Fourth, forests are important for hunting pigs and other animals; for gath-
ering plants used for medicine, foods, ceremonial adornment, and ritual offer-
ings; and for conducting spiritual rituals.

291
chapter seven

Fifth, domains of ‘aumakua or ancestral deities are particular natural and


cultural features where Hawaiians renew their ties to ancestors through expe-
riencing natural phenomena and witnessing ho‘ailona or natural signs.
Sixth, trails and dirt roads are indispensable for affording access to the cul-
tural resources and use areas, both mauka to forests and streams and makai to
streams and the ocean.
Though I have expanded upon this essential list in subsequent studies, these
served as an important starting point in developing a systematic community-
based approach to identifying and documenting the wide range of cultural and
natural resources relied upon for cultural, religious, and subsistence activities.
The most exciting aspect of the study for me was to document the unbro-
ken continuity of cultural and subsistence beliefs, customs, and practices in a
thriving, predominantly Native Hawaiian community at the end of the twen-
tieth century. The kua‘äina of Häna persist in their way of life despite new-
comers, illegal drugs, and economic challenges. As the intensity and pace of
change increases within a global economy, public policies will be needed that
provide funding, regulation, and enforcement to protect the Native Hawaiian
way of life in Häna.

Puna
I spent most of my summers as a child and through college with my grand-
parents in Waiäkea on the edge of Hilo, Hawai‘i. My parents owned papaya
fields and hala forest land in ‘Opihikao in the Puna district adjacent to a small
ranch owned by my mother’s brothers. I spent many summer days in Puna,
both at ‘Opihikao and also at Puna’s wondrous black sands, rugged lava rock
coastline, warm pools and springs, mysterious lava tubes, awesome volcanic
craters, and spectacular eruptions. Puna is an essential and dynamic place that
helps define my life and identity as kanaka ‘öiwi.
My formal research into Puna took shape around the “Native Hawaiian
Ethnographic Study for the Hawai‘i Geothermal Project Environmental
Impact Study” that I conducted with Jon Matsuoka of the School of Social
Work and Luciano Minerbi of the Department of Urban and Regional Plan-
ning. This was the first Native Hawaiian cultural impact study conducted for
a major federal project in Hawai‘i since it began to be required as part of an
environmental impact study. We decided to design the study as a template for
subsequent cultural impact studies and thus set ourselves a high standard.
Although the study areas included both Puna and East Maui, my role was to

292
H A ‘ INA I A M A I : tell the story

conduct the ethnographic study of Puna and the Pele beliefs, customs, and
practices in conjunction with Matsuoka, Kumu Hula Pualani Kanaka‘ole
Kanahele, and a Pacific Islands Studies graduate student, Noenoe Barney-
Campbell. This final report served as an excellent model of how to model and
conduct a cultural impact study. It formed the basis of the Puna chapter in
this book, and I patterned the chapters on Waipi‘o, Häna, and Moloka‘i on
the same model.
The federal and state governments and the private investor terminated the
Hawai‘i Geothermal Project before the entire Environmental Impact Study
was completed and published by Oak Ridge National Laboratory. The archae-
ological assessment was stopped in its initial phase. However, since the cul-
tural impact study component that we were responsible for was already com-
pleted, Oak Ridge printed and distributed it to the informants, government
agencies, and public and university libraries. This book is the first time much
of the information gathered for the ethnographic study will be published and
made available to a larger audience. We shared our model for the conduct of a
cultural impact study with the Hawai‘i Office of Environmental Quality Con-
trol, and it was incorporated into their official guidelines on how to conduct
such studies as part of an environmental impact study.
The ethnographic study of Puna began with an examination of the tradi-
tional cultural significance of the district. The place-names for the district and
the ‘ölelo no‘eau or descriptive proverbs and poetic sayings for which the area
is famous were found and interpreted. Descriptive chants for the area were
researched, translated, and interpreted. A special review of the Pele chants for
the district was conducted in order to identify significant sites and cultural use
areas. Combined, these sources provided valuable insights into the cultural
resources and features for which the area was known and the overall role of
this area in the traditional cultural practices and customs of Hawaiians. Each
of the chapters describing the cultural kïpuka opens with a discussion of the
‘ölelo no‘eau and chants descriptive of and unique to that particular ahupua‘a,
moku, or island.
Traditional mo‘olelo record what the Native Hawaiian ancestors observed
as the primal natural elements and the important natural and physical features
and natural resources of the landscape. They provide, in story form, a descrip-
tion of the natural environmental setting in which the early Hawaiians settled
and established themselves. The primal natural elements were depicted as
manifestations of Hawaiian deities, and the myths and chants relate which
natural elements dominated the landscape and the lives of the early Hawai-

293
chapter seven

ians. Research by archaeologists complements the traditional mo‘olelo with


material evidence and artifacts of the record of human settlement. Archaeo-
logical studies have increased over the years as part of required environmen-
tal impact studies conducted for proposed development. These reports are
available at the University of Hawai‘i at Mänoa’s Hamilton Library and the
Hawai‘i State Historic Preservation Department. An inventory of these
reports is available at the State Historic Preservation Department. The tradi-
tional mo‘olelo for each of the cultural kïpuka follows the discussion of the
‘ölelo no‘eau and chants.
For the period of ruling chiefs, early contact with European and American
traders and the establishment of the Hawaiian monarchy, the four published
Native Hawaiian scholars of the early nineteenth century—David Malo, Sam-
uel Kamakau, John Papa Ii, and Kepelino—provide excellent accounts.6 Abra-
ham Fornander’s Collection of Hawaiian Antiquities and Folk-lore is also impor-
tant. Documents range from the journals of explorers and missionaries to
government records of the Kingdom of Hawai‘i—such as the census, tax rec-
ords, indices of land awards, record of Native Testimony to the Board of Com-
missioners to Quiet Land Titles, and the Boundary Review Commission pro-
ceedings — and Thrum’s Hawaiian Annual.
For the territorial period, from 1898 through statehood, visitor guide
books, magazines, newspapers, monographs such as Handy’s Native Planters in
Old Hawai‘i, and documents in the archives of the State of Hawai‘i and vari-
ous museums are excellent resources. For Puna, the Hawai‘i Volcano National
Park Headquarters Library had documents and records particular to the dis-
trict. Oral history and key informant interviews are important sources of infor-
mation about the lifestyle and livelihoods of the Puna Hawaiians. For Puna, I
drew upon the Kalapana Oral History Project and interviews by Russell
Apple.
For the post-statehood period, I searched for contemporary studies and also
conducted my own oral history interviews in every area except Waipi‘o Val-
ley. In the summer of 1970 the University of Hawai‘i at Mänoa Departments
of Geography and Anthropology and the School of Public Health collabo-
rated on human ecology studies in North Kohala and Puna, work preceded by
a similar study on Moloka‘i by the Department of Anthropology. These stud-
ies include surveys of subsistence activities in each of these areas and provide
a useful baseline of data for comparison of these activities over time into the
twenty-first century.

294
H A ‘ INA I A M A I : tell the story

In 1982 the U.S. Department of Energy commissioned a study by the Puna


Hui ‘Ohana, an organization of Hawaiian families in Puna, as a prelude to a
proposal to develop plants for producing electricity from geothermal energy.
This study also surveyed the subsistence activities of the families in the Puna
community and provides useful data for comparison with the earlier Univer-
sity of Hawai‘i study as well as subsequent ones. A similar study was conducted
on Moloka‘i in 1981 and published in the Moloka‘i Data Book: Community Val-
ues and Energy Development by the Department of Urban and Regional Plan-
ning. There was nothing comparable for Häna or Waipi‘o.
For the geothermal ethnography, Jon Matsuoka and I conducted key
informant interviews in 1994 with members of Native Hawaiian ‘ohana who
had lived in the Puna district for several generations. As a unique feature of the
interviews, informants were asked to identify areas of subsistence, cultural, and
religious use on a topographical map of the district. The data gathered were
analyzed and organized into a report on the customs, beliefs, and practices of
the Puna Hawaiians.
The interviews revealed that there is still a wealth of ancestral knowledge
that has been kept alive and is practiced by Hawaiian families in Puna. More-
over, the living culture is constantly undergoing growth and change. Certainly
the ethnohistories of places such as Puna, Häna, and Moloka‘i can be devel-
oped from a wide variety of written sources. Nevertheless, cultural impact
studies, in order to be complete, need to go beyond the written record and
include interviews with key informants who are members of Hawaiian ‘ohana
and traditional hälau (cultural groups) who live in the area and have established
a relationship of stewardship for the cultural and natural resources of the
area. For Puna, the rich ethnohistorical written sources were complemented
by interviews and focus group discussions and participatory mapping with
Hawaiian families, members of hälau, and cultural groups. All the information
gathered, written and oral, was combined into the final report on the impact
of geothermal energy development on the cultural beliefs, customs, and prac-
tices of the Puna Hawaiians.
At the same time that the ethnographic study was being conducted, the Pele
Defense Fund was preparing its case to protect the traditional and customary
rights of Native Hawaiians in the Wao Kele o Puna Forest (Pele Defense Fund
v. Paty 79 Haw. at 442 [1992]).7 When the case was heard in August 1994, key
informants that were interviewed for the geothermal ethnographic study tes-
tified as witnesses. I also presented testimony based upon the key informant

295
chapter seven

interviews, including the topographic map of Puna on which the informants


had noted with pasted-on dots the location of their subsistence hunting, fish-
ing, and gathering areas, cultural sites, and trails. This testimony and that of
the key informants laid the foundation for a ruling that set a precedent for
Native Hawaiian gathering rights in Puna and throughout the islands.

Moloka‘i
I first began to visit Moloka‘i in 1975 as a class project for the ethnic studies
course on Hawaiians that I teach at the University of Hawai‘i at Mänoa.
Colette Machado, a good friend and former student who had returned to her
home island, welcomed the students so they could be educated about the cul-
tural significance of Moloka‘i and learn about the political struggles of the
homesteaders for water and of the kua‘äina for access. The idea was for us to
be inspired to do research and organize support for Moloka‘i when we went
back to O‘ahu. Colette helped to arrange places for us to camp, vehicles to
transport baggage from the airport, and community speakers to educate the
students.
Through Colette I met people of Moloka‘i who were members of both Hui
Ala Loa and the Protect Kaho‘olawe ‘Ohana—Noa Emmett Aluli, Judy Napo-
leon, Joyce Kainoa, and John Sabas. I already knew George Helm from high
school; he went to the all-boys’ St. Louis High School on O‘ahu at the same
time I attended the all-girls’ sister school, Sacred Hearts Academy. We were
both in a co-ed choral group started by Kumu John Lake for the students from
our schools.
Over the years of visiting Moloka‘i, I grew to appreciate the grassroots
community of Moloka‘i who were guided by their küpuna in a lifelong com-
mitment to protect the Hawaiian way of life on Moloka‘i and on Kaho‘olawe.
My connections to Moloka‘i and Kaho‘olawe were intertwined, although my
first actual visit to Kaho‘olawe was ten years after my first visit to Moloka‘i.
Throughout the years I gave support to community issues on Moloka‘i.
When I wrote my dissertation on cultural kïpuka, I decided to research the
history of the kua‘äina of Moloka‘i as a prime example of an entire island that
had been bypassed by the mainstream of economic and political change. In
this, the interviews of forty-eight küpuna by Mary Kawena Pukui in 1961 and
1964, preserved on eighty-three half-hour tapes in the Bishop Museum’s
Audio-Recording Collection, served as my primary source. I also used a 1941
unpublished manuscript by Southwick Phelps on the cultural resources of the

296
H A ‘ INA I A M A I : tell the story

island and George Cooke’s book about Moloka‘i Ranch. For the section on the
experiences of early Hawaiian homesteaders, I also used the Mary Kawena
Pukui’s taped interviews, together with manuscripts and reports on the estab-
lishment of the Hawaiian Homes Commission and its program of home-
steading on Moloka‘i.
In 1993 I worked with Jon Matsuoka of the School of Social Work and
Luciano Minerbi of the Department of Urban and Regional Planning and a
committee of Moloka‘i kua‘äina, state officials and representatives of Moloka‘i
Ranch to develop the “Governor’s Moloka‘i Subsistence Task Force Report.”
In June 1993 we worked with the task force to conduct a random sample
survey of the Moloka‘i community regarding the extent and importance of
subsistence activities on Moloka‘i. In July and August 1993 the task force con-
ducted focus groups with subsistence fishers, hunters, and gatherers in Kau-
nakakai, East End, Mauna Loa, and Ho‘olehua, as well as with island-wide
commercial fishermen and lä‘au lapa‘au practitioners. One of the key features
of these focus groups was participatory mapping of subsistence hunting, fish-
ing, and gathering areas, as well as trails, on a topographic map of the island.
It was the first time we had used this method; its success led us to use it again
in our Puna Cultural Impact Study.
The Moloka‘i Subsistence Task Force met through September and October
1993 to balance the findings of the random sample survey and the input from
the focus groups to propose policies and recommendations for the commu-
nity to review. The task force held a community meeting in November 1993
and received additional recommendations, which were incorporated into a
final report in December 1993. I outline this process to convey the breadth of
input from the community that was demanded by the task force in order to
complete the final report. Of any community that I have worked with, the
Moloka‘i community is the most committed to finding ways to control its own
destiny, and it values the community process in order to derive and design a
community-based initiative.
As noted in chapter 5, the survey and the focus groups confirmed that
Moloka‘i continued to be a rural island where the kua‘äina engaged heavily in
subsistence farming, fishing, hunting, and gathering. Among the random sam-
ple group surveyed across the entire island, informants stated that 28 percent
of their food comes from subsistence. Among the Native Hawaiian families
surveyed, informants stated that 38 percent of their food come through sub-
sistence activities. Respondents reported obtaining food acquired through
subsistence activities approximately once a week.

297
chapter seven

Through the study, the Moloka‘i kua‘äina hoped to impress the policymak-
ers and economists that subsistence needed to be acknowledged as an impor-
tant sector of the economy—as important as the market sector and the govern-
ment sector. Economic planning for Moloka‘i and other rural communities
needed to factor in subsistence and the impact on the natural resources relied
upon for subsistence. This report was not meant to sit on a shelf and gather
dust. It was used immediately to introduce and justify legislation to designate
the northeast coast of Moloka‘i, from Nihoa Flats in the east to ‘Ïlio Point in
the west, as a community-based subsistence fishing management area under
the Hui Mälama o Mo‘omomi. The legislature granted communities through-
out the islands the ability to designate subsistence fishing management areas
and established Moloka‘i’s northeast coast as a pilot project from 1995 to
1997. The community worked vigorously with the Department of Land and
Natural Resources to adopt management rules for the pilot project that would
be incorporated as general rules for community-based subsistence fishing man-
agement areas throughout the Islands. However, after the pilot project ended,
the department failed to follow through on the general rules. Numerous rural
communities throughout the state would like to collaborate with Hui Mälama
o Mo‘omomi to protect invaluable nearshore fishing grounds throughout the
state from overfishing and depletion.
The study also formed the basis of testimony in water allocation cases for
the Kamiloloa Aquifer regarding the potential impact increased drilling and
withdrawal of water by Moloka‘i Ranch would have on the subsistence
resources both on land and on the nearshore reefs. The Department of Hawai-
ian Home Lands also responded to the study by placing management of
Hawaiian Home Lands hunting grounds under the management of the Molo-
ka‘i homesteaders.
In 1998 the Moloka‘i community embarked on a major effort to boost the
island’s economy. They applied to the Department of Agriculture for a rural
empowerment zone grant. There were hundreds of community meetings, and
I assisted in synthesizing the community’s goals, objectives, and economic
projects throughout the process. Although the Moloka‘i community did not
get the empowerment designation, in 1999 the federal government desig-
nated it an enterprise community (EC). It will receive $250,000 per year for
ten years, a total of $2.5 million, to partner with private entities and govern-
ment agencies to leverage additional monies for forty identified economic
development projects. In 2003 the community incorporated as Ke Aupuni
Lökahi to carry out the projects and programs as an enterprise community.

298
H A ‘ INA I A M A I : tell the story

Designed by a broad cross section of the island for the purposes of the
grant, the Ke Aupuni Lökähi Enterprise Community is guided by an outstand-
ing vision statement, quoted in chapter 5 on Moloka‘i, to guide the island’s
economic development in the twenty-first century. The statement on Com-
munity Values elaborates upon the meaning of the vision statement. It explains
that the vision statement reflects the concept of aloha ‘äina and the belief that
the land or ‘äina is alive and must be respected, treasured, nurtured, and pro-
tected if it is to be productive. Hawaiians believe that in return for good stew-
ardship, the land sustains the people who care for it. The sacred and depen-
dent relationship between the land and the people sustained Moloka‘i for a
thousand years, and the vision statement affirms aloha ‘äina as the bedrock
value upon which Moloka‘i’s economic recovery will be founded:

Throughout the centuries before western explorers arrived, Moloka‘i


people knew every different wind, rain and cloud formation by its own
given name. Fish were never taken out of season. Prayer rituals accompa-
nied every step of farming, from planting to harvest. Fresh water sources
were stringently protected, and an abundance of water was regarded as the
highest symbol of wealth. In return for their protection of the land, the
people of Moloka‘i became renowned for their ability to produce abundant
quantities of food, and the island acquired the name ‘äina momona, abun-
dant land. The Vision Statement expresses our community’s belief that we
can only restore this island’s legendary productivity if we become more
vigilant guardians of its resources.8

By the time of the 1998 grant project, I commuted weekly to Moloka‘i and
was a part-time resident of Ho‘olehua. In January 2003 Ke Aupuni Lökahi set
up a special committee to engage in a community-based master land use plan-
ning process for the Moloka‘i Ranch, which is incorporated as Moloka‘i Prop-
erties Limited (MPL). For more than 100 days and with a total of 300 partic-
ipants, five committees considered hundreds of planning issues that would
affect the lives of the people of Moloka‘i. I participated in the Cultural Issues
Committee. During the summer of 2004 the proposed plan was completed,
and a process of facilitated and community meetings began. Under the plan,
MPL will donate 26,000 of the 65,000 acres it owns to the Moloka‘i Land
Trust. These lands generate annual rental revenues of $250,000, which can be
used to fund the operations of the land trust. An additional 29,000 acres will
be permanently protected in an agricultural and rural reserve and under exist-
ing conservation easements. In return, MPL seeks community support to

299
chapter seven

develop 200 luxury home lots on the last pristine shoreline on Moloka‘i, along
Lä‘au Point. The community is in the process of weighing the benefits and
drawbacks of the overall package.
In January 2004 Ke Aupuni Lökahi Moloka‘i asked me to conduct a Molo-
ka‘i Responsible Tourism Study. I launched the study in January 2005. My
task was to interview and hold focus group discussions with Moloka‘i business
operators, subsistence farmers and fishermen, community organizations, com-
munity contacts for sports and high school events, and Ke Aupuni Lökahi in
order to design a five-year visitor plan. The plan follows community guide-
lines for stewardship of the island’s environmental, cultural, and infrastructure
resources. The challenge is to develop tourism on Moloka‘i while still uphold-
ing Moloka‘i’s renown as “the last Hawaiian island,” which is the reason why
families continue to live on Moloka‘i and why visitors are attracted to its
shores.
As the Ke Aupuni Lökahi initiatives and the Moloka‘i Ranch planning
indicate, the Moloka‘i community is still intent on shaping its own destiny,
keeping Moloka‘i Hawaiian, and actively engaging the community in plan-
ning its own future. Moreover, the community constantly leans upon its tra-
ditional strengths, reflected in the famous sayings for the guidance and direc-
tion of Moloka‘i. These are described in the following closing thoughts for
Moloka‘i:

The saying, “Moloka‘i Nui A Hina” or Great Moloka‘i, Child of Hina,


affirms that Moloka‘i, like a human child, was born to a mother and father:
Wäkea, god of the sky, and Hina, goddess of the moon and weaver of the
clouds. This traditional legend of origin establishes that the island of
Moloka‘i, like a child, is small and fragile — unlike a large continent. The
resources of an island are finite, and these finite resources need to be
nurtured by the island’s “family” if the people are to be strong, healthy,
and prosperous. Many of the people of Moloka‘i trace their roots on the
island back to antiquity, making the island an integral part of their
ancestral family. Moloka‘i’s modern-day stewards have a special responsi-
bility to care for the island as they would care for a member of their own
family—a responsibility bequeathed to them by Hina, birth mother of this
island.
“Moloka‘i Pule O‘o,” island of powerful prayers, is another traditional
name for Moloka‘i. In ancient times this name inspired fear and respect
throughout Hawai‘i, because it was based on the island’s reputation as a

300
H A ‘ INA I A M A I : tell the story

training ground for the most powerful priests in the islands. Legends say
that the people of Moloka‘i could drive invading armies from their shores
by simply uniting in prayer. This name recognizes Moloka‘i as an ancient
center for learning, and honors the spiritual strength of Moloka‘i’s people,
and their historic sovereign control over the island. Although Moloka‘i is
not self-governing today, as discussed below, her people are nevertheless
respected for their ability, thus far, to protect the Hawaiian culture, subsis-
tence lifestyle and the natural resources upon which they are dependent.
Their feat has been accomplished by combining an intimate knowledge of
the island’s resources with strength of character and fearless determination
to deal with threats to their environment and lifestyle. The enduring
description of Moloka‘i as “the last Hawaiian island” affirms the success
of the community in protecting the Hawaiian way of life as the core of
the island’s multi-ethnic, close-knit society.
‘Äina Momona: Land of Plenty: Before Western contact, the economy
of Moloka‘i was agricultural and centered on inshore aquaculture, the
cultivation of various crops, fishing, hunting and gathering. As a result of
the industry of her people, Moloka‘i, with its extensive protected reefs and
fishponds, gained a reputation as the land of “fat fish and kukui nut relish.”
The “fat fish” came from Moloka‘i’s fishponds and the waters surrounding
the island. The mention of “kukui nut relish” refers to the lush resources
of the land. The island as a whole was popularly called “ ‘Äina Momona”
or “Land of Plenty” in honor of the great productivity of the island and
its surrounding ocean.9

Kanaloa
And now, I will close this mo‘olelo where I opened it, with my connection to
Kaho‘olawe/ Kanaloa. As mentioned above, I finally made the ocean crossing
to Kanaloa in November 1984 with Uncle Harry Mitchell, a group of my stu-
dents from the ethnic studies “Hawaiians” course, and the Lu‘uwai family. At
first, I didn’t feel the spirit of the island. There was too much to do, between
cooking, hiking, keeping track of my students, and thatching the Hale
Halawai. When it was time to leave, not everyone could fit on the boat and a
small group of eight or ten of us stayed behind an extra day and night to wait
for another boat to pick us up. It was then, when it grew quiet except for the
natural sounds of the wind and the surrounding ocean, that I sensed the spirit
of the land. And in the evening, when the kua‘äina from Hawai‘i and Maui

301
chapter seven

and Moloka‘i opened up and talked story around the campfire, I felt the pres-
ence of our ancestors in their stories.
When I went back to O‘ahu, the island kept calling inside me to return and
to get involved in the struggle to stop the bombing. I did return and contin-
ued to return, as I will for the rest of my life. I have been steadfast in the work
to stop the bombing and to heal the wounds of the island. And I began to
honor Lono in the Makahiki and Kanaloa in private offerings of ho‘okupu.
The chapter on Kaho‘olawe reflects my research and my experience with the
many members of the ‘Ohana and those who have visited its shores and been
touched by the island. Although tremendous gains have been made in healing
the island, many obstacles remain and we in the ‘Ohana must remain vigilant
and steadfast and work in partnership with the Kaho‘olawe Island Reserve
Commission, which has the financial, personnel, and political resources to
establish and protect Kanaloa as a cultural reserve.
Though title to Kaho‘olawe/ Kanaloa was transferred to the State of
Hawai‘i in May 1994, the U.S. Navy retained control over access to the island
in order to conduct an omnibus cleanup of the island from November 10,
1993, through November 11, 2003. After fifty years of use as a military
weapons range, Kaho‘olawe/ Kanaloa’s 28,800 acres were contaminated with
shrapnel, target vehicles, and unexploded ordnance. The U.S. Navy signed an
agreement with the State of Hawai‘i to clean up 30 percent of the island’s sub-
surface of ordnance. In 1993 the U.S. Congress appropriated $460 million for
the U.S. Navy to fulfill this obligation. The U.S. Navy contracted Parsons-
UXB Joint Venture to conduct what is acknowledged to be the largest unex-
ploded-ordnance remediation project in the history of the United States.
Over 10 million pounds of metal, 370 vehicles, and 14,000 tires were removed
from the island and recycled. However, rather than clearing 30 percent of the
island to a depth of four feet, the Navy contractor cleared only 2,650 acres
( 9 percent) of the island’s subsurface. Another 19,464 acres (68 percent) of
the island’s surface was cleared of ordnance, but 6,686 acres (23 percent) of
Kanaloa has not been cleared at all. Disturbingly, the U.S. Navy can guaran-
tee only that it is 90 percent confident that 85 percent of the ordnance in the
2,650 acres or 9 percent was cleared of ordnance to a depth of four feet.
What does this mean? Access to our beloved island will continue to be lim-
ited to the “cleared” areas, which have the highest priority for cultural activ-
ities and revegetation projects. These include Hakioawa and Hakioawa Iki,
where the ‘Ohana maintains its primary base camp, has restored heiau and

302
H A ‘ INA I A M A I : tell the story

shrines, and has established new cultural sites. Moa‘ulanui, where the KIRC
has established its staging area for the revegetation of the island, is accessible.
Moa‘ulaiki, the traditional center for the training of navigators in traditional
way-finding arts, had the highest priority for clearance. Honokanai‘a was the
central staging area for the clean-up and the center of the communications
system, and was one of the first areas cleared of ordnance. Kealaikahiki, from
which long voyages between Hawai‘i and Kahiki were launched, was identified
as an ideal site for contemporary training in celestial navigation and cleared. In
October 2004 a special platform for navigators to observe the position of the
North Star and Southern Cross over the horizon relative to the piko or cen-
tral point of Kealaikahiki was constructed and dedicated by Master Navigator
Mau Pialug, representatives of each of the Hawaiian voyaging canoes, the Pro-
tect Kaho‘olawe ‘Ohana, and the KIRC.
The adjoining beach at Keanakeiki, where the closing ceremonies for the
annual Makahiki are conducted, will continue to be accessible. Kühe‘eia, cen-
ter of the ranching operations in the early twentieth century; and Kaulana,
site of the kingdom’s prison settlement and school and an ideal planting area;
and Ahupü, another ranching site and location of the island’s largest petro-
glyph field and obsidian glass quarry, were all cleared. Honoko‘a, location of
several fishing shrines and the landing place of King Kaläkaua when he vis-
ited the island, was cleared so that the fishing traditions of the island could be
practiced there and passed on. In addition, a cross-island road and trails link
Hakioawa and Kuhe‘eia to the central part of the island over to Honokanai‘a
and out to Kealaikahiki and Keanakeiki. Portions of a round-the-island trail
were cleared, such as at Kanapou Bay and its surrounding cliffs, where there
are observation points to read the weather and the natural elements. Even the
uncleared areas can be accessed by small groups and with escorts who are
trained to detect and handle unexploded ordnance. Activities will continue to
focus on the healing and restoration of the cultural and natural resources of
Kanaloa and reviving Native Hawaiian spiritual and cultural customs and
practices. The limited clearance of ordnance means that the island will not be
open for general recreational or commercial activities, nor for resorts, golf
courses, or subdivisions.
Is this the end? Under state law, when the Navy transferred control of
Kanaloa to the state, it was to be held in trust for transfer to the sovereign
Hawaiian entity when it is reestablished and recognized by the federal and
state governments. The ‘Ohana is committed to holding the Navy account-

303
chapter seven

able for clearing more of the island as our use and needs expand so that the
island can ultimately be returned to the sovereign Hawaiian entity for safe and
meaningful use as a cultural reserve.
I choose to end this mo‘olelo with the words of Noa Emmett Aluli from
the foreword of Kaho‘olawe Nä Leo o Kanaloa because they reflect my own
thoughts and experiences as we have shared our lives and aloha with Kanaloa
and the ‘ohana who remain connected to this island and to all our islands:

On Kaho‘olawe, we’ve been able to live together as Hawaiians. We’ve been


able to practice the religion and to carry on the traditions we’ve learned
from our küpuna, our elders. In doing this, we connect to the land, and
we connect to the gods. We call them back to the land and back to our
lives . . . We commit for generations, not just for careers. We set things
up now so that they’ll be carried on. We look ahead together so that many
of us share the same vision and dream. To our next generations we say,
Go with the spirit. Take the challenge. Learn something. Give back.10

304
 appendix i 

1851 Petition from Puna Native Hawaiians


to Extend the Deadline to File a Land Claim

puna
O makou ka poe makaainana o Puna nei, ke noi aku nei makou i ka aha olelo

1 O ko makou mau kuleana i komo ole i loko o na Lunahoona, e hookomo koke ia


mai me ka uku ole,
2 E waihoneole ia na kula me ka Alodio ole ia
3 Ona aina aupuni i haawi wale ia i ka poe makemake maoli i ka hana
4 E pau ka noho nui ‘ana o na kanaka ma ka hale hookahi
5 E hoola ia ke kanawai o ka poe palaualelo
6 E pau ka male ana o ka luahine me ka mea opiopio pela ou no ka ele ma kule me
ka mea opiopio

puna
We are the common citizens of Puna and we petition the legislature that:

1 That our kuleana that have not been entered with the Land Commissioners be
immediately entered without fees.
2 That the kula areas be left untouched without be owned in fee simple.
3 That government lands be given only to those who really want to work.
4 That people not be allowed to occupy a single house in large numbers.
5 That a law dealing with laziness be brought up.
6 That marriages between old people and young people be abolished.
(trans. w. h. wilson, october 5, 1977)

305
 appendix ii 

Number of Males Who Paid Taxes in Puna in 1858

over 20 years of age under 20 years of age

1 Apua 4 1
2 Kealakomo 32 15
3 Panau Nui 15 4
4 Paunau Iki 1
5 Laepuki 33 6
6 Kamoamoa 3
7 Pulama 9
8 Kahaualea 36 4
9 Kapaahu 12
10 Kupahua 6 1
11 Kalapana 87 7
12 Kaimu 60 over 25 5
13 Ua Kona 1 1
14 Kehena 32 6
15 Keekee 3 1
16 Kamaili 10 1
17 Kaueleau 3 1
18 Kanane 22 5
19 Opihikao 13 1
20 Iililoa 4 1
21 Kauaea 33 1
22 Malama 5 1
23 Kaukulau 3
24 Keahialaka 11
25 Pohoiki 33 1
26 Oneloa 6
27 Ili kipi kaa Inaina Papoi kou 1
28 Laepaoo 4

306
appendixes

over 20 years of age under 20 years of age

29 Pualaa 5 1
30 Kapoho 49 8
31 Kula 47 8
32 Puua 12 3
33 Koae 1 30 7
34 Koae 2 9 1
35 Kanekiki 1 1
36 Halepuaa 22 4
37 Kahuai 8
38 Waawaa 8
39 Honolulu 3 1
40 Waiakahiula 31 3
41 Keonepoko 3
42 Halona 4
43 Popoki 2
44 Makuu 37 7
45 Keaau 56 15
46 Olaa 85 8

307
 appendix iii 

Moloka‘i, Petition of July 2, 1845

he palapala hoopii na ka poe makaainana


Molokai, hoouna ia ku i ka malama o Julai, 1845.
Aloha nui ka Moi hanohano o ko makaou mau kupuna mai ka po mai a hiki mai ia
makou i na keiki. Pela no ke Kuhina nui o ke Aupuni Hawaii nui, a me na lii o kou
aupuni a pau. Eia ko makou manao noi aku ia oe e ko makou Moi a me ko makou mau
alii malalo ou maloko o ka ahaolelo.

1 No ke kuokoa ana o kou aupuni e ka Moi III aole o makou makemake e noho luna
na haole au i hoonoho ai maluna o ke Aupuni Hawaii.
2 Aole o makou makemake e hoohiki na haole i kanaka ( Hawaii).
3 Aole o makou makemake e kuai aku oe i kekahi apana aina o kou Aupuni i na haole.
4 Aole e kauia ka Auhau pohihihi i kau poe huna lepo.

Ma keia mau manao o makou a hoike aku la ia ia oe e ka Moi a me kou poe alii.
Ke kakau nei makou i ko makou mau inoa malalo iho.
Eia ka huina o na inoa (1344*)

Aloha ka mea hanohano i hoonohoia ma ka mole o ka Moi Kamehameha II a me ka


Moi Kamehameha I.
Na kou poe Kauwa makaainana huna lep o kou Paeaiana
Na Keaumaea i haawi aku.

petition by the common people of moloka‘i


Moloka‘i, Sent in the month of July, 1845

Greetings Honorable King of our ancestors from the time of the gods (pö) down to
us the descendants, as well as to the Kuhina nui of our Hawaiian Kingdom and all the
ali‘i of your entire nation.
The following is what we desire to request of you, our King, and our ali‘i under you
in the legislature.

1 For the independence of your nation, King (Kamehameha) III, we do not want the
haole you have appointed over the Hawaiian government to serve as officials.

308
appendixes

2 We do not want haole to be made naturalized Hawaiian citizens.


3 We do not want you to sell any portion of your nation to haole.
4 Do not place confusing taxes upon your humble people (huna lepo—bits of earth).

May these feelings of ours be shown to you, Your Majesty, and to your ali‘i.
We sign our names.
The following is the total amount of names 1344.*
Aloha honorable one who has been appointed to the root of King Kamehameha II
and King Kamehameha I.
Your humble servants, the commoners of your islands,
Given by Keaumaea
[trans. w. h. wilson, august 10, 1977]

1 Lokane 2 Kulau 3 Kapuhenehone 4 Kaikaina


5 Kalaukaa 6 Kalaumano 7 Mahi 8 Naihe
9 Owao 10 Kaakau 11 Lae 12 Kaapuiki
13 Makalilio 14 Kauipuniho 15 Wahine 16 Kalia
17 Kaheewahine 18 Kuakea 19 Kaneinei 20 Kaheana
21 Kalawaia 22 Nalua 23 Peleleleu 24 Kaunahi
25 Hoaehae? 26 Kawalua 27 Pookee 28 Kauokuu
29 Maiawa 30 Kaakau 31 Paaluhi 32 Kalele
33 Uwe 34 Kauwehia 35 Kahualii 36 Kokoniho
37 Kaiwiahuula 38 Kailianu 39 Haipu 40 Laakapa
41 Molokai 42 Makana 43 Kekiiokalani 44 Petero
45 Kawaa 46 Kahookuu 47 Kaleoku 48 Kawaaliilii
49 Kaila 50 Kaula 51 Keanoalii 52 Pauloa
53 Kaapolani 54 Amalu 55 Kamailuna 56 Kuaili
57 Naleilehua 58 Kauhimahalupe 59 Kauao 60 Kama
61 Paana 62 Akahi 63 Wahineaukai 64 Hamakaia
65 Lahana 66 Pihanui 67 Kaleiwahine 68 Kahaomaneo
69 Kuaana 70 Kaena 71 Papu 72 Kaehukanakaliilii
73 Muoouo 74 Kealakai 75 Nakahili 76 Kaikapu
77 Kamakaunui 78 Kawaalauioili 79 Kau 80 Kealo
81 Kailikole 82 Kau 83 Puapua 84 Daniela
85 Upai 86 Ukuala 87 Kahaikupuna 88 Mahoe
89 Kioi 90 Kaiwipalaoa 91 Kapahi 92 Kamamaka
93 Kamoku 94 Kamaiahulu 95 Kalikelike 96 Kama
97 Kalama 98 Kalahale 99 Kaehumakalani 100 Kauwe

309
appendixes

101 Kamalino 102 Kalawe 103 Kalepale 104 Kaialau


105 Kaahi 106 Kaluaokamoku 107 Kekahuna 108 Kekolohe
109 Kekahuna 2 110 Keliikanakaole 111 Timoteo 112 Pua
113 Kuli 114 Kumaikalala 115 Laukua 116 Laa
117 Helekaihuelani 118 Lono 119 Lupua 120 Malia
121 Mahaole 122 Mahiai 123 Manoa 124 Maa
125 Meeau 126 Nahuaai 127 Kaakau 128 Naeole
129 Kaole 130 Keawekane 131 Kalua 132 Waimea
133 Mahi 134 Kekaha 135 Kaii 136 Naahu
137 Kaohilo 138 Kani 139 Kuokoa 140 Pehu
141 Pehu 2 142 Apuia 143 Kaaea 144 Kamokuahiole
145 Kuahamano 146 Puna 147 Kaneliilii 148 Kamoku
149 Piha 150 Keaka 151 Paanoio 152 Pualoha
153 Pinea 154 Nahoalawelawe 155 Kailaa 156 Keahi
157 Kanepauloa 158 Kaimu 159 Paku 160 Haleloa
161 Kalama 162 Naono 163 Pua 164 Kapohuli
165 Lonopuawela 166 Pilaliohe 167 Keawe 168 Malue
169 Kekahuna 170 Okou 171 Kauluahewa 172 Nalaukau
173 Kauhai 174 Kaaulaukini 175 Hanapilo 176 Koele
177 Kauwelani 178 Kalua 179 Kala’ 180 Pahaha
181 Kuheleloa 182 Ehu 183 Alaala 184 Pulehu
185 Mahu 186 Palahuli 187 Palena 188 Kawiino
189 Ailepo 190 Kau 191 Kuanea 192 Keaweeli
193 Naeleele 194 Keaweelie 195 Kahaka 196 Opunui
(+1 grant)
197 Nanoha 198 Aewai 199 Palahe 200 Manuai
201 Kilauakea 202 Kamai 203 Nainea 204 Kamaunu
205 Kanakaole 206 Kanaau 207 Kaneaikai 208 Hihipalani
209 Kuhiheauau 210 Kamai 211 Pupule 212 Kahaule
213 Luha 214 Pehipolani 215 Kahaule 2 216 Kekupuwale
217 Kanunu 218 Hilea 219 Naili 220 Maoma
221 Kanakake 222 Kanakaloa 223 Kapule 224 Paele
225 Aea 226 Kailikea 227 Kahuaono 228 Haole
229 Kaanaana 230 Kaleikau 231 Kaaimoku 232 Kaka
233 Nalii 234 Kahuliaikau 235 Luhia 236 Elemakule
237 Kaiamalani 238 Kale 239 Heleonalama 240 Opae
241 Nakukulani 242 Paele 243 Kapua 244 Waikonale
245 Uilama 246 Kamai 247 Kaniau 248 Kuapa
(+1 grant)
310
appendixes

249 Oopa 250 Kanakaole 251 Kaahui 252 Meahole


253 Akaaina 254 Halama 255 Wahalumilumi 256 Maikai
257 Maikuli 258 Kaiu 259 Makaole 260 Kaiahauna
261 Kamaka 262 Kalino 263 Kokaehe 264 Kaaiaiamoku
265 Oleole 266 Upai 267 Nakaula 268 Kamakaalua
269 Kahaulepo 270 Kaheana 271 Nalu 272 Mahoe
273 Koloka 274 Ika 275 Wahineaela 276 Ana
277 Ohina 278 Kauwiki 279 Kaakau 280 Elisai
281 Ualanui 282 Kunui 283 Kanui 284 Mokeha
285 Molia 286 Kaaikapuu 287 Kaoha 288 Uihaai
289 Naopu 290 Paiahi 291 Puapuu 292 Heneriaka
293 Paaluhi 294 Hihia 295 Puahiohi 296 Kawaihua
297 Lepelua 298 Kawaipi 299 Kaai 300 Laie
301 Loiloi 302 Kaakua 303 Nakai 304 Keawe
305 Kuheleloa 306 Kahue 307 Kalua 308 Kaina
309 Naiole 310 Kanoho 311 Kanioi 312 Mahoe
313 Hopu 314 Kaaikaula 315 Kalualoa 316 Kapalena
(+1 grant)
317 Kuaea 318 Awala 319 Inoino 320 Kealakahi
321 Hawaii 322 Kaiwi 323 Kekua 324 Namaile
325 Kaualo 326 Aikino 327 Wahalepo 328 Daniela
329 Kelaukila 330 Kamala 331 Kane 332 Kaniho
333 Kanakanui 334 Kaoha 335 Kahua 336 Kanakaole
337 Waimoe 338 Penopeno 339 Naonohi 340 Kama
341 Lima 342 Kanene 343 Kauhanui 344 Nahoiha
(+1 grant) (+1 grant)
345 Kauhi 346 Kapule 347 Kanui 348 Kaaukai
349 Kahapana 350 Kaai 351 Kanoao 352 Kaheiki
353 Kahiawa 354 Kukeaa 355 Palapala 356 Kapai
357 Ihi 358 Mulihele 359 Lawelawe 360 Aoao
(+1 grant)
361 Kanakaole 362 Kekolohe 363 Naili 364 Kawaiwai
365 Halualani 366 Ohule 367 Kikoikoi 368 Naili
369 Kahooaha 370 Pauahi 371 Kameheu 372 Kawainui
(+1 grant)
373 Keanini 374 Paele 375 Kaaea 376 Ku
377 Keleau 378 Makaila 379 Kapaiwahea 380 Kaili
381 Kahaiola 382 Luia 383 Opapa 384 Kauahooloku
385 Mose 386 Kuhoe 387 Nakoholua 388 Malaihi
311
appendixes

389 Kikau 390 Kahookano 391 Kaiamoku 392 Namakaelua


(+1 grant)
393 Maalahia 394 Kauaawa 395 Kanehaole 396 Kaluau
397 Kaleo 398 Kukahana 399 Kuaana 400 Kewalo
401 Kekahuna 402 Kapu 403 Nahia 404 Nahale
405 Pupule 406 Mahu 407 Mauoha 408 Laa
409 Moewaa 410 Paahao 411 Lolo 412 Pupuka
413 Puaa 414 Kekau 415 Kamaunu 416 Kupololu
417 Nuipoohiwi 418 Lani 419 Luka 420 Hooipo
421 Pio 422 Kaolelo 423 Kaninau 424 Ika
425 Pupuka 426 Malao 427 Kapae 428 Muouou
429 Lolo 430 Kahili 431 Kahunui 432 Kepaa
433 Kahilina 434 Puali 435 Keneau 436 Kane
437 Kailaa 438 Kaakau 439 Ilikealani 440 Hawea
441 Kaluahine 442 Ohule 443 Puaahaliu 444 Kanemanaole
445 Opu 446 Ehu 447 Haole 448 Kauku
449 Nalaalaau 450 Kulohaaina 451 Kauhi 452 Solomona
453 Kealia 454 Kaui 455 Aberakama 456 Mahina
457 Piipii 458 Makalii 459 Kaohia 460 Umauma
461 Lulii 462 Paahao 463 Kaili 464 Wahineaukai
465 Pahoa 466 Kawakea 467 Kahalau 468 Pau
469 Poke 470 Pueaina 471 Honu 472 Kahue
473 Paele 474 Lakaba 475 Kahaohuli 476 Aukupu
477 Naupa 478 Maeikaukane 479 Nakoaolelo 480 Wahinekoliola
481 Ninihua 482 Halulu 483 Hulu 484 Napaepae
485 Kauhi 486 Piiaimoku 487 Nakaa 488 Kahuha
489 Paaluhi 490 Ioane 491 Aukai 492 Iopa
493 Pukuhe 494 Oopa 495 Ihu 496 Ueuele
497 Hemahema 498 Haipu 499 Kamakahi 500 Keliihooleia
501 Kaohimaunu 502 Kalele 503 Kaheana 504 Kaiue
505 Kalaauala 506 Kahukanokoliilii 507 Kualualii 508 Kila
509 Kaiakea 510 Kawelo 511 Kahiau 512 Luuloa
513 Mahiai 514 Makaihuia 515 Moaku 516 Nawaaloloa
517 Pulehi 518 Pohuehue 519 Waha 520 Mahiai
521 Mahina 522 Napihelua 523 Kahalekai 524 Hokai
525 Naonealaa 526 Namakakaia 527 Kamai 528 Kameheu
529 Kulani 530 Namakaokeawe 531 Kewalo 532 Kelakela
533 Kaloa 534 Opuloa 535 Kalauahea 536 Haole

312
appendixes

537 Paikaniau 538 Kakii 539 Nahuaai 540 Nahuaai


541 Kuohau 542 Inoino 543 Haa 544 Kahau
545 Inui 546 Mahoe 547 Kanalualii 548 Aweka
549 Kuaeau 550 Uwelani 551 Ualani 552 Aloi
553 Kuai 554 Kane 555 Moi 556 Kumaikalala
557 Kalawe 558 Kau 559 Kaluaokamoku 560 Kekahuna
561 Naliikawaa 562 Ouwe 563 Wahineino 564 Keakui
565 Nuka 566 Makaino 567 Alaala 568 Wahinekapu
569 Kaumaka 570 Wahinealii 571 Paku 572 Kepaa
573 Naleilehua 574 Iliwai 575 Kekaha 576 Kamakahelei
578 Kapuli 579 Hanaloa 580 Kaakua 581 Lihue
582 Makahoohano 583 Kaiwi 584 Mahu 585 Waimea
586 Kolia 587 Kaeha 588 Laukua 589 Kalahale
590 Aukai 591 Kaumakua 592 Kaianui 593 Kalokane
594 Kaipuniho 595 Naihe 596 Kaikaina 597 Manoa
598 Uluhoa 599 Mahi 600 Kalaumano 601 Kua
602 Owau 603 Kahakukaalina 604 Imahiai 605 Kuloa
606 Kuanea 607 Kanalualii 608 Kepaa 609 Inui
610 Kaluau 611 Kahoowaha 612 Wahinealii 613 Kaumaka
614 Mahiai 615 Kuohao 616 Kawahahei 617 Nahuaai
618 Pioe 619 Aloalo 620 Aukai 621 Wahinekapu
622 Luina 623 Kaha 624 Makaino 625 Kealohi
626 Nawahine 627 Ponoino 628 Kelakela 629 Kahaleolani
630 Waihinalo 631 Kai 632 Lona 633 Pehu
634 Kahuakailuhi 635 Kamanohelii 636 Mahoe 637 Pohaku
638 Mahoe 639 Hea 640 Puaohi 641 Kahiko
642 Kalua 643 Lea 644 Paiuma 645 Hikihewa
646 Kelono 647 Keanoalii 648 Kalanahea 649 Kahalau
650 Kaili 651 Kale 652 Kaloa 653 Kaiwinui
654 Kaianui 655 Kahinu 656 Hoki 657 Kaie
658 Namakaokaia 659 Namealoaa 660 Laakapu 661 Namakaokeawe
662 Kalani 663 Kameheu 664 Kamaka 665 Kaha
666 Wahahee 667 Maui 668 Pakaha 669 Kahue
670 Kaia 671 Umi 672 Keliikuewa 673 Ukuula
674 Naomi 675 Pukuinui 676 Maleka 677 Kiekie
678 Alapai 679 Keawa 680 Niho 681 Pupule
682 Kahoomaika 683 Kapalo 684 Pulehuhua 685 Kanui
686 Kehuluaulani 687 Kaialau 688 Hinaele 689 Kiula

313
appendixes

690 Kaumu 691 Maihuki 692 Keliiaukai 693 Lakalo


694 Kahuna 695 Paahao 696 Kanawi 697 Kuaai
698 Kamakahoo- 699 Malae 700 Lehua 701 Kau
hano
702 Poopuu 703 Paahao 704 Upai 705 Kahaule
706 Naone 707 Kualawa 708 Napae 709 Kuku (+1 grant)
710 Kauwe 711 Paulaulaiki 712 Kaikulani 713 Kanuku
714 Oiai 715 Kaahaialii 716 Kalualohi 717 Paakua
718 Kekaha 719 Hanaloa 720 Kuaeou 721 Ilo
722 Haole 723 Nalehua 724 Naonealaa 725 Naholowaa
726 Wahineaea 727 Haipu 728 Ikaa 729 Makapo
730 Puana 731 Kapa 732 Kaialau 733 Kalaipaka
734 Nakalina 735 Healani 736 Aki 737 Mueau
738 Kekahuna 739 Lono 740 Kalahale 741 Kaukaliu
742 Papalana 743 Kuli 744 Kama 745 Kama
746 Kapiolamau 747 Paahao 748 Paoloa 749 Puapua
750 Waiale 751 Kahananui 752 Kaaiau 753 Halulu
754 Puamana 755 Iliwai 756 Luahine 757 Pauma
758 Kauaeau 759 Kaupuna 760 Nuuu 761 Holi
762 Hakii 763 Kama 764 Kaumakae 765 Kaa
766 Kumu 767 Kua 768 Kalua 769 Peihalu
770 Heke 771 Luhulu 772 Naoka 773 Wahineine
774 Kali 775 Keakui 776 Puu 777 I
778 Alalo 779 Halawai 780 Kaale 781 Amuamu
782 Naowa 783 Nohunohu 784 Palema 785 Auwaepaa
786 Kumahoe 787 Oehu 788 Paku 789 Ahi
790 Makaula 791 Kapaahu 792 Kuala 793 Kowilikopaa
794 Apua 795 Pahi 796 Hoohano 797 Kane
798 Nahaka 799 Kaehana 800 Kea 801 Moloku
802 Hakuole 803 Kawelo 804 Puupuu 805 Pipi
806 Kaule 807 Kuapuu 808 Kamauoha 809 Paele
(+1 grant)
810 Kahue 811 Kahaaweai 812 Kuluwaimaka 813 Kahuine
814 Hulihee 815 Kuhihewa 816 Nahoaai 817 Kahu
818 Kawelo 819 Muolo 820 Hilo 821 Paele
822 Paele 3 823 Kaheiau 824 Kaheaaku 825 Koemakaia
826 Pala 827 Ihu 828 Keanui 829 Paele 4

314
appendixes

830 Kaululaau 831 Onei 832 Hua 833 Paiwi


834 Kapale 835 Kalaeone 836 Mauku 837 Kuanoni
838 Wawae 839 Kupuwale 840 Nukai 841 Waiki
842 Mahoe 843 Kaumauina 844 Miki 845 Kawai
846 Keo 847 Paumanao 848 Nauaua 849 Lauae
850 Paakea 851 Kaaiia 852 Kanae 853 Kumuhonua
854 Mahoe 855 Kuaiwa 856 Kauwanui 857 Auweka
858 Aa 859 Inoino 860 Imahiai 861 Ikoa
862 Opuloa 863 Ualani 1 864 Ualani 2 865 Ualani 3
866 Uuoe 867 Uu 868 Haa 869 Haole
870 Haole 2 871 Kealani 872 Kaahui 873 Kaahu
874 Kahue 875 Kakii 876 Kahipa 877 Opanui
878 Kaea 879 Kaanui 880 Poonui 881 Minamina
882 Kekahuna 883 Kaniau 884 Kanikau 885 Kukapu
886 Wahinehue 887 Kaukini 888 Wahinehue 889 Namilo
890 Naimihale 891 Keala 892 Kaae 893 Kahula
894 Makaole 895 Kaialiilii 896 Kaililua 897 Nakao
898 Kaolala 899 Piko 900 Keha 901 Nahooikaika
902 Nakeokeo 903 Kaheananui 904 Kaiwa 905 Naanana
906 Kahaaheo 907 Kekua 908 Kawelowahine 909 Paiwa
910 Kahapu 911 Maneo 912 Mahiai 913 Moo
914 Kanakaole 915 Kualii 916 Kaulana 917 Pahia
918 Kekuinaa 919 Kahikanaka 920 Kakani 921 Naoo
922 Waialua 923 Nahaupia 924 Kekahuna 925 Nikipala
926 Nakala 927 Kepio 928 Kekahuna 929 Kaai
930 Kaakau 931 Kahai 932 Kainui 933 Wahauwia
934 Opiopio 935 Kahina 936 Waianu 937 Nakihei
938 Loika 939 Ohimai 940 Mukoi 941 Maihu
942 Paalou 943 Kahioumae 944 Nakai 945 Kekuewa
946 Ieremia 947 Kahoeua 948 Naki 949 Kaupalolo
950 Kauia 951 Pimoa 952 Kanahele 953 Mahoe
954 Kamai 955 Kaai 956 Napela 957 Kahue
958 Hawaii 959 Makapo 960 Kapukila 961 Kahema
962 Nakahuna 963 Kaiaiki 964 Kaunuino 965 Kukanalua
966 Pauahi 967 Palama 968 Upai 969 Keaa
970 Naahuelua 971 Kanewanui 972 Kaneheana 973 Kaneakua
(+1 grant)

315
appendixes

974 Pokepa 975 Maiola 976 Koihoomoe 977 Paana


978 Pahio 979 Keawe 980 Kukoa 981 Kaukaukapa
(+2 grants)
982 Kalilikane 983 Kauhi 984 Solomona 985 Iosepa
986 Kaia 987 Lolo 988 Kaailaau 989 Kaneelele
990 Iimona 991 Ahulili 992 Kaniu 993 Kapopule
994 Koowahie 995 Kanakanui 996 Kaeku 997 Poohina
998 Heehua 999 Pali 1000 Puhikea 1001 Kaialawaia
1002 Daniela 1003 Waialoha 1004 Naohe 1005 Paauao
1006 Kaihe 1007 Kaakau 1008 Palima 1009 Palima 2
1010 Kanae 1011 Ukunui 1012 Upai 1013 Makuai
1014 Pukuha 1015 Kaihumua 1016 Omino 1017 Kaloulu
1018 Puli 1019 Haole 1020 Kanui 1021 Puhaumea
1022 Kaheleino 1023 Kalamaikai 1024 Kumalaua 1025 Kaha
1026 Paipaiku 1027 Naoha 1028 Kawehena 1029 Loheau
1030 Kaialuau 1031 Mahoe 1032 Hanunu 1033 Palapu
1034 Hainoa 1035 Kauiki 1036 Keaki 1037 Kamoku
(+1 grant)
1038 Naonea 1039 Ai 1040 Kahiapaiole 1041 Kalapauila
1042 Hapuku 1043 Uaiaholo 1044 Kikoopaoa 1045 Kaluau
(+1 grant) (+3 grants)
1046 Kekipi 1047 Luaka 1048 Hololoa 1049 Kahoohalahala
1050 Kalaihae 1051 Kuaana 1052 Kaiu 1053 Kalaeone
1054 Kalua 1055 Kanui 1056 Kaai 1057 Kahui
1058 Kaiwa 1059 Kawelo 1060 Kakau 1061 Pueo
1062 Kaelehiwa 1063 Iai 1064 Kaleo 1065 Geogi
(+1 grant)
1066 Wahie 1067 Makalawelawe 1068 Naiakapu 1069 Kaiheelua
1070 Huewaa 1071 Kepio 1072 Naope 1073 Puaanui
1074 Mahoe 1075 Keaka 1076 Maheai 1077 Kahana
1078 Puuone 1079 Piikoi 1080 Mai 1081 Moo
1082 Wailele 1083 Koiole 1084 Kai 1085 Kanakaole
1086 Kalili 1087 Mu 1088 Aki 1089 Kamohai
1090 Puailelewale 1091 Kupa 1092 Kanakaokai 1093 Laa
(+1 grant)
1094 Kala 1095 Kaholopo 1096 Kahaipu 1097 Mai
1098 Kenaumoku 1099 Naone 1100 Kalahili 1101 Amalu
1102 Kaluaoko 1103 Kelohe 1104 Lai 1105 Manamana

316
appendixes

1106 Debora 1107 Moo 1 1108 Moo 2 1109 Papalua


1110 Paele 1111 Kamakolu 1112 Makali 1113 Kaanaana
1114 Kanakaole 1115 Waa 1116 Kaona 1117 Hoe
1118 Kai 1119 Kamili 1120 Keahiaena 1121 Kamai
1122 Kuhilani 1123 Noa 1124 Nahuina 1125 Nahuka
1126 Nalu 1127 Pakaka 1128 Paumano 1129 Paouahi
1130 Naonea 1131 Kaialiili 1132 Lili 1133 Ikeole
1134 Nauele 1135 Pule 1136 Kaili 1137 Kaheaku
1138 Uhai 1139 Kaokako 1140 Kaaihue 1141 Kaholopo
1142 Aola 1143 Kamauoha 1144 Ukehaui 1145 Kauakahi
(+2 grants)
1146 Kikau 1147 Nopahi 1148 Kekaowai 1149 Kekaowai
1150 Nika 1151 Moo 1152 Haena 1153 Loita
1154 Hoolulu 1155 Debora 1156 Makalohi 1157 Makolu
1158 Kahula 1159 Kaiawaawa 1160 Kaoloa 1161 Kahaawepala
1162 Kauhihai 1163 Kaheananui 1164 Kapela 1165 Kaiwi
1166 Kaikaalele 1167 Kaoihia 1168 Kekui 1169 Kupanihi
1170 Lilia 1171 Lihue 1172 Moewale 1173 Mareka
1174 Nakoana 1175 Naehuelua 1176 Niau 1177 Wahapuaa
1178 Kealoha 1179 Kualana 1180 Kawaianuhea 1181 Kekoo
1182 Naaiholei 1183 Kekini 1184 Kalakia 1185 Geogi
1186 Laa 1187 Keikiama 1188 Aki 1189 Kapahi
1190 Kaha 1191 Kapuahiauli 1192 Poopuu 1193 Keaka
1194 Kaapuiki 1195 Palauaiki 1196 Naone 1197 Kahaule
1198 Kamuku 1199 Kamamaka 1200 Kaiwipalana 1201 Upai
1202 Lukua 1203 Mahiai 1204 Keliikanakaole 1205 Keliikanakaole
1206 Kehuliaulani 1207 Kanaau 1208 Kahee 1209 Kaahanui
1210 Ilihune 1211 Uhilau 1212 Kaaehule 1213 Ehu
1214 Kaakau 1215 Kawelowahine 1216 Helenihi 1217 Kupuna
1218 Kuapuu 1219 Paaluhi 1220 Nahaomama 1221 Nooui
1222 Lea 1223 Hilihewa 1224 Kelono 1225 Kahiko
1226 Pehu 1227 Kanaauohelii 1228 Kahela 1229 Pohaki
1230 Poihe 1231 Waihinalo 1232 Kahalolani 1233 Kealo
1234 Kaha 1235 Nawahine 1236 Kini 1237 Kaaukai
1238 Nuhi 1239 Paakiki 1240 Kaheewahine 1241 Waiole
1242 Kalama 1243 Kau 1244 Kolia 1245 Niho
1246 Nakua 1247 Kaiwini 1248 Kahananui 1249 Pelelao
1250 Kaunahi 1251 Kalawaia 1252 Kaheana 1253 Pua

317
appendixes

1254 Noa 1255 Kaahu 1256 Nalua 1257 Hoaeae


1258 Poonui 1259 Maiaawa 1260 Awailua 1261 Poohee
1262 Kalikelike 1263 Kalele 1264 Kakau 1265 Paaluhi
1266 Kale 1267 Kaaiai 1268 Nalauha 1269 Nainaelua
1270 Makaole 1271 Uwe 1272 Kaneiahula 1273 Kaaipuhi
1274 Kaiwiahuula 1275 Kokoniho 1276 Kahuailio 1277 Limahanaiau
1278 Kalawehau 1279 Makana 1280 Molokai 1281 Kahula
1282 Kanui 1283 Pinui 1284 Muoone 1285 Aki
1286 Nahoomana 1287 Poonui 1288 Moi 1289 Kaupo
1290 Kaakau 1291 Kawelowahine 1292 Kapuna 1293 Paloe
1294 Kuapuu 1295 Paialuhi 1296 Upai 1297 Pupule
1298 Kahele 1299 Kawauki 1300 Naouealaa 1301 Kahee
1302 Uhilau 1303 Naholowaa 1304 Kaahanui 1305 Mae
1306 Ilihune 1307 Makapo 1308 Kaaepule 1309 Kehuluaulani
1310 Kaahilila 1311 Kuka 1312 Ehu 1313 Kanakaole
1314 Kahaikukuna 1315 Waimea 1316 Mahu 1317 Naou
1318 Kalia 1319 Kaeha 1320 Aukai 1321 Maakua
1322 Kaimakua 1323 Naainaelua 1324 Nakua 1325 Nalehua
1326 Naliikoa 1327 Nuhi 1328 Puna 1329 Kuana
1330 Pakaikua 1331 Puanana 1332 Ohikulani 1333 Kaho‘owaha
1334 Kapahi 1335 Napae 1336 Kauwe 1337 Poopua
1338 Namai 1339 Kaahu 1340 Kahinu 1341 Kaiwilana
1342 Kewalo

*Two names are missing. The 268 names that match names of persons who received
Land Commission Awards through the Mähele and Kuleana Acts are shown in
boldface.
Source: Hawai‘i State Archives, ser. 222, box 2, folder 3.

318
 notes 

notes to chapter one


1. The primary sources for this monograph are oral histories. As a historian I also
draw upon available written documents, records, newspapers, and monographs for each
cultural kïpuka. However, since most written accounts of Hawai‘i focus on the major
islands and centers of trade, written sources for these rural communities are limited; oral
history sources complement and enhance the history or mo‘olelo of these districts and
the kua‘äina who lived there from generation to generation.
2. In Mary Kawena Pukui and Samuel H. Elbert, Hawaiian Dictionary (Honolulu:
University of Hawai‘i Press, 1975), kua‘äina is translated as “person from the country,
rustic; of the country, countrified, rustic. Lit. back land.”
3. Eric Enos, an educator who reopened taro patches in the uplands of Wai‘anae on
the slopes of Mount Ka‘ala, speaks of kua‘äina as backbone in the video He Makana No
Nä Kumu: A Gift to the Teachers of Hawaiian Students, produced by Diane Kahanu and Nä
Maka o Ka‘Äina, 1991.
4. Noa Emmett Aluli, family physician to ‘ohana on the island of Moloka‘i and one of
the founders of the Protect Kaho‘olawe ‘Ohana, speaks of kua‘äina as those who bend
their backs to work the land.
5. Daniel Pahupu, interview by Mary Kawena Pukui, March 9, 1961, Mana‘e, Molo-
ka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi Bishop
Museum).
6. Ibid. The Audio-Recording Collection, Anthropology Department, Bernice Pauahi
Bishop Museum, was a major source of information for this mo‘olelo. The collection was
published in Mary Kawena Pukui, Samuel H. Elbert, and Esther T. Mookini, Place Names
of Hawaii (Honolulu: University of Hawai‘i Press, 1984). The Moloka‘i information was
also included in Catherine Summers, Molokai: A Site Survey, Pacific Anthropological
Records 14 (Honolulu: Bernice Pauahi Bishop Museum, 1971).
7. This version of the chant and the translation is in Pualani Kanakaole Kanahele,
“Ka Honua Ola/ The Living Earth” (unpublished ms., funded by the Center for Hawai-
ian Studies at the University of Hawai‘i at Mänoa from the State Foundation on Culture
and the Arts), pp. 101–3.
8. Andrew Lind, An Island Community: Ecological Succession in Hawaii (New York:
Greenwood Press, 1968), pp. 102–3.

319
notes to pages 10 – 23

9. The general patterns noted here are based upon data and sources referred to and
discussed in the chapters on Waipi‘o, Häna, and Moloka‘i.
10. E. S. Craighill Handy and Elizabeth Green Handy with Mary Kawena Pukui,
Native Planters in Old Hawaii: Their Life, Lore, and Environment (Honolulu: Bishop
Museum Press, 1972), p. vi.
11. Rubellite Johnson, Kumulipo: Hawaiian Hymn of Creation, vol. 1 (Honolulu: Top-
gallant, 1981); The Kumulipo: An Hawaiian Creation Myth, trans. Queen Lili‘uokalani
(Kentfield: Pueo Press, 1997).
12. Handy, Handy, and Pukui, Native Planters in Old Hawaii, p. 64.
13. Kaho‘olawe, Nä Leo O Kanaloa: Chants and Stories of Kaho‘olawe (Honolulu: ‘Ai
Pöhaku Press, 1995).
14. Sitiveni Halapua, “Sustainable Development: From Ideal to Reality in the Pacific
Islands,” paper prepared for the Fourth Pacific Islands Conference of Leaders, Tahiti,
French Polynesia, June 24–26, 1993, sponsored by the East-West Center, Honolulu.
15. Davianna Pömaika‘i McGregor, “Kupa‘a I Ka ‘Äina: Persistence on the Land”
(Ph.D. diss., University of Hawai‘i, 1989).
16. I have also conducted an ethnographic study of the ahupua‘a of Waiähole, Wai-
käne, Hakipu‘u, and Kahana on O‘ahu as the foundation for testimony I prepared for the
Waiähole Water Case before the Hawai‘i State Water Commission. Waiähole and Wai-
käne were the focus of a significant struggle by resident Hawaiian and local farmers and
fishermen in the 1970s to retain the lands in agriculture and sustain the rural Hawaiian
and local lifestyle characteristic of cultural kïpuka. In my study I concluded that these val-
leys are still cultural kïpuka, but their location on the island of O‘ahu and proximity to
the city of Honolulu place them at risk of being transformed into suburbs of Honolulu.
The conscious efforts of the new generation of Hawaiian kua‘äina in these valleys to farm
taro and to sustain the near-shore fishponds and fishing grounds and to reestablish the
natural flow of the stream waters of these valleys away from diversion networks and tun-
nels and toward leeward O‘ahu will sustain these rural valleys as cultural kïpuka.
17. These periods are discussed and summarized in Patrick V. Kirch, Feathered Gods
and Fishhooks: An Introduction to Hawaiian Archaeology and Prehistory (Honolulu: University
of Hawai‘i Press, 1985); Patrick V. Kirch, Legacy of the Landscape: An Illustrated Guide to
Hawaiian Archaeological Sites (Honolulu: University of Hawai‘i Press, 1996); and Malcolm
Naea Chun and Matthew Spriggs, “New Terms Suggested For Early Hawaiian History,”
Ka Wai Ola O OHA (February 1987): 4. Other sources for dating these periods are Abra-
ham Fornander, Fornander Collection of Hawaiian Antiquities and Folk-lore: The Hawaiians’
Account of the Formation of Their Islands and Origins of Their Race, with the Traditions of Their
Migrations as Gathered from Original Sources, ed. Thomas G. Thrum, Bernice Pauahi

320
notes to pages 23– 27

Bishop Museum Memoirs, vols. 4–6 ( Honolulu: Bishop Museum, 1916–20); Martha W.
Beckwith, Hawaiian Mythology (Honolulu: University of Hawai‘i Press, 1970); Samuel
Kamakau, Ruling Chiefs of Hawaii (Honolulu: Kamehameha Schools Press, 1961); Samuel
Kamakau, Ka Po‘e Kahiko: The People of Old, Bernice Pauahi Bishop Museum Special Pub-
lication 51 (Honolulu: Bernice Pauahi Bishop Museum Press, 1964); Samuel Kamakau,
The Works of the People of Old, Bernice Pauahi Bishop Museum Special Publication 61
(Honolulu: Bernice Pauahi Bishop Museum Press, 1976); and David Kaläkaua, King of
Hawaii, The Legends and Myths of Hawaii: The Fables and Folklore of a Strange People (Tokyo
& Rutland, Vt.: Charles E. Tuttle, 1973).
18. Ross Cordy, Exalted Sits the Chief: The Ancient History of Hawai‘i Island (Honolulu:
Mutual, 2000), pp. 104–9.
19. Samuel Kamakau, Tales and Traditions of the People of Old: Nä Mo‘olelo a ka Po‘e
Kahiko, ed. Dorothy B. Barrere, trans. Mary Kawena Pukui (Honolulu: Bishop Museum
Press, 1991), p. 112.
20. Kamakau, Tales and Traditions, p. 136. See also Thomas G. Thrum, Hawaiian Folk
Tales (Chicago: McClurg, 1921); Thrum, More Hawaiian Folk Tales (Chicago: McClurg,
1923); W. D. Westervelt, Legends of Ma-ui (New York: AMS Press, 1979).
21. E. S. Craighill Handy and Mary Kawena Pukui, The Polynesian Family System in
Ka-‘u, Hawai‘i (Tokyo and Rutland, Vt.: Charles E. Tuttle, 1976), pp. 5–6.
22. At the time of Cook, 1778–79, Kalaniopu‘u controlled Hawai‘i island, while Kahe-
kili controlled Maui, O‘ahu, Moloka‘i, Läna‘i, Kaho‘olawe, Kaua‘i, and Ni‘ihau.
23. T. K. Earle, Economic and Social Organization of a Complex Chiefdom: The Halele‘a
District, Kaua‘i, Hawai‘i, Anthropological Papers of the Museum of Anthropology, Uni-
versity of Michigan, No. 63, 1978 (unpublished ms.); Caroline Ralston, “Hawaii, 1778–
1854: Some Aspects of Maka‘äinana Response to Rapid Cultural Change,” Journal of
Pacific History 29, no. 1 ( January 1984): 23.
24. On O‘ahu, the ahupua‘a were bounded on each side by mountain ridges. On Maui,
the ahupua‘a were bounded on each side by streams. On Hawai‘i, cinder hills or pu‘u were
used as boundary markers.
25. Handy, Handy, and Pukui, Native Planters in Old Hawaii. For a detailed description
of traditional Hawaiian land divisions see David Malo, Mo‘olelo Hawai‘i, trans. Nathaniel
B. Emerson as Hawaiian Antiquities, Bernice Pauahi Bishop Museum Special Publication
2 (Honolulu: Bishop Museum Press, 1951).
26. Mary Kawena Pukui, ‘Ölelo No‘eau: Hawaiian Proverbs and Poetical Sayings, Bernice
Pauahi Bishop Museum Special Publication 71 (Honolulu: Bishop Museum Press, 1983),
p. 198.
27. Ibid., no. 1149, p. 125.

321
notes to pages 29– 36

28. Kamakau, Ruling Chiefs, pp. 1–21; Marion Kelly, Majestic Ka‘u: Mo‘olelo of Nine
Ahupua‘a, Department of Anthropology Report Series 80–2 (Honolulu: Bishop Museum
Press, 1990); Pukui and Green, “Nä Ali‘i,” pp. 74–77, 131–33.
29. Pukui, ‘Ölelo No‘eau, no. 1150, p. 125.
30. David Malo, “Causes for the Decrease of the Population in the Islands,” trans. with
comments by Lorrin Andrews, Hawaiian Spectator 2, no. 2 (1839): 125.
31. Kirch, Legacy of the Landscape, p. 6.
32. The estimate from Cook’s voyage was 400,000. A recent study places the precon-
tact population as high as 800,000; see David Stannard, Before the Horror (Honolulu: Uni-
versity of Hawai‘i Press, 1989).
33. Handy and Pukui, The Polynesian Family System, pp. 234–35.
34. Malo, “Causes for the Decrease,” p. 125.
35. Ralph Kuykendall, The Hawaiian Kingdom (Honolulu: University of Hawai‘i Press,
1968), 1:29–60.
36. Kamakau, Ruling Chiefs, pp. 219–28; Kaläkaua, Legends and Myths, pp. 429–46;
Marshall Sahlins, Historical Metaphors and Mythical Realities: Structure in the Early History of
the Sandwich Islands Kingdom (Ann Arbor: University of Michigan Press, 1981), pp. 55–64;
William Davenport, “The Hawaiian ‘Cultural Revolution’: Some Economic and Politi-
cal Considerations,” American Anthropologist 71 (1969): 1–20; A. L. Kroeber, Anthropology
(New York: Harcourt, Brace, 1948).
37. Kanakaole Kanahele, Ola Honua.
38. Kamakau, Ruling Chiefs; Kuykendall, Hawaiian Kingdom.
39. The excerpt is from “Nä Kumukanawai O Ka Makahiki 1839 A Me Ka 1840,”
reproduced in Ka Ho‘oilina, “The Legacy: Puke Pai ‘ölelo Hawai‘i,” Journal of Hawaiian
Language Sources 1, no. 1 (March 2002): 32–33. Note that the journal translates pae ‘äina
o Hawai‘i nei as Sandwich Islands, but I’ve translated it more precisely as Hawaiian archi-
pelago.
40. Ibid., pp. 40–41. I changed the translation to more accurately reflect the Hawaiian.
I wrote “from Hawai‘i to Ni‘ihau” instead of “from one end of the Islands to the other”;
the Hawaiian stated, “It belonged to the people and the chiefs in common,” although the
journal translation changed the word order so that it reads “It belonged to the chiefs and
people in common.”
41. Office of the Commissioner of Public Lands, Indices of Awards Made by the Board
of Commissioners to Quiet Land Titles in the Hawaiian Islands (Honolulu: Star-Bulletin Press,
1929), p. 2.
42. Ibid., pp. 3, 14.
43. Ibid., pp. 3, 15.

322
notes to pages 37– 43

44. Louis Cannelora, The Origin of Hawaii Land Titles and of the Rights of Native Ten-
ants (Honolulu: Security Title Corporation, 1974).
45. Ibid.; Jon J. Chinen, The Great Mähele: Hawaii’s Land Division of 1848 (Honolulu:
University of Hawai‘i Press, 1958), p. 8.
46. Office of the Commissioner of Public Lands, Indices of Awards.
47. Ibid.; Chinen, The Great Mähele, p. 8.
48. Study by Marion Kelly cited in Neil Levy, “Native Hawaiian Land Rights,” Cali-
fornia Law Review 63, no.4 ( July 1975): 856.
49. Prince Jonah Kühiö Kalaniana‘ole, “The Story of the Hawaiians,” Mid-Pacific
Magazine 21, no. 2 (February 1921).
50. Hawai‘i Revised Statutes, sec. 7-1 (1985).
51. Kuykendall, The Hawaiian Kingdom; Davianna McGregor, “Voices of Today Echo
Voices of the Past,” in Mälama: Hawaiian Land and Water, ed. Dana Naone Hall (Hono-
lulu: Bamboo Ridge Press, 1985).
52. William Goodale, “The Hawaiian as Unskilled Laborer,” Hawaiian Almanac and
Annual (1914): 183.
53. Ka Nühou, May 23, 1873.
54. U.S. Department of State, Papers Relating to the Mission of James H. Blount, United
States Commissioner to the Hawaiian Islands (Washington: Government Printing Office,
1893; also referred to as “Blount Report”), pt. 2, p. 5.
55. U.S. Congress, House Report No. 243, “Intervention of United States Govern-
ment in Affairs of Foreign Friendly Governments,” 53rd Congress, 2d sess., December
21, 1893 (Washington: Government Printing Office, 1893); U.S. Congress, Senate Com-
mittee on Foreign Relations, “Hawaiian Islands,” in Report of the Committee on Foreign
Relations with Accompanying Testimony and Executive Documents Transmitted to Congress from
January 1, 1893 to March 19, 1894, vols. 1–2 (Washington: Government Printing Office,
1894 [also referred to as “The Morgan Report”]); U.S. Congress, Senate Committee on
Foreign Relations, Report No. 227, “Report from the Committee on Foreign Relations and
Appendix in Relation to the Hawaiian Islands, February 26, 1894,” 53d Congress, 2d sess.
(Washington: Government Printing Office, 1894); U.S. Department of State, “Blount
Report.”
56. Robert M. C. Littler, The Governance of Hawaii: A Study in Territorial Administra-
tion (Stanford, Calif.: Stanford University Press, 1929).
57. U.S. Congress, "Congressional Debates on Hawaii Organic Act, Together with
Debates and Congressional Action on Other Matters Concerning the Hawaiian Islands,”
56th Congress, 1st sess., 1899–1900, Congressional Record 33, pts. 1–8.
58. McGregor, “Küpa‘a I Ka‘äina.” The Big Five are C. Brewer, Theo H. Davies, Cas-

323
notes to pages 44– 52

tle & Cooke, AmFac, and Alexander and Baldwin. Lawrence Fuchs, Hawaii Pono: A Social
History (San Diego: Harcourt, Brace & World, 1961); Noel Kent, Hawaii: Islands under
the Influence (New York: Monthly Review Press, 1983); Andrew Lind, An Island Commu-
nity: Ecological Succession in Hawaii (Chicago: University of Chicago Press, 1938).
59. McGregor, “Küpa‘a I Ka‘äina.”
60. Eckbo, Dean, Austin, and Williams, with Morris Fox, consultant, “H-3 Socio-Eco-
nomic Study: The Effects of Change on a Windward Oahu Rural Community” (unpub-
lished report, Honolulu, 1973).

notes to chapter two


1. Colin Lennox, “A Report to the Trustees of the Bernice P. Bishop Museum on the
Resources of Waipi‘o Valley, Island of Hawaii: Their Past and Present Uses and an
Analysis of the Problems Facing their Fuller Use in the Future,” unpublished manuscript
(1954), Bernice Pauahi Bishop Museum Library, p. 3; Alfred E. Hudson, “Archaeology
of East Hawai‘i, 1930–32,” unpublished manuscript, Bernice Pauahi Bishop Museum
Library, pp. 132–34.
2. Samuel H. Elbert and Noelani Mahoe, Na Mele o Hawai‘i Nei: 101 Hawaiian Songs
(Honolulu: University of Hawai‘i Press, 1978), pp. 49, 94.
3. Martha Beckwith, Hawaiian Mythology (Honolulu: University of Hawai‘i Press,
1970), p. 155.
4. Ibid.
5. Ibid., pp. 86, 122.
6. Ibid., pp. 350–51.
7. Ibid., p. 142; Westervelt, in Paradise of the Pacific, June 1906, p. 10.
8. Roger Rose, Reconciling the Past: Two Basketry, Ka‘ai, and the Legendary Liloa and Lono-
ikamakahiki (Honolulu: Bernice Pauahi Bishop Museum Press, 1992).
9. Beckwith, Hawaiian Mythology, pp. 353, 357–58, 509–10.
10. Fornander attempted to place the legendary chiefs of the migratory period and
thereafter in specific years. He dated Pilikaeaea at 1090. Fornander, Fornander Collection,
vol. 6, pt. 2, pp. 312–16.
11. Abraham Fornander, An Account of the Polynesian Race, Its Origins and Migrations,
and the Ancient History of the Polynesian People to the Times of Kamehameha I, vol. 2 (Lon-
don: Trubner, 1880), p. 73.
12. Malo, Mo‘olelo Hawaii, pp. 251–54.
13. Ross Cordy, in Exalted Sits the Chief: The Ancient History of Hawai‘i Island (Hono-
lulu: Mutual, 2000), using 20 years per generation, rather than the 30 years per generation
used by Fornander, placed Pili’s reign at 1320 ce. He dated Kaha‘imoele‘a to 1460–80,
Kalaunuiohua to 1480–1500, Kihanuilulumoku to 1560–80, and ‘Umialïloa to 1600–1620.

324
notes to pages 53– 59

14. Handy, Handy, and Pukui, Native Planters in Old Hawaii, pp. 534–35; Beckwith,
Hawaiian Mythology, pp. 389–92; Kamakau, Ruling Chiefs, pp. 1–21.
15. Kamakau, Ruling Chiefs, pp. 19–20.
16. Cordy, Exalted Sits the Chief, p. 200.
17. John F. G. Stokes, Heiau of the Island of Hawai‘i: A Historic Survey of Native Hawai-
ian Temple Sites (Honolulu: Bernice Pauahi Bishop Museum Press, 1991), pp. 159–62.
18. Ibid., p. 151.
19. Stephen L. Desha, Kamehameha and His Warrior Kekuhaupi‘o, trans. Frances N.
Frazier (Honolulu: Kamehameha Schools Press, 2000), pp. 275–89.
20. Stokes, Heiau, p. 160.
21. Desha, Kamehameha, pp. 299–301.
22. William Ellis, Journal of William Ellis, Narrative of a Tour of Hawaii, or Owhyhee;
With Remarks on the History, Traditions, Manners, Customs and Language of the Inhabitants
of the Sandwich Islands (Honolulu: Advertiser, 1963), p. 256.
23. Ibid., pp. 365–66.
24. Kamakau, Ruling Chiefs, p. 285.
25. Rose, Reconciling the Past, p. 11.
26. Hudson, “Archaeology of East Hawaii,” pp. 141–42, 151–53.
27. Cited in Handy, Handy, and Pukui, Native Planters in Old Hawaii, p. 534.
28. Dorothy Barrere, comp., “The King’s Mähele: The Awardees and Their Lands,”
unpublished bound manuscript, 1994 (Hamilton Library, University of Hawai‘i at
Mänoa). Information on the ali‘i of the mähele is drawn from this manuscript.
29. Office of the Commissioner of Public Lands [OCPL], Indices of Awards Made by
the Board of Commissioners to Quiet Land Titles in the Hawaiian Islands (Honolulu: Star-Bul-
letin Press, 1929), pp. 3–7, 17–18, 432–36.
30. Cited in Hudson, “Archaeology of East Hawaii,” p. 142, and in Ethnic Studies Oral
History Project [ESOHP], Waipi‘o Mano Wai: An Oral History Collection, 2 vols. (Hono-
lulu: Ethnic Studies Program, University of Hawai‘i at Mänoa, 1978), p. C-23.
31. ESOHP, Waipi‘o Mano Wai, p. C-23.
32. Ibid., p. C-24.
33. Ethel Damon, Father Bond of Kohala: A Chronicle of Pioneer Life in Hawaii (Hono-
lulu: The Friend, 1927), p. 209.
34. Lennox, “Report to the Trustees,” pp. a, 4.
35. Cited in Handy, Handy, and Pukui, Native Planters in Old Hawaii, p. 534; printed
in Paradise of the Pacific, May 1895, p. 67.
36. Lennox, “Report to the Trustees,” p. 24.
37. Glenn Petersen, “Taro Farming in Waipio Valley on the Island of Hawaii,” in
North Kohala Studies: Preliminary Research in Human Ecology, ed. R. Warwick Armstrong

325
notes to pages 60 – 66

and Henry T. Lewis (Honolulu: University of Hawai‘i Press, 1970), p. 26; Stella Jones,
“Field Notes on Waipi‘o, 1931,” from an informant called Kahimoku. These are unpub-
lished field notes for work by Jones sponsored by the Bishop Museum and are located in
the Bernice Pauahi Bishop Museum Library.
38. Lennox, “Report to the Trustees,” p. 24.
39. Pacific Commercial Advertiser, June 27, 1906, p. 5.
40. Kalani and Sam Kaaekuahiwi, in Jones, “Field Notes on Waipi‘o, 1931.”
41. Sam Kaaekuahiwi, in Jones, “Field Notes on Waipi‘o, 1931.”
42. David Makaoi, interview by Vivien Lee and Yukie Yoshinaga, February 9, 1978,
Nu‘uanu, O‘ahu, ESOHP, Waipi‘o, Mano Wai, p. 844.
43. ESOHP, Waipi‘o, Mano Wai, p. i, Appendix I.
44. Kalani in Jones, “Field Notes on Waipio, 1931”; Lennox, “Report to the Trust-
ees,” pp. 24–25.
45. Mrs. Kapahu in Jones, “Field Notes on Waipi‘o, 1931.”
46. Lennox, “Report to the Trustees,” p. 25.
47. Petersen, “Taro Farming in Waipio Valley,” p. 26.
48. U.S. Bureau of the Census, Fifteenth Census of the United States: 1930, Occupation
Statistics Hawaii (Washington, D.C.: Government Printing Office, 1932), table 22, p. 70.
The census provides an enumeration by election precinct. The election precincts are
identified in Hawai‘i State Archives, “Governors’ Proclamations, 1926–1930,” pp. 6–21,
128–47.
49. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 850.
50. Ibid., pp. 847–48, 850–52; Rachel Thomas, April 8, 1978, ESOHP, Waipi‘o, Mano
Wai, pp. 1015–17.
51. Ted Kaaekuahiwi, interview by Vivien Lee and Yukie Yoshinaga, March 13, 1978,
ESOHP, Waipi‘o, Mano Wai, pp. 395–96; William Kanekoa, April 7, 1978, ESOHP, Wai-
pi‘o, Mano Wai, pp. 700–701; David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano
Wai, pp. 853–54.
52. Ted Kaaekuahiwi, March 13, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 392–94,
397–98; David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 858.
53. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 858.
54. Jones, “Field Notes”; Rachel Thomas, interview by Vivien Lee and Yukie Yoshi-
naga, April 8, 1978, Kukuihaele, Hawai‘i, ESOHP, Waipi‘o, Mano Wai, pp. 1010–11.
55. Sam Kaaekuahiwi in Jones, “Field Notes,” and Jones’ own observations.
56. Sam Kaaekuahiwi in Jones, “Field Notes.”
57. Sam Kaaekuahiwi in Jones, “Field Notes.”
58. Nelson Chun, interview by Vivien Lee and Yukie Yoshinaga, March 11, 1978,
Honoka‘a, Hawai‘i, ESOHP, Waipi‘o, Mano Wai, p. 174.

326
notes to pages 67– 75

59. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 861; Ted Kaae-
kuahiwi, March 13, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 400–402.
60. Jones, “Field Notes.”
61. Comment on fertilizer: Nelson Chun, March 11, 1978, ESOHP, Waipi‘o, Mano
Wai, p. 174.
62. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 844–45.
63. Sam Kaaekuahiwi in Jones, “Field Notes.”
64. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 854.
65. Ibid., pp. 842, 857–58.
66. George Farm, interview by Vivien Lee and Yukie Yoshinaga, April 4, 1978, Kukui-
haele, Hawai‘i, ESOHP, Waipi‘o, Mano Wai, 301–2.
67. Jones, “Field Notes”; Fannie Hauanio Duldulao, interview by Vivien Lee and
Yukie Yoshinaga, March 14, 1978, Kukuihaele, Hawai‘i, ESOHP, Waipi‘o, Mano Wai,
p. 275.
68. Ted Kaaekuahiwi, March 13, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 403–5; Rachel
Thomas, April 8, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 1020–21.
69. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 854.
70. Leslie Chang, interview by Vivien Lee and Yukie Yoshinaga, June 4, 1978, Hilo,
Hawai‘i, ESOHP, Waipi‘o, Mano Wai, pp. 86–87; William Kanekoa, interview by Vivien
Lee and Yukie Yoshinaga, April 7, 1978, Kukuihaele, Hawai‘i, ESOHP, Waipi‘o, Mano
Wai, p. 702; David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 869–70.
71. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 961.
72. Leslie Chang, June 4, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 90–91.
73. Hawai‘i State Archives, Land Matters, Public Lands Petitions File, Naleilehua to
Governor Carter, January 1, 1904.
74. In 1947 the people of the valley asked the county to widen the road, but the county
did not want to spend the estimated $10,000–$15,000 that it would cost. The road was
wide enough for mules and for narrow-based four-wheel-drive vehicles. Lennox, “Report
to the Trustees,” pp. 64–66.
75. Fannie Hauanio Duldulao, March 14, 1978, ESOHP, Waipi‘o, Mano Wai, p. 297.
76. Leslie Chang, June 4, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 91–92; David Makaoi,
February 9, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 868–69.
77. Jones, “Field Notes.”
78. Daughter-in-law of Mrs. Kanekoa in Jones, “Field Notes.”
79. Fannie Hauanio Duldulao, March 14, 1978, ESOHP, Waipi‘o, Mano Wai, p. 276.
80. Ibid.
81. Daughter-in-law of Mrs. Kanekoa in Jones, “Field Notes.”

327
notes to pages 76 – 86

82. Jones, “Field Notes.”


83. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 863.
84. Fannie Hauanio Duldulao, March 14, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 271,
274.
85. Ibid., pp. 272–73.
86. David Makaoi, February 9, 1978, ESOHP, Waipi‘o, Mano Wai, p. 842.
87. Ibid., p. 859.
88. Ted Kaaekuahiwi, March 13, 1978, ESOHP, Waipi‘o, Mano Wai, pp. 412–15.
89. ESOHP, Waipi‘o, Mano Wai, p. C-28.
90. Paradise of the Pacific, June 1948, p. 11.
91. Lennox, “Report to the Trustees,” pp. a, 4.
92. University of Hawai‘i Land Study Bureau. “Preliminary Survey of Resource
Development and Rehabilitation Opportunities in Waipi‘o Valley, Island of Hawai‘i,”
Special Study Series, L.S.B. Report 1, January 1960.
93. R. Warwick Armstrong and Henry T. Lewis, eds. North Kohala Studies: Preliminary
Research in Human Ecology (Honolulu: University of Hawai‘i Press, 1970).
94. Honolulu Star-Bulletin, March 12, 1964.
95. Honolulu Star-Bulletin, November 17, 1972.
96. ESOHP, Waipi‘o, Mano Wai, pp. C-31–C-32.
97. Department of Urban and Regional Planning, University of Hawai‘i at Mänoa,
“Waipi‘o Valley: Towards Community Planning and Ahupua‘a Management,” technical
report (Fall 1999), p. 17.
98. Honolulu Star-Bulletin, November 10, 1963.
99. Ibid.

notes to chapter three


1. Pukui, ‘Ölelo No‘eau, no. 460, p. 55.
2. Curtis J. Lyons, “Land Matters in Hawai‘i,” The Islander, no. 2 (Honolulu, 1875),
p. 111.
3. Jocelyn Linnekin, Children of the Land: Exchange and Status in a Hawaiian Community
(New Brunswick, N.J.: Rutgers University Press, 1985); Leonard Lueras, On the Hana
Coast: Being an Accounting of Adventures, Past and Present, in a Land Where the Hand of Man
Seems to Rest Lightly (Honolulu: Emphasis International, 1983); Bob Jones, “Kaulana na
Pua/ Famous Are the Children” [KGMB news special], 1987.
4. Thomas K. Maunupau, Nüpepa Kü‘oko‘a, June 15, 1922, trans. Mary Kawena Pukui
(Hawaiian Ethnographic Notes), Newspapers 1922–24, Bernice Pauahi Bishop Museum
Library. After I used this version for my dissertation it was published as Thomas K.
Maunupau, Huakai Makaikai a Kaupö, Maui: A Visit to Kaupö, Maui, as published in Ka

328
notes to pages 87– 90

Nüpepa Kü‘oko‘a, June 1, 1922–March 15, 1923, ed. Naomi Noelanioko‘olau Clarke
Losch, trans. Mary Kawena Pukui and Malcolm Naea Chun (Honolulu: Bernice Pauahi
Bishop Museum Press, 1998), pp. 103–4.
5. W. M. Walker, “Archaeology of Maui,” unpublished manuscript (Bernice Pauahi
Bishop Musuem, 1931), pp. 34–35; Beckwith, Hawaiian Mythology, pp. 226–37.
6. Beckwith, Hawaiian Mythology, p. 230.
7. Martha W. Beckwith, ed., Kepelino’s Traditions of Hawai‘i, Bernice Pauahi Bishop
Museum Bulletin 95 (Honolulu: Bernice Pauahi Bishop Museum; New York: Kraus
Reprint, 1971), p. 229.
8. Beckwith, Hawaiian Mythology, p. 64.
9. Handy, Handy, and Pukui, Native Planters in Old Hawaii, p. 510.
10. Beckwith, Hawaiian Mythology, p. 65.
11. Ibid., pp. 99–100.
12. Julia Naone, interview by Mary Kawena Pukui, May 3, 1960, Hämoa, Kïpahulu,
Maui (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi Bishop
Museum), no. 85.6.
13. Julia Naone, interview by Mary Kawena Pukui, May 3, 1960, no. 85.4; Josephine
Kauakeaohana Roback Medeiros, interview by Mary Kawena Pukui, May 5, 1960, Hämoa,
Kïpahulu, Maui (Audio-Recording Collection, Department of Anthropology, Bernice
Pauahi Bishop Museum), no. 87.4.2.
14. Julia Naone, interview by Mary Kawena Pukui, May 3, 1960, no. 85.4.
15. Josephine Marciel, interview by Mary Kawena Pukui, May 3, 1960, Hämoa, Kïpa-
hulu, Maui (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum).
16. Walker, “Archaeology of Maui,” p. 40.
17. Mrs. Paul Fagan, interview by Mary Kawena Pukui, May 5, 1960, Hana, Ke‘anae,
Maui (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi Bishop
Museum), no. 87.2.1.
18. Beckwith, Hawaiian Mythology, pp. 40–41.
19. Walker, “Archaeology of Maui,” pp. 37–38.
20. Beckwith, Hawaiian Mythology, p. 22.
21. Ibid., pp. 22–23.
22. Jon K. Matsuoka et al., “Native Hawaiian Ethnographic Study for the Hawai‘i
Geothermal Project Proposed for Puna and Southeast Maui,” technical report, Oak Ridge
National Laboratory, Lockheed Martin (May 1996), pp. 172–73; Walker, “Archaeology of
Maui,” p. 7; Beckwith, Hawaiian Mythology, p. 170.
23. Maunupau, Huakai Makaikai a Kaupö, Maui /A Visit to Kaupö, Maui, pp. 97, 100,
101.

329
notes to pages 91– 98

24. The information in this section is based on Handy, Handy, and Pukui, Native
Planters in Old Hawaii, pp. 498–502.
25. In conversations with the author, küpuna Harry Kunihi Mitchell identified the
flume as being made out of an old canoe.
26. E. S. Craighill Handy, The Hawaiian Planter, vol. 1, His Plants, Methods and Areas
of Cultivation, Bernice Pauahi Bishop Museum Bulletin 161 (Honolulu: Bernice Pauahi
Bishop Museum, 1940), pp. 111–12.
27. Minnie and Sam Po, Elspeth Sterling, and Peter Chapman, interview by Mary
Kawena Pukui, July 5, 1966, Häna, Maui: tape 4 (Audio-Recording Collection, Depart-
ment of Anthropology, Bernice Pauahi Bishop Museum).
28. Sam Po, interview by Mary Kawena Pukui, July 5, 1966, Häna, Maui: tapes 5–7
(Audio-Recording Collection, Department of Anthropology, Bernice Pauahi Bishop
Museum).
29. Several testimonies in McGregor, “Küpa‘a I Ka‘äina,” p. 372.
30. Beckwith, Hawaiian Mythology, p. 383; Handy, The Hawaiian Planter, 1:491.
31. Beckwith, Hawaiian Mythology, pp. 307.
32. Walker, “Archaeology of Maui,” pp. 13, 14.
33. Fornander, Fornander Collection, vol. 6, pt. 2, p. 313, and vol. 4, pt. 2, pp. 214–18,
236–56.
34. Patrick V. Kirch, Legacy of the Landscape: An Illustrated Guide to Hawaiian Archaeo-
logical Sites (Honolulu: University of Hawai‘i Press, 1996), p. 72.
35. Kamakau, Ruling Chiefs, p. 385.
36. Handy, Handy, and Pukui, Native Planters in Old Hawaii, pp. 502–4.
37. Pukui, ‘Ölelo No‘eau, no. 2548, p. 278.
38. Beckwith, Hawaiian Mythology, p. 379.
39. Kamakau, Ruling Chiefs, p. 80.
40. Ibid., p. 380.
41. Ibid., p. 80.
42. Ibid., pp. 22–33.
43. Ibid. See also Trust for Public Land with Bay Pacific Consulting, “East Maui
Resource Inventory” (Honolulu: Rivers, Trails, and Conservation Assistance Program,
National Park Service, 1998).
44. Handy, Handy, and Pukui, Native Planters in Old Hawaii, pp. 503–4; Beckwith,
Hawaiian Mythology, p. 380.
45. Kamakau, Ruling Chiefs, p. 84; Lueras, On the Hana Coast, pp. 40–41.
46. Walker, “Archaeology of Maui,” pp. 25, 124.
47. Ibid., p. 16.
48. Kamakau, Ruling Chiefs, p. 78.

330
notes to pages 98– 105

49. Ibid., p. 84.


50. Desha, Kamehameha and His Warrior Kekuhaupi‘o, p. 31.
51. Captain Cook had first stopped at Waimea Kaua‘i in January 1778. He then pro-
ceeded to the North Pacific in search of the Northwest Passage. When winter icebergs
formed and obstructed his exploration, he returned to Hawai‘i to wait out the winter.
52. Kamakau, Ruling Chiefs, pp. 97–100; Desha, Kamehameha and His Warrior Keku-
haupi‘o, pp. 58–61.
53. Kamakau, Ruling Chiefs, p. 116.
54. The accounts of the battles are based upon Desha, Kamehameha and His Warrior
Kekuhaupi‘o, pp. 216–22, and Kamakau, Ruling Chiefs, pp. 128–58.
55. Desha, Kamehameha and His Warrior Kekuhaupi‘o, pp. 239–302.
56. Jonathan S. Green, letter, Dec. 1835.
57. John Forster, “Social Organization and Differential Social Change in Two Hawai-
ian Communities,” International Journal of Comparative Sociology 3 (December 1962): 202.
58. Linnekin, Children of the Land, p. 37.
59. Wailuku Station Report, Hawai‘i Mission Children’s Society Library (1837), cited
in Linnekin, Children of the Land, p. 18.
60. Kamakau, Ruling Chiefs, pp. 236–37.
61. Ibid., p. 418.
62. Barrere, “The King’s Mähele.”
63. Lueras, On the Hana Coast, pp. 44–45; Forster, “Social Organization,” pp. 202–3.
64. Lueras, On the Hana Coast, p. 45.
65. Maui Visitors Bureau, “Hana Visitors Guide,” 1998.
66. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988, Ke‘a-
nae, Maui.
67. D. T. Conde to D. Baldwin, February 7, 1844; D. T. Conde to L. Chamberlain and
S. N. Castle, December 26, 1838; D. T. Conde to Titus Coan, c. 1845, in Hawai‘i Mis-
sion Children’s Society Library, cited in Linnekin, Children of the Land, pp. 18–19.
68. Sereno Bishop, “Report to Board of Sandwich Islands Mission” (1861), in Hawai‘i
Mission Children’s Society Library, cited in Linnekin, Children of the Land, p. 18.
69. Josephine Marciel, interview by Mary Kawena Pukui, May 3, 1960, no. 85.7.
70. Lueras, On the Hana Coast, pp. 46–47.
71. Ibid., p. 60; Harry Kunihi Mitchell, interview by Davianna McGregor, April 22,
1988.
72. Pacific Commercial Advertiser, September 4, 1910.
73. Ibid.
74. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988.
75. U.S. Bureau of the Census, Fifteenth Census of the United States: 1930, Occupation

331
notes to pages 105– 111

Statistics Hawaii (Washington: Government Printing Office, 1931), p. 72, table 22. The
precincts were identified in “Governors’ Proclamations,” 1926–30, pp. 6–21.
76. Douglas Yamamura, “A Study of Some of the Factors in the Education of the
Child of Hawaiian Ancestry in Hana, Maui,” master’s thesis (University of Hawai‘i, 1941),
pp. 21, 24, 108, 151; Harry Kunihi Mitchell, interview by Davianna McGregor, April 22,
1988.
77. Yamamura, “A Study of Some of the Factors,” pp. 39–43. Yamamura’s information
is based on a study of 52 families whose children attended Häna School in 1939. This is
after the 1930 cut-off, but the patterns observed in 1939 had been carried over from the
earlier period and were not a recently introduced pattern.
78. Ibid., pp. 40, 42.
79. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988; Lueras,
On the Hana Coast, p. 60.
80. Yamamura, “A Study of Some of the Factors,” pp. 45, 46.
81. Ibid., p. 151.
82. Ibid., pp. 151–52, 155.
83. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988.
84. Yamamura, “A Study of Some of the Factors,” pp. 124–25.
85. Ibid., pp. 126–31.
86. Ibid., pp. 140–41.
87. Mrs. Kapeka Ka‘auamo, interview by Mary Kawena Pukui, December 2, 1961,
Wailua, Ke‘anae, Maui (Audio-Recording Collection, Department of Anthropology, Ber-
nice Pauahi Bishop Museum), no. 122.2, transcribed and translated by Larry Kimura.
88. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988.
89. Josephine Medeiros, interview by Mary Kawena Pukui, no. 87.3.1.
90. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988.
91. Moewale and Joseph Pu, interview by Mary Kawena Pukui, May 4–5, 1960, Hana
and Kïpahulu, Maui (Audio-Recording Collection, Department of Anthropology, Bernice
Pauahi Bishop Museum), no. 87.1; Mrs. Kapeka Ka‘auamo, interview by Mary Kawena
Pukui, December 2, 1961, no. 122.1.1.
92. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988.
93. Mrs. Kapeka Ka‘auamo, interview by Mary Kawena Pukui, December 2, 1961, no.
122.1.2.
94. Sam Po, interview by Mary Kawena Pukui, July 5, 1966, no. 4, Hana and Kïpahulu,
Maui (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi Bishop
Museum).
95. Josephine Marciel, interview by Mary Kawena Pukui, May 3, 1960; Yamamura, “A

332
notes to pages 111– 116

Study of Some of the Factors,” p. 35; Harry Kunihi Mitchell, interview by Davianna
McGregor, April 22, 1988.
96. Josephine Marciel, interview by Mary Kawena Pukui, May 3, 1960; Francis Mar-
ciel and Dolly Mahalo, interview by Mary Kawena Pukui, December 1, 1961, Kaupö,
Maui (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi Bishop
Museum), no. 86.3.
97. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988.
98. Agnes Mailou and Daisy Lind, interview by Mary Kawena Pukui, March 15–16,
1963, Häna (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), nos. 137.3.2, 200.03, 200.04; Una Walker, Karen Pryor, Elizabeth Haia
Chang, Babes Hanchett, Leimamo Lee, and Mrs. Charles Pohaku, interview by Mary
Kawena Pukui, August 10, 1968, Häna (Audio-Recording Collection, Department of
Anthropology, Bernice Pauahi Bishop Museum), nos. 200.3, 200.4.
99. Mrs. Kapeka Ka‘auamo, interview by Mary Kawena Pukui, December 2, 1963, no.
122.2; Agnes Mailou and Daisy Lind, interview by Mary Kawena Pukui, March 16, 1963;
Una Walker, Karen Pryor, Elizabeth Haia Chang, Babes Hanchett, Leimamo Lee, and
Mrs. Charles Pohaku, interviews by Mary Kawena Pukui, August 10, 1968; Josephine
Marciel, interview by Mary Kawena Pukui, May 3, 1960, no. 85.7.
100. Craighill Handy, The Hawaiian Planter, 1:110.
101. Pacific Commercial Advertiser, September 4, 1910.
102. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988.
103. Forster, “Social Organization,” p. 203.
104. Harry Kunihi Mitchell, interview by Davianna McGregor, April 22, 1988.
105. Ibid.
106. Ibid.
107. Linnekin, Children of the Land, pp. 76–77.
108. Handy, The Hawaiian Planter, 1:110.
109. Josephine Medeiros, interview by Mary Kawena Pukui, May 5, 1960, no. 87.4,
1–2.
110. Group 70, Davianna McGregor, and Cultural Surveys Hawai‘i, “Kalo Kanu o Ka
‘Äina: A Cultural Landscape Study of Ke‘anae and Wailuanui, Island of Maui” (technical
report for the County of Maui Planning Department, 1995), pp. 38, 84.
111. Handy, The Hawaiian Planter, 1:111.
112. Lueras, On the Hana Coast, pp. 96–97.
113. Ibid.; Yamamura, “A Study of Some of the Factors,” p. 9.
114. Lueras, On the Hana Coast, pp. 96–97.
115. Ibid.

333
notes to pages 116 – 126

116. Handy, The Hawaiian Planter, 1:111.


117. Lueras, On the Hana Coast, p. 60.
118. Ibid., p. 46.
119. Ibid., p. 60.
120. Handy, Handy, and Pukui, Native Planters in Old Hawaii, pp. 504–5.
121. Handy, The Hawaiian Planter, 1:111, 112.
122. Kuoko‘a [newspaper], June 1, 1922, trans. Mary Kawena Pukui, Hawaiian Ethno-
graphic Notes, Newspapers 1922–24, Bernice Pauahi Bernice Pauahi Bishop Museum
Library.
123. Hawaiian Gazette, September 6, 1910.
124. Lueras, On the Hana Coast, p. 46.
125. Handy, The Hawaiian Planter, 1:112–13.
126. Lueras, On the Hana Coast, p. 133; Sam O. Hirota, Inc., “Environmental Impact
Statement for the Kaupö Water System Improvements, Kaupö, Island of Maui, Hawaii”
(Honolulu, 1983), Hawaiian Collection, Hamilton Library, University of Hawai‘i.
127. Lueras, On the Hana Coast, p. 46.
128. Hawaiian Gazette, September 6, 1910; Pacific Commercial Advertiser, September 5,
1910.
129. Pacific Commercial Advertiser, September 9, 1910.
130. Kuoko‘a, June 1, 1922.
131. Kuoko‘a, June 20, 1922, trans. Mary Kawena Pukui, HEN, Newspapers 1922–24,
Bernice Pauahi Bishop Museum Library.
132. Handy, The Hawaiian Planter, 1:113.
133. Josephine Marciel, interview, May 3, 1960, no. 85.7.
134. Ibid.
135. Lueras, On the Hana Coast, p. 65.
136. Maui News, December 9, 1942, cited in Group 70, McGregor, and Cultural Sur-
veys Hawai‘i, “Kalo Kanu o Ka ‘Äina,” p. 38.
137. Group 70, McGregor, and Cultural Surveys Hawai‘i, “Kalo Kanu o Ka ‘Äina,”
p. 38.
138. Lueras, On the Hana Coast, pp. 67–70.
139. Maui Visitor’s Bureau, “Hana Visitors Guide,” 1998; F. P. Shepard, G. A. Mac-
Donald, and D. C. Cox, The Tsunami of April 1, 1946 (Berkeley and Los Angeles:
University of California Press, 1950); Linnekin, Children of the Land, cited in Group 70,
McGregor, and Cultural Surveys Hawai‘i, “Kalo Kanu o Ka ‘Äina,” pp. 38–39.
140. Maui Visitor’s Bureau, “Hana Visitors Guide,” 1998.
141. Trust for Public Land with Bay Pacific Consulting, “East Maui Resource Inven-
tory,” p. 35.

334
notes to pages 127– 145

142. Group 70, McGregor, and Cultural Surveys Hawai‘i. “Kalo Kanu o Ka ‘Äina,
p. 40.
143. Telephone conversation with Charmain Day, December 3, 1994.
144. For the Ko‘olau district of Maui, all of the ahupua‘a boundaries converge and
originate at the northern rim of Haleakalä at Pöhaku Pälaha; Lyons, “Land Matters in
Hawai‘i,” p. 111. According to James Hueu, “In Ke‘anae, they had their ahupua‘a. Hono-
manü to Makapipi towards Nähiku.” James Hueu, interview by Davianna McGregor,
November 12, 1994, Ke‘anae, Maui.
145. Group 70, McGregor, and Cultural Surveys Hawai‘i, “Kalo Kanu o Ka ‘Äina,”
pp. 102–3.
146. Ibid., p. 104.
147. Ibid., pp. 104–5.
148. Ibid., p. 105.
149. Ibid., pp. 106–7.
150. Ibid., p. 107.
151. Ibid., p. 108.
152. Ibid., p. 109.
153. Ibid.
154. Ibid., p. 110.
155. Ibid., pp. 110–11.
156. Ibid., p. 111.
157. Ibid., pp. 111–12.
158. Ibid., p. 112.
159. Ibid., pp. 112–13.
160. Moki Day, interview by Davianna McGregor, November 3, 1994, Wailuanui,
Maui.
161. Group 70, McGregor, and Cultural Surveys Hawai‘i, “Kalo Kanu o Ka ‘Äina,”
p. 113.

notes to chapter four


1. Pukui, ‘Ölelo No‘eau, no. 2747.
2. Ibid., no. 2058.
3. Ibid., no. 1458.
4. Ibid., no. 1587.
5. Pualani Kanaka‘ole Kanahele, lecture to student body of Kamehameha Schools,
May 2, 1990.
6. Kanaka‘ole Kanahele, “Ka Honua Ola: The Living Earth.” The following transla-
tion and interpretation of the chant is excerpted from Kanahele’s report.

335
notes to pages 146– 151

7. Ibid., p. 71.
8. Ibid., p. 73.
9. Pukui, ‘Ölelo No‘eau, no. 1777.
10. Excerpt from “A Legend told by Moses Manu, Ka Loea Kalai‘aina, May 1899–
Feb. 1900,” trans. Mary Kawena Pukui, in Dorothy Barrere, “Political History of Puna,”
manuscript for Archaeology Reconnaissance of the Kalapana Extension, Bernice Pauahi
Bishop Museum, 1959.
11. Beckwith, Hawaiian Mythology; G. W. Kahiolo, He Moolelo No Kamapua‘a: The Story
of Kamapua‘a, trans. Esther T. Mookini and Erin C. Neizmen with the assistance of David
Tom (Hawaiian Studies Program, University of Hawai‘i at Mänoa, 1978).
12. Greg Burtchard and Pennie Moblo, “Archaeology in the Kilauea East Rift Zone
Kapoho, Kama‘ili and Kilauea Geothermal Subzones, Puna District, Hawai‘i Island,”
Report ORNL/SUB/94-SN150/1-2, July (Oak Ridge, Tenn.: Oak Ridge National Lab-
oratory, 1994).
13. Jules Remy, Contributions of a Venerable Savage to the Ancient History of the Hawai-
ian Islands, trans. William T. Brigham (Boston: Press of A. A. Kingman, 1868).
14. Ellis, Journal; Hawaiian Majesty King David Kalakaua, Legends and Myths of
Hawaii; Mary Kawena Pukui and C. Curtis, Pikoi and Other Legends of the Island of Hawai‘i
(reprint, Honolulu: Kamehameha Schools Press, 1949).
15. Beckwith, Hawaiian Mythology, pp. 190–92.
16. Kamakau, Tales and Traditions, p. 100.
17. “Tradition ascribes to Paao the introduction of human sacrifice into the temple rit-
ual, the walled heiau, and the red-feather girdle as a sign of rank; all typical, says Handy,
of late Tahitian culture and not found in Samoa. Other institutions ascribed to him are the
pulo‘ulo‘u tapu sign, the prostrating tapu (tapu moe or -o), and the feather god Kaili; some
would call Paao rather than La‘a-mai-kahiki the introducer of image worship . . . That
Paao took his ideas from Tahiti is further indicated by reference to ‘Vavau’ and ‘Upolo’
as places where he owned land, probably in districts so named in northern Tahiti in the
Aha-roa division of that island, and the name Aha-ula ( later called Waha-ula) for the first
heiau erected by his party on Hawai‘i suggests such a connection.” Beckwith, Hawaiian
Mythology, p. 370.
18. Fornander, An Account of the Polynesian Race, pp. 35–36.
19. Kamakau, Tales and Traditions, p. 100.
20. Barrere, “Political History of Puna”; Bruce Cartwright, “Some Aliis of the Migra-
tory Period,” Bernice Pauahi Bishop Museum Occasional Papers, vol. 10, no. 7 (Hono-
lulu: Bernice Pauahi Bishop Museum, 1933).
21. Fornander, Fornander Collection, pp. 514–19; Walker, “Archaeology of Maui,”
p. 41.

336
notes to pages 151– 163

22. Beckwith, Hawaiian Mythology, pp. 363–70.


23. Pukui, ‘Ölelo No‘eau, no. 994.
24. Ibid., no. 260.
25. Barrere, “Political History of Puna,” p. 15.
26. Kamakau, Tales and Traditions, p. 10.
27. Ibid., p. 17.
28. Barrere, “Political History of Puna.”
29. Kamakau, Tales and Traditions, pp. 18–19.
30. Barrere, “Political History of Puna,” 17–18.
31. Kamakau, Tales and Traditions, p. 106.
32. Ibid.
33. Ibid., pp. 108–9.
34. J. C. Beaglehole, The Journals of Captain James Cook On His Voyages of Discovery, vol.
3, The Voyage of the Resolution and Discovery, 1776–1780 (Cambridge: Hakluyt Society,
1955–74), pp. 196, 606.
35. Barrere, “Political History of Puna,” p. 19.
36. Pukui, ‘Ölelo No‘eau, no. 826.
37. Ellis, Journal of William Ellis.
38. Ibid.
39. Charles Langlas, The People of Kalapana, 1832–1950 (Hilo: The author, 1990),
pp. 17–18.
40. Tommy Holmes, “A Preliminary Report on the Early History and Archaeology of
the Puna Forest Reserve / Wao Kele ‘o Puna Natural Area Reserve,” report prepared for
True /Mid-Pacific Geothermal ( November 1985).
41. Charles Wilkes, Narrative of the U.S. Exploring Expedition During the Years 1838,
1839, 1840, 1841, 1842 (Philadelphia, 1849), vol. 4, p. 181.
42. Chester S. Lyman, Around the Horn to the Sandwich Islands and California, 1845–
1850 (New Haven, Conn.: Yale University Press, 1924), p. 19.
43. Seven lands in Puna were left unassigned during the Mähele: Kahue, Hulunanai,
Iililoa, Kaunaloa, Ki (B), Keekee, and Keonepoko 2. In 1888 it was decided that these
would be government lands. See Melinda Sue Allen, “The Kalapana Extension in the
1800’s, A Research of the Historical Records,” prepared for the National Park Service,
Hawaii Volcanoes National Park, 1979.
44. Allen, “Kalapana Extension.”
45. Ibid..
46. Hawai‘i State Archives, Interior Department Letters, March 26, 1857.
47. In Holmes, “Preliminary Report.”
48. Ibid., p. 17.

337
notes to pages 164– 181

49. Hawai‘i State Archives, Boundary Commission Hawai‘i, 3rd & 4th Circuits, Peti-
tioner’s Exhibit B, “The Ahupuaa of Keaau, District of Puna, Island of Hawaii, 3d, J.C.,”
before the Commissioner of Boundaries, Fourth Judicial Circuit, Territory of Hawai‘i,
In the Matter of the Boundaries of Waiakahekahe-‘iki upon the petition of W. H. Ship-
man, owner (Hilo, June 18, 1914).
50. Hawai‘i State Archives, Boundary Commission Hawai‘i, 3rd & 4th Circuits, In Re:
Boundaries, Ahupuaa of Keahialaka, Puna, Hawai‘i, brief of Hitchcock & Wise, filed
March 20, 1897.
51. Thomas G. Thrum, Thrum’s Hawaiian Almanac and Annual (Honolulu: Honolulu
Star-Bulletin, 1895).
52. Roger Skolmen, “Hawaii’s Forest Products Industry,” paper presented at the 18th
Annual Hawaii Forestry Conference, November 18–19, 1976, Honolulu.
53. Hawai‘i State Archives, Governor’s Proclamation, June 29, 1911; Governor’s Proc-
lamation, December 22, 1928.
54. Charles Baldwin, Geography of the Hawaiian Islands (New York: American Book,
1908), pp. 78–79.
55. Henry Walsworth Kinney, The Island of Hawai‘i (Hilo: Hilo Board of Trade,
1913).
56. Ibid.
57. Handy, Handy, and Pukui, Native Planters in Old Hawaii, p. 541.
58. Langlas, People of Kalapana, pp. 35–36.
59. Langlas, People of Kalapana.
60. Russell Apple, “Transcriptions of a 1974 Interview by Russell Apple with Former
Superintendent Wingate Concerning the Kalapana Extension” and “Homesite Provisions
of the 1938 Kalapana Act,” Hawai‘i Volcano National Park Headquarters Library, Janu-
ary 5, 1971.
61. U.S. Congress, Act of June 20, 1938 (52 Stat. 781 et seq.).
62. Langlas, People of Kalapana, pp. 92–94.
63. George Cooper and Gavan Daws, Land and Power in Hawai‘i (Honolulu: Bench-
mark Books, 1985), p. 259.
64. Ibid., p. 262.
65. Ibid., p. 265.
66. Ibid., p. 263.
67. County of Hawai‘i, “The General Plan Hawai‘i County,” November 1989.
68. County of Hawai‘i, “General Plan.”
69. Burdette E. Bostwick and Brian Murton, Puna Studies: Preliminary Research in
Human Ecology (Honolulu: Department of Anthropology, University of Hawai‘i, 1971).

338
notes to pages 181– 200

70. Puna Hui ‘Ohana, “Assessment of Geothermal Impact on Aboriginal Hawaiians”


(Washington, D.C.: Department of Energy, 1982).
71. Matsuoka et al., “Native Hawaiian Ethnographic Study.” The following section
draws upon the ethnographic report that resulted from the interviews Jon Matsuoka and
I conducted with ‘ohana of Puna.
72. The final findings of fact and conclusions of law; Pele Defense Fund v. the Estate
of James Campbell (Civil No. 89-089).

notes to chapter five


1. Koko Willis and Pali Jae Lee, Tales from the Night Rainbow (Honolulu: Night Rain-
bow Publishing/ Native Books, 1990), pp. 24–25.
2. Summers, Molokai, pp. 15, 198–201; Kamakau, Ruling Chiefs, pp. 150, 166, 179;
Desha, Kamehameha and His Warrior Kekuhaupi‘o, pp. 264–66.
3. Molokai Enterprise Committee, “Moloka‘i Rural Empowerment Zone Application
submitted to the United States Department of Agriculture,” October 9, 1998, vol. 2, pt.1,
sec. 1, p. 1.
4. Summers, Molokai. The review of the ruling chiefs of Moloka‘i in this and the next
paragraph is based on Summers, Molokai, pp. 11–13.
5. Summers, Molokai, pp. 14–15; Robert Schmitt, Demographic Statistics of Hawaii,
1778–1965 (Honolulu: University of Hawai‘i Press, 1968), p. 42, table 6.
6. Summers, Molokai, pp. 16–17.
7. “Bligh’s Notes on Cook’s Last Voyage,” Mariner’s Mirror 14 (October 1828): 385.
8. Summers, Molokai, p. 3.
9. Lucille De Loach, “Land and People of Molokai: An Overview” (master’s thesis,
University of Hawai‘i at Mänoa, 1975), pp. 53–54.
10. David Malo, “Causes for the Decrease,” pp. 121–30.
11. Summers, Molokai, p. 3.
12. Ibid. Summers concludes that Moloka‘i’s population was not exposed to the epi-
demic diseases that decimated the Native Hawaiian people and that the decrease of the
population should be attributed to emigration to one of the larger islands.
13. American Board of Commissioners for Foreign Missions, “Molokai Station
Reports, 1833–1849,” unpublished typescript, 1937, University of Hawai‘i Library, pp.
4–5.
14. Schmitt, Demographic Statistics, p. 42, table 6.
15. Original in Hawaiian in the archives, trans. W. H. Wilson, August 10, 1977. See
Appendix III.
16. Kanepu‘u, “Traveling About on Moloka‘i,” Ke Au Okoa, September 5 and 26,
1867.

339
notes to pages 201– 206

17. G. W. Bates, Sandwich Island Notes by a Haole (New York: Harper & Brothers,
1854), pp. 274–75, 277.
18. Lucille De Loach, “Molokai: An Historical Overview,” in Molokai Studies: Prelim-
inary Research in Human Ecology, ed. Henry Lewis (Honolulu: Department of Anthropol-
ogy, University of Hawai‘i, 1970), pp. 130–32.
19. De Loach, “Molokai,” p. 134.
20. The figure of 2,132 includes the Kalawao Hansen’s disease patients. Republic of
Hawaii, Report of the General Superintendent of the Census, 1896 (Honolulu: Hawaiian Star
Press, 1897).
21. Hawaii State Archives, Laws, 1864–65, pp. 62–64.
22. Hawaiian Kingdom Board of Health, Hansen’s Disease in Hawaii (1866), pp. 27–28.
23. This trade is described in the preceding chapter.
24. Mary Kawena Pukui, Chant H-41 c Webcor timer 753 Sel. 11, transcribed, trans-
lated, and performed by Mary Kawena Pukui, Hawaii Ethnographic Notes, Bernice
Pauahi Bishop Museum Library. The entire chant may be read at the Bishop Museum
Library. It is also transcribed and translated in Mary Kawena Pukui and Alfons Korns,
The Echo of Our Song (Honolulu: University of Hawai‘i Press, 1973).
25. Linda Greene, Exile in Paradise: The Isolation of Hawai‘i’s Hansen’s Disease Victims
and Development of Kalaupapa Settlement, 1865 to the Present, 1985 (Denver: Branch of Plan-
ning, Alaska–Pacific Northwest–Western Team, U.S. Department of Interior, National
Park Service, 1985), p. 38.
26. Ibid., p. 53.
27. Ibid., pp. 83–179. For the life of Father Damien and an excellent account of his
work at Kalaupapa see Gavan Daws, Holy Man: Father Damien of Molokai (Honolulu: Uni-
versity of Hawai‘i Press, 1984).
28. U.S. Bureau of the Census, Fifteenth Census, table 5, p. 44.
29. Geritt Judd, Pule O‘o: The Story of Molokai (Honolulu: Porter Printing, 1936),
p. 13.
30. U.S. Bureau of the Census, Fifteenth Census, table 22, p. 72.
31. George P. Cooke, Moolelo o Molokai: A Ranch Story of Molokai (Honolulu: Hono-
lulu Star-Bulletin, 1949), pp. 59–60; Vernon Charles Bottenfield, “Changing Patterns of
Land Utilization on Molokai” (master’s thesis, University of Hawai‘i at Mänoa, 1958),
pp. 88, 89; Judd, Pule O‘o, p. 17.
32. Albert Kahinu, interview by Mary Kawena Pukui, May 1, 1961, Kaunakakai, Molo-
ka‘i (Audio-Recording Collection, Anthropology Department, Bernice Pauahi Bishop
Museum), no. 108.4. I called the Moloka‘i Ranch offices on Moloka‘i and on O‘ahu to get
a record of wages and the ethnic composition of the workers from 1900 to 1930, but they
did not have employment records for that period.
33. Waldemar Duvauchelle, interview by Mary Kawena Pukui, March 7, 1961, Puko‘o,
340
notes to pages 206– 210

Moloka‘i (Audio-Recording Collection, Anthropology Department, Bernice Pauahi


Bishop Museum), no. 107.4.1.
34. Daniel Pahupu, interview by Mary Kawena Pukui, March 9, 1961, Kalama‘ula,
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 107.10; Mrs. Grace Hagerman, interview by Mary Kawena Pukui,
June 23, 1967, Kalua‘aha, Moloka‘i (Audio-Recording Collection, Department of Anthro-
pology, Bishop Museum), no. 194.1; Bottenfield, “Changing Patterns,” p. 89.
35. Baldwin, Geography of the Hawaiian Islands, p. 107.
36. John Nathan Cobb, Commercial Fisheries of the Hawaiian Islands. Agent of the U.S.
Fish Commission Report for 1901, 1902, pp. 429–30. The following fishponds were iden-
tified as being maintained: Punalau in Naiwa with 20 acres; Kamahuehue in Kamalo with
37 acres; Kainaohe in Kaamola with 17 acres; Hinau in Keawanui with 54.5 acres; Paha-
loa in Manawai with 6 acres; two nameless ones in Kaluaaha, one with 11 acres and one
with 9 acres; Kaopeahina in Kaluaaha with 20.5 acres. Niaupala in Kaluaaha with 33.5
acres; Pipio in Manulehu with 14 acres; Ilae’s pond in Puko‘o with 25 acres; a nameless
one in Kupeke with 30 acres; Nahiole in Ahaino (West) with 1 acre; Kihaloko in Ahaino
(east) with 5 acres; Waihilahila in Kailiuia with 3.5 acres; Kulaalamihi in Honomuni with
6 acres.
37. Southwick Phelps, “Regional Study of Moloka‘i, Hawai‘i,” 1941, manuscript, Ber-
nice Pauahi Bishop Museum Library, p. 14.
38. Kahaulelio, “Fishing Lore,” Nüpepa Kü‘oko‘a, trans. and comp. Mary Kawena
Pukui, Bernice Pauahi Bishop Museum Library, p. 88.
39. Phelps, “Regional Study of Molokai,” pp. 71–75, lists the ahupua‘a as follows:
Ahaino: bad prayer; Honomuni: small places close together; Honoulimalo‘o: dark, shel-
tered, dry place; Honouliwai: dark sheltered water; Kaamola: turning round; Kahananui:
great undertaking; Kailiula: red skinned; Kainalu: billowy sea; Kaluaaha: the coir net pit;
Kamalo: the dry spell; Kamanoni: place where noni grows; Kamiloloa: long milo tree;
Kapaakea: whitish rock; Kapulei: sacred wreath; Kapuokoolau: hill on windward side;
Kaunakakai: go along in company of four; Kawaikapu: forbidden water; Kawela: heat;
Keawanui: big bay; Keonokuino: six bad places; Keopukaloa: far-reaching sound; Keopu-
kauuku: small hole for sound to go through; Kumimi: poisonous crab; Kupeke: dwarflike;
Lupehu: universal plenty; Makakupaia: sentinel-like eyes; Makolelau: shriveled leaf;
Manawai: branch stream; Mapulehu: rising ash cloud; Moakea: white fowl; Moanui: great
fowl; Ohia: ohia tree; Pohakupili: nearby stone; Puaahala: bunched hala roots; Puelelu:
trumpet shell; Pukoo: supporting conch shell; Punaula: reddish spring; Puniuohua: fam-
ily coconut cup; Ualapue: hilled sweet potatoes; Waialua: 2 streams.
40. Ann Perrells, “Environmental Resources and Neighborhood Food Exchange” in
Lewis, 1970, pp. 45–51.
41. Handy, The Hawaiian Planter, pp. 101–2.
341
notes to pages 210 – 216

42. Peter Namakaeha, interview by Mary Kawena Pukui, March 10, 1961, Honouli-
wai, Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pau-
ahi Bishop Museum), no. 107.15.1.
43. Handy, The Hawaiian Planter, 1:102.
44. Daniel Pahupu, interview by Mary Kawena Pukui, March 9, 1961, no. 107.10.
45. Peter Namakaeha, interview by Mary Kawena Pukui, March 10, 1961, no.
107.15.1.
46. Phelps, “Regional Study of Moloka‘i,” p. 70, lists them as follows: Hälawa: valley
of sufficient water; Kaiamiki: reduced fish; Kiloa: tall ti plants; Pelekunu: strong smelling;
Waikolu: three waters; Wailau: many waters; Wawaelepe: twisted foot.
47. Fred Tollefson, interview by Mary Kawena Pukui, June 23, 1967, Keoneniuomana,
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 194.1–194.2.
48. Kenneth P. Emory, “Windward Molokai,” Mid-Pacific Magazine, November 1916,
p. 446.
49. Phelps, “Regional Study of Moloka‘i,” p. 42.
50. James Poaha, interview by Mary Kawena Pukui, March 8, 1961, Ho‘olehua, Kala-
ma‘ula, Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice
Pauahi Bishop Museum), no. 107.9.
51. Emory, “Windward Molokai,” pp. 443–47.
52. Harriet Ne in Michael Dooley and Harry James Mowat, Na Manao O Na Kupuna:
An Oral History of Hawaii (Kaunakakai: Puu-o-Hoku Media Service, 1979), p. 5.
53. Ibid., pp. 5, 7, 9.
54. Ibid., p. 6.
55. Ibid., p. 9.
56. Daniel Napela Naki, interview by Mary Kawena Pukui, March 6, 1961, Honouli-
wai, Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pau-
ahi Bishop Museum), no. 107.1.2. Concerning the school, see Bottenfield, “Changing
Patterns,” p. 75. Concerning the date when Mrs. Wilson got married and moved out, see
speech by Mrs. Kealiinohomoku in University of Hawai‘i Music Department forum,
March 19, 1987.
57. Information received from Mrs. Harriet Ne in response to inquiry from me
through her grandson Edward Ayau on August 24, 1989. Also information found in Mar-
ion Kelly, “Cultural History of Pelekunu Valley, Moloka‘i,” manuscript for the Nature
Conservancy (Honolulu, March 9, 1988), p. 21.
58. Nüpepa Kü‘oko‘a, August 19, 1921.
59. Daniel Napela Naki, interview by Mary Kawena Pukui, March 6, 1961, no.
107.1.2.

342
notes to pages 216– 221

60. Ibid.
61. Amoy Duvauchelle, interview by Mary Kawena Pukui, March 7, 1961, Puko‘o,
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 107.5.
62. Mrs. Emma Apana, interview by Mary Kawena Pukui, May 1, 1961, Kamalo,
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 108.06.1, 2.
63. J. Kaopuiki, interview by Mary Kawena Pukui, March 9, 1961, Kaimiloloa, Puko‘o,
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 107.13.
64. Daniel Naki, interview by Mary Kawena Pukui, March 6, 1961, no. 107.1.2.
65. Ibid.
66. Ibid.
67. Thomas Riley, Wet and Dry in a Hawaiian Valley: The Archaeology of an Agricultural
System (Ph.D. diss., University of Hawai‘i at Mänoa, 1973), p. 79.
68. Honolulu Star-Bulletin, May 18, 1935, 3rd section, p. 1.
69. Handy, Hawaiian Planter, p. 101.
70. Honolulu Star-Bulletin, May 18, 1935; Riley, 1973, p. 81.
71. John and Edith Akina, interview by Mary Kawena Pukui, May 1, 1961, Kumuele,
Kamalo, Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice
Pauahi Bishop Museum), no. 108.6.1.
72. Rebecca Uahinui, interview by Mary Kawena Pukui, May 1, 1961, Kalama‘ula
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 108.5.1.
73. Ibid.
74. Edith Akina, interview by Mary Kawena Pukui, May 1, 1961, no. 108.6.1.
75. James Poaha, interview by Mary Kawena Pukui, March 8, 1961, no. 107.9.
76. Sarah Naoo, interview by Mary Kawena Pukui, March 6, 1961, Honouliwai, Molo-
ka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi Bishop
Museum), no. 107.1.
77. Ibid.
78. Sarah Ka‘ai Kalima, interview by Mary Kawena Pukui, May 1, 1961, Kalua‘aha,
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 108.7.
79. James Poaha, interview by Mary Kawena Pukui, March 8, 1961, no. 107.9; Sarah
Ka‘ai Kalima, interview by Mary Kawena Pukui, May 1, 1961, no. 108.7.
80. Sarah Ka‘ai Kalima, interview by Mary Kawena Pukui, May 1, 1961, no. 108.7.
81. Ibid.; Mitchell Pau‘ole, interview by Mary Kawena Pukui, March 8, 1961, Ho‘ole-

343
notes to pages 221– 223

hua Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pau-


ahi Bishop Museum), no. 107.7.1.
82. Waldemar Duvauchell, interview by Mary Kawena Pukui, March 7, 1961, no.
107.4.1; Sarah Ka‘ai Kalima, interview by Mary Kawena Pukui, May 1, 1961, no. 108.7;
Zellie Duvauchelle Sherwood, interview by Mary Kawena Pukui, April 21, 1964, Puko‘o,
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 153.1.3.
83. Sarah Ka‘ai Kalima, interview by Mary Kawena Pukui, May 1, 1961, no. 108.7.
84. Zellie Duvauchelle Sherwood, interview by Mary Kawena Pukui, April 21, 1964,
no. 153.1.3.
85. Fred Tollefson, interview by Mary Kawena Pukui, June 23, 1967, no. 194.1–194.2;
J. Kaopuiki, interview by Mary Kawena Pukui, March 9, 1961, no. 107.13; James Poaha,
interview by Mary Kawena Pukui, March 8, 1961, no. 107.9.
86. Grace Hagerman, interview by Mary Kawena Pukui, June 23, 1967, no. 194.1
87. Mitchell Pau‘ole, interview by Mary Kawena Pukui, March 8, 1961, no. 107.7.1;
Rachel Dudoit, interview by Mary Kawena Pukui, May 2, 1961, Kumimi, Moloka‘i
(Audio-Recording Collection, Department of Anthropology, Bernice Pauahi Bishop
Museum), no. 108.8.1.2.
88. Waldemar Duvauchelle, interview by Mary Kawena Pukui, March 7, 1961, no.
107.4.1.
89. Daniel Naki, interview by Mary Kawena Pukui, March 6, 1961, no. 107.1.2.
90. Waldemar Duvauchelle, interview by Mary Kawena Pukui, March 7, 1961, no.
107.4.1. He spoke of the use of motorboats, sampans, and modern fishing equipment.
91. Mrs. Hagerman said, “All Hawaiian words spoken on the school grounds were
punishable.” She said that she was kept in after school for speaking in Hawaiian at school.
Interview by Mary Kawena Pukui, June 23, 1967, no. 194.1.
92. Sung by Mrs. Sam K. Enos in interview by Mary Kawena Pukui, April 29, 1961,
Moanui, Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice
Pauahi Bishop Museum), no. 108.3. The song, “Nani Hälawa,” was written by David
Kala‘au and has seven more verses (only three of these are translated here): Hanohano o
Kalua‘aha / Iä Moloka‘i Nui A Hina / Kü kilakila i ka la‘i / Me ke kai nehe i ke one //
Ho‘oheno o Kamalo / Ka waiho ka hela mai i ka la‘i / Me püpü kani ‘oe / Me ka makani
lawe ‘ehu kai // Nani wale ku‘u ‘ike ana / Iä Kawela i ka la‘i anu / Ilaila mäkou i ‘ike iho
ai / I ke ‘one holu o Kamiloloa // Uluwehi no la‘i a ka mau‘u / I ka leo honehone o ke kio-
wea / Ilaila ho‘ohihi ka mana‘o / I ka malu lau niu o Kalama‘ula // Kilakila Kauluwai kau
mai i luna / Ipu ia i ke ‘ala me ka onaona / Home noho ou e kuke / Ka mäkua o ka lehu-
lehu // Uluwehi Kala‘e i ka iuiu / I ka pä kolonahi a ke kehau / Ho‘ope ne i ka lehua / Me

344
notes to pages 224– 230

maile lau li‘i o ke kuahiwi // Upu a‘e ka mana‘o ho‘ohihi / E ‘ike i ka ‘äina ho‘opulapula /
Ka waiho ka hele Ho‘olehua / O ke heke no ia i ka‘u‘ike [Distinguished Kalua‘aha / Of
Moloka‘i Great Child of Hina / Standing majestic and silent / With the rustling ocean
on the sand // Cherished Kamalo / Lying there spread out calmly / With the singing land
shell / With the wind that brings the sea spray // Singularly beautiful my viewing /
Toward Kawela in the cool calm / There we came to know / The sand carried back and
forth by the sea of Kamiloloa . . .]
93. Milton Bloombaum and Ted Gugelyk, Ma‘i Ho‘oka‘awale: The Separating Sickness
(Honolulu: Social Science Research Institute, 1979), p. 11.
94. Ibid., p. 79.
95. Ibid., p. 37.
96. Greene, Exile in Paradise, pp. 312–82.
97. Hawaii (Territory), Governor, Report of the Governor’s Advisory Committee on Lep-
rosy in Hawaii (October 1930), p. 7.
98. Ibid., p. 8.
99. Ibid., p. 10.
100. Greene, Exile in Paradise, p. 384.
101. Bloombaum and Gugelyk, Ma‘i Ho‘oka‘awale, p. 27.
102. Grace Humphries, “Hawaiian Homesteading: A Chapter in the Economic Devel-
opment of Hawai‘i” (master’s thesis, University of Hawai‘i at Mänoa, 1937), p. 37.
103. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission to the Leg-
islature of Hawaii, Regular Session (1925), p. 9.
104. Felix M. Keesing, Hawaiian Homesteading on Moloka‘i, University of Hawaii
Research Publications 12 (New York: AMS Press, 1936), p. 56.
105. Ibid., pp. 19, 27.
106. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission to the Leg-
islature of Hawaii, Regular Session (1923), pp. 5–11.
107. Ibid., p. 11.
108. Ibid., p. 12.
109. Keesing, Hawaiian Homesteading on Molokai, p. 57.
110. Hawaii ( Territory), Governor, Report of Hawaiian Homes Commission (1923),
p. 12.
111. Ibid., pp. 6, 13.
112. Ibid., p. 11.
113. Ibid., p. 13.
114. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission (1925),
pp. 10, 19–22.

345
notes to pages 231– 237

115. Keesing, Hawaiian Homesteading on Molokai, p. 28. Hawaii (Territory), Governor,


Report of the Hawaiian Homes Commission to the Legislature of Hawaii Regular Session (1927),
pp. 7, 26–33.
116. Keesing, Hawaiian Homesteading on Molokai, p. 57.
117. Ibid.
118. Henry P. Judd, The Friend, September 1936, p. 200.
119. Gertrude and Mitchell Pau‘ole, interview by Mary Kawena Pukui, March 8, 1961,
Ho‘olehua, Moloka‘i (Audio-Recording Collection, Department of Anthropology,
Bernice Pauahi Bishop Museum), no. 107.6.
120. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission to the Leg-
islature of Hawaii, Regular Session (1929), p. 10.
121. Keesing, Hawaiian Homesteading on Molokai, p. 30.
122. Ibid., p. 65.
123. Ibid., pp. 72–74.
124. Ibid., pp. 71–84; Humphries, “Hawaiian Homesteading,” pp. 69–74.
125. Keesing, Hawaiian Homesteading on Molokai, p. 76.
126. Harry Hanakahi, interview by Mary Kawena Pukui, March 9, 1961, Kalama‘ula
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum), no. 107.12.2.
127. Gertrude Pau‘ole, interview by Mary Kawena Pukui, March 8, 1961, Ho‘olehua,
Moloka‘i (Audio-Recording Collection, Department of Anthropology, Bernice Pauahi
Bishop Museum).
128. Mitchell Pau‘ole, interview by Mary Kawena Pukui, March 8, 1961, no. 107.7.1.
129. “Homesteaders’ Future Looms Bright on Island of Molokai; Ask Even Chance
to Make Good,” Honolulu Advertiser, March 7, 1927, p. 9.
130. Ibid.
131. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission (1929),
p. 5; U.S. Congress, House Committee on the Territories, 70th Cong., 1st sess., H.R.
6989, A Bill to Amend the Hawaiian Homes Commission Act, 1920, approved July 9, 1921, As
Amended by Act of February 3, 1923, pp. 5–6.
132. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission (1929),
pp. 5–6.
133. Mitchell Pau‘ole, interview by Mary Kawena Pukui, March 8, 1961, no. 107.7.1.
134. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission to the Leg-
islature of Hawaii, Regular Session (1931), p. 7.
135. Wallace R. Farrington, “Two Days with Hawaiian Homesteaders,” Star-Bulletin,
October 4, 1930, p. 3.

346
notes to pages 237– 246

136. Wallace R. Farrington, “Two Days with Hawaiian Homesteaders,” Star-Bulletin,


October 7, 1930, p. 7. Although there were no written reports, there were general criti-
cisms from planter interests who still opposed the program.
137. Star-Bulletin, October 6, 1930, p. 7.
138. Ibid., October 8, 1930, p. 1.
139. Ibid., p. 7.
140. Ibid., October 6, 1930, p. 7.
141. Ibid., October 9, 1930, p. 4.
142. Humphries, “Hawaiian Homesteading,” p. 53.
143. Ibid., pp. 53–54.
144. Keesing, Hawaiian Homesteading on Molokai, pp. 101–4.
145. Ibid., p. 106.
146. Mitchell and Gertrude Pau‘ole, interview by Mary Kawena Pukui, March 8,
1961, no. 107.6.
147. Keesing, Hawaiian Homesteading on Molokai, p. 115.
148. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission (1929),
p. 11.
149. Keesing, Hawaiian Homesteading on Molokai, p. 109.
150. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission (1925),
pp. 18, 40.
151. Keesing, Hawaiian Homesteading on Molokai, pp. 31–32.
152. Hawaii (Territory), Governor, Report of the Hawaiian Homes Commission (1925),
pp. 16–17.
153. Mitchell Pau‘ole, interview by Mary Kawena Pukui, March 8, 1961, no. 107.7.1.
154. Cooke, Moolelo o Molokai, pp. 22–23.
155. Lewis, Molokai Studies.
156. Penelope Canan et al., Moloka‘i Data Book: Community Values and Energy Develop-
ment (Honolulu: Urban and Regional Planning Program, University of Hawai‘i at Mänoa,
1981).
157. Roger Anderson, Nick Huddleston, and Masa Yokota, “A Feasibility Study for the
Implementation of the Concept of Traditional Hawaiian Land Use in Pelekunu Valley,
Moloka‘i,” University of Hawai‘i School of Architecture, December 21, 1982.
158. Hawai‘i, Department of Hawaiian Home Lands, Kalama‘ula Development Plan
(Honolulu: DHHL, 1983).
159. Decision Analysts Hawaii, “An Economic Development Strategy and Implemen-
tation Program for Moloka‘i,” 1985.
160. Matsuoka, McGregor, and Minerbi, “Governor’s Moloka‘i Subsistence Task
Force Final Report.”

347
notes to pages 248– 260

161. Informants reported that subsistence rates increased after the closure of Del
Monte, yet because there are no baseline measures, this belief cannot be empirically
verified.
162. Molokai Enterprise Community, “Moloka‘i Rural Empowerment Zone,” vol. 2,
pt. 1, sec. 1, p. 1.
163. Ibid., secs. 2, 3, 4.

notes to chapter six


1. For a discussion of these changes see Kent, Hawaii.
2. Kaho‘olawe Island Conveyance Commission (KICC), Kaho‘olawe: Restoring a Cul-
tural Treasure, Final Report of the Kaho‘olawe Island Conveyance Commission to the
Congress of the United States, March 31 (Maui, 1993; referred to hereinafter as KICC
report, 1993).
3. “Kaho‘olawe: Rebirth of a Sacred Hawaiian Island,” exhibit by Bishop Museum at
the Smithsonian Institution’s Arts and Industries Building, June 5–September 2, 2002.
4. Statistics cited in Davianna McGregor-Alegado, “Hawaiians: Organizing in the
1970s,” Amerasia 7, no. 2 (1980): 29–55, such as: In 1972 one-third of all Native Hawai-
ians earned poverty-level incomes of $4,000 or below. In 1970 Hawaiians comprised 30
percent of the welfare recipients and 49.5 percent of all adult prison inmates. The unem-
ployment rate for Native Hawaiians was higher than that of the general population. Only
50 percent of adult Hawaiians over twenty-five years of age had graduated from high
school, and the dropout rate for Native Hawaiians was 23 percent, compared to the state
rate of 13 percent.
5. Fornander, Fornander Collection, vol. 6, p. 360.
6. Ibid., vol. 4, p. 12.
7. Malo, Mo‘olelo Hawai‘i, p. 243. I have reinterpreted the Hawaiian translation.
8. Beckwith, Hawaiian Mythology, p. 170.
9. Beckwith, ed., Kepelino’s Traditions of Hawai‘i, pp. 187–88.
10. The other three sites are located in Halema‘uma‘u Crater at Palikapuokamoho-
ali‘i; in a shark cave in the reef near the entrance to Pearl Harbor; and on Ni‘ihau.
11. Beckwith, Hawaiian Mythology, p. 129.
12. Fornander, Fornander Collection, vol. 5, pp. 514–19.
13. Nainoa Thompson and Gordon Pi‘ianaia related this to me in October 1991 when
the Hokule‘a and its training crew visited Kaho‘olawe.
14. Fornander, Fornander Collection, vol. 4, p. 128.
15. Ibid., vol. 6, p. 281.
16. Kamakau, Tales and Traditions, pp. 104–5.

348
notes to pages 260 – 275

17. Peter H. Buck, Arts and Crafts of Hawai‘i, Section VI: Canoes, Bernice Pauahi Bishop
Museum Special Publication 45 (Honolulu: Bishop Museum Press, 1964), p. 283.
18. Kamakau, Tales and Traditions, p. 112.
19. Ibid., p. 114.
20. Beckwith, Hawaiian Mythology, p. 20.
21. Ibid., p. 22.
22. Thrum, “Aiai, Son of Ku‘ula,” in Hawaiian Folk Tales, p. 238.
23. Ibid., March 7, 1902.
24. Kahaulelio, “He Mau Kuhikuhi.”
25. Ibid.
26. Ibid.
27. Harry Künihi Mitchell, interview by Davianna McGregor, November 1989.
28. Nathan Napoka, “Kahoolawe Place Names,” in Carol Silva, Kahoolawe Cultural
Study, Part 1: Historical Documentation, prepared for the Pacific Division, Naval Facilities
Engineering Command, U.S. Navy, Pearl Harbor, Honolulu, April 1983.
29. Ibid.
30. Ibid.
31. Edith Kanaka‘ole Foundation, “E Mau Ana o Kanaloa, Ho‘i Hou: The Persever-
ance of Kanaloa, Return! The Cultural Practices and Values Established at Kanaloa/
Kaho‘olawe Past and Present,” Kaho‘olawe Island Conveyance Commission Consultant
Report No. 12 (Wailuku, 1993), pp. 45–46.
32. Shuzo Uemoto, Nana I Na Loea Hula: Look to the Hula Resources, with narratives by
Hula Resources (Honolulu: Kalihi-Palama Culture and Arts Society, 1997), p. 54.
33. Edith Kanaka‘ole Foundation, http://www.edithkanakaolefoundation.org (accessed
December 30, 2005).
34. Edith Kanaka‘ole Foundation, “E Mau Ana O Kanaloa,” pp. 52–53.
35. Anderson explained this to Noa Emmett Aluli and me in September 1990 when
he asked us to provide a white paper about the island of Kaho‘olawe.
36. B. J. Reyes, “With Little Fanfare, Kahoolawe Island Returns to Hawaiian Con-
trol,” Associated Press, November 12, 2003, http://zwire.com /site/news.cfm?newsid
=10504278&BRD=1817&PAG=461&dept_id=222077&rfi=6 (accessed January 4, 2006).
37. George Bush, “Memorandum on the Kaho‘olawe, Hawaii, Weapons Range,”
October 22, 1990. The text of the memo reads: “Memorandum for the Secretary of
Defense. Subject: Use of the Island of Kaho‘olawe, Hawaii, as a Weapons Range. You are
directed to discontinue use of Kaho‘olawe as a weapons range effective immediately. This
directive extends to use of the island for small arms, artillery, naval gunfire support, and
aerial ordnance training. In addition, you are directed to establish a joint Department of

349
notes to pages 276 – 279

Defense-State of Hawaii commission to examine the future status of Kaho‘olawe and


related issues.” Jet planes ready to leave Kane‘ohe Marine Corps Air Station for bomb-
ing runs on Kaho‘olawe were grounded on the morning of October 22, 1990. http://
bushlibrary.tamu.edu/research/papers/1990/90102203.html (accessed December 30,
2005).
38. Hawai‘i Revised Statutes, Chapter 6K.
39. George S. Kanahele, Hawaiian Renaissance (Honolulu: Project Waiaha, 1982), and
George Hu‘eu Stanford Kanahele, Ku Kanaka: Stand Tall; A Search For Hawaiian Values
(Honolulu: University of Hawai‘i Press and Waiaha Foundation, 1986), documented and
contributed to the Native Hawaiian cultural renaissance.
40. U.S. Department of Interior, 1993–1998, “Federal Indian Policies,” June 12, 1998.
41. U.S. Department of Interior and Department of Justice, “From Mauka to Makai:
The River of Justice Must Flow Freely; Report on the Reconciliation Process between the
Federal Government and Native Hawaiians” (Washington, D.C., October 23, 2000).
42. Ibid., p. 56.
43. In the Apology Resolution, Native Hawaiian (both words capitalized) is defined
as “any individual who is a descendant of the aboriginal people who, prior to 1778, occu-
pied and exercised sovereignty in the area that now constitutes the State of Hawai‘i.” The
Hawaiian Homes Commission Act and the Admission Act use the term native Hawaiian
(lowercase n, capital H) to mean “Any descendant of not less than one-half part of the
blood of the races inhabiting the Hawaiian Islands previous to 1778.” The term Kanaka
Maoli is promoted as the indigenous name for Native Hawaiians. However, Kanaka Maoli
simply means native or indigenous, while Kanaka Maoli Hawai‘i means native or indige-
nous Hawaiian. In the 1859 Civil Code of the Kingdom of Hawai‘i, chapter VIII, kanaka
Hawaii is used to translate “native of the Hawaiian Islands” and ke kanaka maoli is used
to translate “native.” In the 1878 Census of the Kingdom of Hawai‘i, He kane kanaka
maoli (Hawaii) was used for “Native Male” and He wahine kanaka maoli (Hawaii) was used
for “Native Female.” However, this must have referred to pure Hawaiians only, as there
were categories for “Half-Caste Male”—He hapahaole kane—and “Half-Caste Female”—
He hapahaole wahine. In this chapter, I will use “Native Hawaiian” to refer to anyone who
has Hawaiian ancestry, i.e., who is descended from a Kanaka Maoli Hawai‘i ancestor, and
“native Hawaiian” to refer to those who are of half or more Hawaiian ancestry. Under the
law, thus far only “native Hawaiians” are the beneficiaries of the Hawaiian Home Lands
and the ceded public lands trusts, as discussed below.
44. Rice v. Cayetano, 528 US 495 (2000) Docket No. 98-818, February 23, 2000. Jus-
tice Anthony Kennedy delivered the opinion of the Court, in which Chief Justice William
Rehnquist and Justices Sandra Day O’Connor, Antonin Scalia, and Clarence Thomas
joined. Justice Stephen Breyer filed an opinion concurring in the result, in which Justice

350
notes to pages 280 – 286

David Souter joined. Justice John Paul Stevens filed a dissenting opinion in which Justice
Ruth Bader Ginsburg joined as to Part II. Justice Ginsburg filed a dissenting opinion.
45. These questions were raised in the following statement: “If Hawaii’s restriction
were to be sustained under Mancari we would be required to accept some beginning prem-
ises not yet established in our case law. Among other postulates, it would be necessary to
conclude that Congress, in reciting the purposes for the transfer of lands to the State—
and in other enactments such as the Hawaiian Homes Commission Act and the Joint
Resolution of 1993—has determined that native Hawaiians have a status like that of Indi-
ans in organized tribes, and that it may, and has, delegated to the State a broad authority
to preserve that status. These propositions would raise questions of considerable moment
and difficulty.”
46. Rice v. Cayetano. 528 US 495 (2000).
47. Daniel Akaka, “Statements on Introduced Bills and Joint Resolutions,” S. 746,
107th Cong., 1st sess., Congressional Record (April 6, 2001), p. S3757.
48. The complete vision statement reads: “The kino (body) of Kanaloa is restored.
Forests and shrublands of native plants and other biota clothe its slopes and valleys. Pris-
tine ocean waters and healthy reef ecosystems are the foundation that supports and sur-
rounds the island. Nä po‘e Hawai‘i (the people of Hawai‘i) care for the land in a manner
which recognizes the island and ocean of Kanaloa as a living spiritual entity. Kanaloa is
a pu‘uhonua (spiritual refuge) and wahi pana (sacred place) where Native Hawaiian cul-
tural practices flourish. The piko (navel) of Kanaloa is the crossroads of past and future
generations from which the Native Hawaiian lifestyle spreads throughout the islands.”
49. PBR-Hawai‘i with Pualani Kanahele et al., Community Planning, Inc., and Geo-
graphic Decision Systems International, Palapala Ho‘onohonoho Moku‘äina O Kaho‘olawe:
Kaho‘olawe Use Plan (Wailuku: Kaho‘olawe Island Reserve Commission, 1995), pp. 3-1 to
3-5.
50. At Kalae, South Point, on the island of Hawai‘i a fishing ko‘a marks a fishing
ground that is eight miles from the shoreline.
51. Pualani Kanaka‘ole Kanahele, “He Ko‘ihonua No Kanaloa, He Moku,” in Kaho‘o-
lawe Nä Leo o Kanaloa: Chants and Stories of Kaho‘olawe, photographs by Wayne Levin et
al. (Honolulu: ‘Ai Pöhaku Press, 1995), pp. 97–109.

notes to chapter seven


1. Kanaka maoli has been popularized by advocates of Native Hawaiian sovereignty and
independence as the appropriate indigenous term for Native Hawaiians. Nevertheless, I
prefer to use the term kanaka ‘öiwi, which also means Native Hawaiian, because it implies
ancestry as a component of being Native Hawaiian. Maoli means native, indigenous, gen-
uine, true, and real, whereas ‘öiwi, which means native and native son, can be literally

351
notes to pages 288–304

translated as “of the ancestral bone.” For Native Hawaiians, the bones of our ancestors
and ourselves are sacred and hold the essence of the soul and spirit of ourselves, our pred-
ecessors, and our descendants. Thus, within our iwi resides our mana, which in large part
has been transmitted to us over the generations from our ancestors and will pass on
through us to our descendants. Herein also is the core of our ancestral memory and
knowledge.
2. Honolulu Star-Bulletin, March 18, 2004.
3. http://www.bishopmuseum.org/research/natsci /waipiostudy/ background/index
.html (accessed January 4, 2006).
4. Group 70, McGregor, and Cultural Surveys Hawai‘i, “Kalo Kanu o Ka ‘Äina,” p. 5.
5. Kanahele, foreword to James, Ancient Sites of O‘ahu, ix–xi.
6. New translations of these works provide more accurate accounts of the mo‘olelo
written by these scholars.
7. The Hawai‘i State Supreme Court ruled that Native Hawaiians continued to have
gathering rights in the Wao Kele o Puna Forest even after the State of Hawai‘i had trans-
ferred ownership of the forest to the Campbell Estate under a land exchange. The Hilo
district court held hearings in August 1994 and received testimony of Puna residents
regarding their traditional and customary usage of the Wao Kele o Puna forest.
8. Moloka‘i Enterprise Community, “Moloka‘i Rural Empowerment Zone Applica-
tion Submitted to the United States Department of Agriculture” (October 9, 1998), vol.
2, pt. 1, sec. 1, p. 3.
9. Ibid., vol. 1, sec. 3, p. 3.
10. Noa Emmett Aluli, foreword to Kaho‘olawe, Nä Leo o Kanaloa: Chants and Stories of
Kaho‘olawe (Honolulu: ‘Ai Pöhaku Press, 1995), p. xiv.

352
 bibliography 

Published and Online Materials, Reports, Oral Histories, and Interviews

Allen, Melinda Sue. “The Kalapana Extension in the 1800s: A Research of the Historical
Records.” Unpublished manuscript prepared for the National Park Service, 1979.
Hawaii Volcanoes National Park.
American Board of Commissioners for Foreign Missions, p. 40. “Molokai Station Reports,
1833–1849.” Unpublished typescript, 1937. University of Hawai‘i Library.
Anderson, Roger, Nick Huddleston, and Masa Yokota. “A Feasibility Study for the
Implementation of the Concept of Traditional Hawaiian Land Use in Pelekunu Val-
ley, Moloka‘i.” Unpublished manuscript, December 21, 1982. University of Hawai‘i
School of Architecture.
Apple, Russell. “Homesite Provisions of the 1938 Kalapana Act.” Unpublished manu-
script, January 5, 1971. Hawai‘i Volcano National Park Headquarters Library.
———. “Transcriptions of a 1971 Interview by Russell Apple with Former Superinten-
dent Wingate Concerning the Kalapana Extension.” Unpublished manuscript, Janu-
ary 5, 1971. Hawai‘i Volcano National Park Headquarters Library.
Armstrong, R. Warwick, and Henry T. Lewis, eds. North Kohala Studies: Preliminary
Research in Human Ecology. Honolulu: University of Hawai‘i Press, 1970.
Baldwin, Charles. Geography of the Hawaiian Islands. New York: American Book, 1908.
Barrere, Dorothy. “Political History of Puna.” Unpublished manuscript for Archaeology
Reconnaissance of the Kalapana Extension, 1959. Bishop Museum.
———, comp. “The King’s Mähele: The Awardees and Their Lands.” Unpublished
bound manuscript, Hamilton Library, University of Hawai‘i at Mänoa, 1994.
Bates, G. W. Sandwich Island Notes by a Haole. New York: Harper & Brothers, 1854.
Beaglehole, Ernest. Some Modern Hawaiians. Research Publications 19. Honolulu: Uni-
versity of Hawai‘i Press, 1937.
Beaglehole, J. C. The Journals of Captain James Cook on His Voyages of Discovery. Vol. 3, The
Voyage of the Resolution and Discovery, 1776–1780. Cambridge: Hakluyt Society,
1955–1974.
Beckwith, Martha W. Hawaiian Mythology. Honolulu: University of Hawai‘i Press, 1970.
———, ed. Kepelino’s Traditions of Hawai‘i. Bernice Pauahi Bishop Museum Bulletin 95.
Honolulu: B. P. Bishop Museum Press; New York: Kraus Reprint, 1971.

353
bibliography

Bishop Museum. Waipo Valley Stream Restoration Study. Hawai‘i Biological Survey. http://
www.bishopmuseum.org/research/natsci /waipiostudy/ background/index.html
(accessed December 30, 2005).
“Bligh’s Notes on Cook’s Last Voyage.” Mariner’s Mirror 14 (October 1828).
Bloombaum, Milton, and Ted Gugelyk. Ma‘i Ho‘oka‘awale: The Separating Sickness. Hono-
lulu: Social Science Research Institute, 1979.
Bostwick, Burdette E., and Brian Murton. Puna Studies: Preliminary Research in Human
Ecology. Honolulu: Department of Anthropology, University of Hawai‘i, 1971.
Bottenfield, Vernon Charles. “Changing Patterns of Land Utilization on Molokai.” Mas-
ter’s thesis, University of Hawai‘i, 1958.
Buck, Peter H. Arts and Crafts of Hawai‘i, Section VI: Canoes. Bernice Pauahi Bishop
Museum Special Publication 45. Honolulu: Bishop Museum Press, 1964.
Burtchard, Greg, and Pennie Moblo. “Archaeology in the Kilauea East Rift Zone
Kapoho, Kama‘ili and Kilauea Geothermal Subzones, Puna District, Hawai‘i Island.”
Report ORNL/SUB/94-SN150 /1-2, July. Oak Ridge, Tenn.: Oak Ridge National
Laboratory, 1994.
Canan, Penelope, Michael Hennessy, Kathleen Kinsella Miyasiro, Michael Shiroma, Lee
Sicher, Debra Lewis, David C. Matteson, Lynette Kono, William Dendle, and Jeffrey
M. Melrose. Moloka‘i Data Book: Community Values and Energy Development. Honolulu:
Urban and Regional Planning Program, University of Hawai‘i Press, 1981.
Cannelora, Louis. “The Origin of Hawaii Land Titles and of the Rights of Native Ten-
ants.” Honolulu: Security Title, 1974.
Cartwright, Bruce. “Some Aliis of the Migratory Period.” Bishop Museum Occasional
Papers, vol. 10, no. 7. Honolulu: Bernice Pauahi Bishop Museum, 1933.
Chinen, J. H. “The Great Mähele: Hawaii’s Land Division of 1848.” Honolulu: Univer-
sity of Hawai‘i Press, 1978.
Chun, Malcolm Naea, and Matthew Spriggs. “New Terms Suggested for Early Hawai-
ian History.” Ka Wai Ola O OHA (February 1987).
Cobb, John Nathan. Commercial Fisheries of the Hawaiian Islands. U.S. Fish Commission
Report for 1901. Washington: Government Printing Office, 1902.
Cooke, George P. Moolelo o Molokai: A Ranch Story of Molokai. Honolulu: Honolulu Star-
Bulletin, 1949.
Cooper, George, and Gavan Daws. Land and Power in Hawai‘i. Honolulu: Benchmark
Books, 1985.
Cordy, Ross. Exalted Sits the Chief: The Ancient History of Hawai‘i Island. Honolulu: Mutual,
2000.
Coulter, John W. Population and Utilization of Land and Sea in Hawai‘i, 1853. Bernice Pau-
ahi Bishop Museum Bulletin 88. Honolulu: Bernice Pauahi Bishop Museum, 1931.

354
bibliography

Damon, Ethel. Father Bond of Kohala: A Chronicle of Pioneer Life in Hawaii. Honolulu: The
Friend, 1927.
Davenport, William. “The Hawaiian ‘Cultural Revolution’: Some Economic and Politi-
cal Considerations.” American Anthropologist 71 (1969).
Daws, Gavan. Holy Man: Father Damien of Molokai. Honolulu: University of Hawai‘i Press,
1984.
Decision Analysts Hawaii. “An Economic Development Strategy and Implementation
Program for Moloka‘i.” 1985.
De Loach, Lucille. “Land and People of Molokai: An Overview.” Master’s thesis, Uni-
versity of Hawai‘i at Mänoa, 1975.
———. “Molokai: An Historical Overview.” In Molokai Studies: Preliminary Research in
Human Ecology, edited by Henry Lewis. Honolulu: Department of Anthropology, Uni-
versity of Hawai‘i, 1970.
Desha, Stephen L. Kamehameha and His Warrior Kekuhaupi‘o. Translated by Frances N.
Frazier. Honolulu: Kamehameha Schools Press, 2000.
Dooley, Michael, and Harry James Mowat. Na Manao o Na Kupuna: An Oral History of
Hawaii. Kaunakakai: Puu-o-Hoku Media Service, 1979.
Duvauchelle, Waldemar. Interview by Mary Kawena Pukui. March 7, 1961. Puko‘o,
Moloka‘i. Audio-Recording Collection, Anthropology Department, Bernice Pauahi
Bishop Museum, no. 107.4.1.
Earle, Timothy K. Economic and Social Organization of a Complex Chiefdom: The Halele‘a
District, Kaua‘i, Hawai‘i. Unpublished manuscript, 1978. Anthropological Papers,
Museum of Anthropology 63, University of Michigan.
Eckbo, Dean, Austin, and Williams, with Morris Fox, consultant. “H-3 Socio-Economic
Study: The Effects of Change on a Windward Oahu Rural Community.” Unpublished
report. Honolulu, 1973.
Edith Kanaka‘ole Foundation. http://www.edithkanakaolefoundation.org (accessed
December 30, 2005).
———. “E Mau Ana O Kanaloa, Ho‘i Hou: The Perseverance of Kanaloa, Return! The
Cultural Practices and Values Established at Kanaloa/Kaho‘olawe Past and Present.”
Kaho‘olawe Island Conveyance Commission Consultant Report 12. Wailuku, 1993.
Elbert, Samuel H., and Noelani Mahoe. Na Mele o Hawai‘i Nei: 101 Hawaiian Songs.
Honolulu: University of Hawai‘i Press, 1970.
Ellis, William. Journal of William Ellis, Narrative of a Tour of Hawaii, or Owhyhee: With
Remarks on the History, Traditions, Manners, Customs and Language of the Inhabitants of
the Sandwich Islands. Honolulu: Advertiser, 1963.
Emory, Kenneth P. “Windward Molokai.” Mid-Pacific Magazine, November 1916.
Ethnic Studies Oral History Project. “Waipi‘o, Mano Wai: An Oral History Collection.”

355
bibliography

2 vols. Unpublished manuscript, 1978. Honolulu: Ethnic Studies Program, University


of Hawai‘i at Mänoa.
Fornander, Abraham. An Account of the Polynesian Race: Its Origins and Migrations and the
Ancient History of the Polynesian People to the Times of Kamehameha I. Vol. 2. London:
Trubner, 1880.
———. Fornander Collection of Hawaiian Antiquities and Folk-lore: The Hawaiians’ Account
of the Formation of Their Islands and Origins of Their Race, with the Traditions of Their
Migrations as Gathered from Original Sources. Edited by Thomas G. Thrum. Bernice
Pauahi Bishop Museum Memoirs, vols. 4–6. Honolulu: Bishop Museum, 1916–1920.
Forster, John. “Social Organization and Differential Social Change in Two Hawaiian
Communities.” International Journal of Comparative Sociology 3 (December 1962).
Fuchs, Lawrence. Hawaii Pono: A Social History. San Diego: Harcourt, Brace & World,
1961.
Greene, Linda. Exile in Paradise: The Isolation of Hawaii's Leprosy Victims and Development
of Kalaupapa Settlement, 1865 to Present, 1985. Denver: Branch of Planning, Alaska–
Pacific Northwest–Western Team, U.S. Department of Interior, National Park Ser-
vice, 1985.
Group 70, Davianna McGregor, and Cultural Surveys Hawai‘i. “Kalo Kanu o Ka ‘Äina:
A Cultural Landscape Study of Ke‘anae and Wailuanui, Island of Maui.” Technical
Report for the County of Maui Planning Department, 1995.
Halapua, Sitiveni. “Sustainable Development: From Ideal to Reality in the Pacific
Islands.” Paper prepared for the Fourth Pacific Islands Conference of Leaders, Tahiti,
French Polynesia, June 24–26, 1993, sponsored by the East-West Center, Honolulu.
Handy, E. S. Craighill. The Hawaiian Planter. Vol. 1, His Plants, Methods and Areas of Cul-
tivation. Bernice Pauahi Bishop Museum Bulletin 161. Honolulu: Bernice P. Bishop
Museum, 1940.
Handy, E. S. Craighill, and Elizabeth Green Handy, with Mary Kawena Pukui. Native
Planters in Old Hawaii: Their Life, Lore, and Environment. Honolulu: Bishop Museum
Press, 1972.
Handy, E. S. Craighill, and Mary Kawena Pukui. The Polynesian Family System in Ka-‘u,
Hawai‘i. Tokyo and Rutland, Vt.: Charles E. Tuttle, 1976.
Hawaiian Gazette. September 6, 1910.
Hawaii Ethnographic Notes. Bernice Pauahi Bishop Museum.
He Makana No Nä Kumu: A Gift to the Teachers of Hawaiian Students. Produced by Diane
Kahanu and Nä Maka O Ka ‘Äina. 1991.
Hirota, Sam O. “Environmental Impact Statement for the Kaupö Water System Improve-
ments, Kaupö, Island of Maui, Hawaii.” Unpublished report. Honolulu, 1983.
Holmes, Tommy. “A Preliminary Report on the Early History and Archaeology of the

356
bibliography

Puna Forest Reserve/Wao Kele ‘o Puna Natural Area Reserve.” Report prepared for
True/Mid-Pacific Geothermal, November 1985.
Hudson, Alfred E. “Archaeology of East Hawaii, 1930–32.” Unpublished manuscript.
Bernice Pauahi Bishop Museum Library.
Humphries, Grace. “Hawaiian Homesteading: A Chapter in the Economic Development
of Hawai‘i.” Master’s thesis, University of Hawai‘i at Mänoa, 1937.
Jarves, James Jackson. History of the Hawaiian Islands: Embracing Their Antiquities, Mythol-
ogy, Legends, Discovery by Europeans in the Sixteenth Century, Re-Discovery by Cook, With
Their Civil, Religious and Political History, From the Earliest Traditionary Period to the Year
1846. Honolulu: H. M. Whitney, 1872.
Johnson, Rubellite. Kumulipo: The Hawaiian Hymn of Creation. Honolulu: Topgallant,
1981.
Jones, Stella. “Field Notes on Waipio, 1931.” Unpublished manuscript. Bernice Pauahi
Bishop Museum Library.
Judd, Geritt. Pule O‘o: The Story of Molokai. Honolulu: Porter Printing, 1936.
Judd, Henry P. The Friend. September 1936.
Kahaulelio, A. D. “He Mau Kuhikuhi No Ka Lawai‘a Ana.” Nüpepa Kü‘oko‘a, February
28, 1902–July 4, 1902. Translated by Mary Kawena Pukui as “Fishing Lore.” Bernice
Pauahi Bishop Musuem Library; Hawaiian Collection, Hamilton Library, University
of Hawai‘i.
Kahiolo, G. W. He Moolelo No Kamapua‘a: The Story of Kamapua‘a. Translated by Esther
T. Mookini and Erin C. Neizmen with the assistance of David Tom. Hawaiian Stud-
ies Program, University of Hawai‘i at Mänoa, 1978.
Kaho‘olawe Island. http://www.kahoolawe.org/ (accessed December 30, 2005).
Kaho‘olawe Island Conveyance Commission. Kaho‘olawe: Restoring a Cultural Treasure.
Final Report of the Kaho‘olawe Island Conveyance Commission to the Congress of
the United States. March 31. Maui, 1993.
Kaho‘olawe, Nä Leo o Kanaloa: Chants and Stories of Kaho‘olawe. Foreword by Noa Emmett
Aluli. Honolulu: ‘Ai Pöhaku Press, 1995.
Kaläkaua, Hawaiian Majesty King David. The Legends and Myths of Hawaii: The Fables and
Folklore of a Strange People. Tokyo and Rutland, Vt.: Charles E. Tuttle, 1972.
Kalaniana‘ole, Prince Jonah Kühiö. “The Story of the Hawaiians.” Mid-Pacific Magazine
21, no. 2 (February 1921).
Kamakau, Samuel. Ka Po‘e Kahiko: The People of Old. Bernice Pauahi Bishop Museum
Special Publication 51. Honolulu: Bishop Museum Press, 1964.
———. Ruling Chiefs of Hawaii. Honolulu: Kamehameha Schools Press, 1961.
———. Tales and Traditions of the People of Old: Nä Mo‘olelo a ka Po‘e Kahiko. Edited by

357
bibliography

Dorothy B. Barrere. Translated by Mary Kawena Pukui. Honolulu: Bishop Museum


Press, 1991.
———. The Works of the People of Old. Bernice Pauahi Bishop Museum Special Publica-
tion 61. Honolulu: Bishop Museum Press, 1976.
Kanahele, Edward. Introduction to Van James, Ancient Sites of O‘ahu: A Guide to Hawai-
ian Archaeological Places of Interest. Honolulu: Bishop Museum Press, 1991.
Kanahele, George Hueu Stanford. Hawaiian Renaissance. Honolulu: Project Waiaha, 1982.
———. Ku Kanaka: Stand Tall; A Search for Hawaiian Values. Honolulu: University of
Hawai‘i Press , 1986.
Kanaka‘ole Kanahele, Pualani. “He Ko‘ihonua No Kanaloa, He Moku.” In Kaho‘olawe,
Nä Leo o Kanaloa: Chants and Stories of Kaho‘olawe, photographs by Wayne Levin, Row-
land B. Reeve, Franco Salmoiraghi, and David Ulrich. Honolulu: ‘Ai Pöhaku Press,
1995.
———. “Ka Honua Ola: The Living Earth.” Unpublished manuscript. Funded by the
Center for Hawaiian Studies at the University of Hawai‘i at Mänoa from the State
Foundation on Culture and the Arts, 1992.
Kanepu‘u. “Traveling About on Moloka‘i.” Ke Au Okoa, September 5 and 26, 1867.
Ka Nühou. May 23, 1873.
“Kaulana na Pua: Famous Are the Children.” KGMB news special by Bob Jones, 1987.
Keesing, Felix. Hawaiian Homesteading on Molokai. Research Publications 12. Honolulu:
University of Hawai‘i Press, 1936.
Kelly, Marion. “Cultural History of Pelekunu Valley, Moloka‘i.” Manuscript for the
Nature Conservancy. Honolulu, March 9, 1988.
———. Majestic Ka‘u: Mo‘olelo of Nine Ahupua‘a. Department of Anthropology Report
Series 80-2. Honolulu: Bishop Museum Press, 1990.
Kent, Noel. Hawaii: Islands under the Influence. New York: Monthly Review Press, 1983.
Kinney, Henry Walsworth. The Island of Hawaii. Hilo: Hilo Board of Trade, 1913.
Kirch, Patrick V. Feathered Gods and Fishhooks: An Introduction to Hawaiian Archaeology and
Prehistory. Honolulu: University of Hawai‘i Press, 1985.
———. Legacy of the Landscape: An Illustrated Guide to Hawaiian Archaeological Sites. Hono-
lulu: University of Hawai‘i Press, 1996.
Kroeber, A. L. Anthropology: Race, Language, Culture, Psychology, Prehistory. New York: Har-
court, Brace and World, 1948.
Kuykendall, Ralph. The Hawaiian Kingdom. Vol. 1, Foundation and Transformation, 1778–
1854. Honolulu: University of Hawai‘i Press, 1968.
Langlas, Charles. The People of Kalapana, 1832–1950. Hilo: The author, 1990.
Lennox, Colin. “A Report to the Trustees of the Bernice P. Bishop Museum on the
Resources of Waipio Valley, Island of Hawaii: Their Past and Present Uses and an

358
bibliography

Analysis of the Problems Facing Their Fuller Use in the Future.” Unpublished man-
uscript, 1954. Bernice Pauahi Bishop Museum Library.
Levy, Neil. “Native Hawaiian Land Rights.” California Law Review 63, no. 4 ( July 1975):
848–85.
Lewis, Henry, ed. Molokai Studies: Preliminary Research in Human Ecology. Honolulu:
Department of Anthropology, University of Hawai‘i , 1970.
———. North Kohala Studies: Preliminary Research in Human Ecology. Honolulu: Depart-
ment of Anthropology, University of Hawai‘i, 1970.
Lili‘uokalani. The Kumulipo: An Hawaiian Creation Myth. Kentfield, Calif.: Pueo Press,
1997.
Lind, Andrew. An Island Community: Ecological Succession in Hawaii. Chicago: University
of Chicago Press, 1938.
Linnekin, Jocelyn. Children of the Land: Exchange and Status in a Hawaiian Community.
New Brunswick, N.J.: Rutgers University Press, 1985.
Littler, Robert M. C. The Governance of Hawaii: A Study in Territorial Administration. Stan-
ford, Calif.: Stanford University Press, 1929.
Losch, Naomi Noelanioko‘olau Clarke, ed. Huaka‘i Mäkai‘ka‘i a Kaupö, Maui: A Visit to
Kaupö, Maui, by Thomas K. Maunupau. Translated by Mary Kawena Pukui and Mal-
colm Naea Chun. Honolulu: Bishop Museum Press, 1998.
Lueras, Leonard. On the Hana Coast: Being an Accounting of Adventures, Past and Present,
in a Land Where the Hand of Man Seems to Rest Lightly. Honolulu: Emphasis Interna-
tional, 1983.
Lyman, Chester S. Around the Horn to the Sandwich Islands and California, 1845–1850. New
Haven, Conn.: Yale University Press, 1924.
Lyons, Curtis J. “Land Matters in Hawai‘i.” The Islander 2. Honolulu, 1875. Reprint in
Hamilton Library, Hawaiian Collection, University of Hawai‘i at Mänoa.
Malo, David. “Causes for the Decrease of the Population in the Islands.” Translated with
comments by Lorrin Andrews. Hawaiian Spectator 2, no. 2 (1839).
———. Mo‘olelo Hawaii. Translated by Nathaniel B. Emerson as Hawaiian Antiquities.
Bernice Pauahi Bishop Museum Special Publication 2. Honolulu: Bishop Museum
Press, 1951 .
Matsuoka, Jon, Davianna McGregor, and Luciano Minerbi. “Governor’s Moloka‘i Subsis-
tence Task Force.” Final report, June 1994. Moloka‘i: Department of Business, Eco-
nomic Development, and Tourism. Honolulu: The Department, 1994.
Matsuoka, Jon K., Davianna Pomaika‘i McGregor, Luciano Minerbi, Pualani Kanahele,
Marion Kelly, and Noenoe Barney-Campbell. “Native Hawaiian Ethnographic Study
for the Hawai‘i Geothermal Project Proposed for Puna and Southeast Maui.” Tech-
nical report, Oak Ridge National Laboratory, Lockheed Martin, May 1996.

359
bibliography

Maui Visitors Bureau. “Hana Visitors Guide.” 1998.


Maunupau, Thomas K. Huakai Makaikai a Kaupö, Maui: A Visit to Kaupö, Maui, as pub-
lished in Nüpepa Kü‘oko‘a, June 1, 1922–March 15, 1923. Translated by Mary Kawena
Pukui. Hawai‘i Ethnographic Notes, Newspapers 1922–1924. Bernice Pauahi Bishop
Museum.
McGregor, Davianna. Interviews, November 1–4, 1994. Author’s collection.
———.“Küpa‘a I Ka ‘Äina: Persistence on the Land.” Ph.D. dissertation, University of
Hawai‘i at Mänoa, 1989.
———. “Voices of Today Echo Voices of the Past.” In Mälama: Hawaiian Land and Water,
edited by Dana Naone Hall. Honolulu: Bamboo Ridge Press, 1985.
McGregor-Alegado, Davianna. “Hawaiians: Organizing in the 1970s.” Amerasia 7, no. 2
(1980).
Moloka‘i Enterprise Community. “Moloka‘i Rural Empowerment Zone Application Sub-
mitted to the United States Department of Agriculture.” October 9, 1998.
Morales, Rodney. Ho‘iHo‘i Hou: A Tribute to George Helm & Kimo Mitchell. Honolulu:
Bamboo Ridge Press, 1984.
“Nä Kumukanawai o Ka Makahiki 1839 A Me Ka 1840.” Ka Ho‘oilina: The Legacy; Puke
Pai ‘ölelo Hawai‘i, Journal of Hawaiian Language Sources 1, no. 1 ( March 2002).
Napoka, Nathan. “Kahoolawe Place Names.” In Carol Silva, Kahoolawe Cultural Study,
Part 1: Historical Documentation. Prepared for the Pacific Division, Naval Facilities
Engineering Command, U.S. Navy, Pearl Harbor, Honolulu, April 1983.
Nüpepa Kü‘oko‘a. 1918–1921.
Office of the Commissioner of Public Lands. Indices of Awards Made by the Board of Com-
missioners to Quiet Land Titles in the Hawaiian Islands. Honolulu: Star-Bulletin Press,
1929.
Pacific Commercial Advertiser. 1900–1930.
Paradise of the Pacific. 1900–1958.
PBR-Hawai‘i with Pualani Kanahele, Leslie Kuloloio, Davianna McGregor, Rowland
Reeve, Hardy Spoehr, Community Planning, and Geographic Decision Systems Inter-
national. Palapala Ho‘onohonoho Moku‘äina O Kaho‘olawe: Kaho‘olawe Use Plan. Kahului:
Kaho‘olawe Island Reserve Commission, 1995.
Perrells, Ann. “Environmental Resources and Neighborhood Food Exchange.” In Molo-
kai Studies: Preliminary Research in Human Ecology, ed. Henry Lewis. Honolulu: Depart-
ment of Anthropology, University of Hawai‘i, 1970.
Petersen, Glenn. “Taro Farming in Waipio Valley on the Island of Hawaii.” In North
Kohala Studies: Preliminary Research in Human Ecology, ed. R. Warwick Armstrong and
Henry T. Lewis. Honolulu: University of Hawai‘i Press, 1970.

360
bibliography

Phelps, Southwick. “Regional Study of Molokai, Hawaii.” Unpublished manuscript, 1941.


Bernice Pauahi Bishop Museum Library.
Pukui, Mary. Translation of Nüpepa Kü‘oko‘a, June 1, 1922. Hawai‘i Ethnographic Notes,
Newpapers 1922–1924. Bernice Pauhahi Bishop Museum Library.
Pukui, Mary Kawena. Chant H-41 c Webcor timer 753 Sel. 11. Wire recording. Tran-
scribed, translated, and performed by Mary Kawena Pukui. Hawaiian Ethnographic
Notes. Bishop Museum Library.
——. Interviews, 1960–1968. Audio-Recording Collection, Department of Anthropol-
ogy, Bernice Pauahi Bishop Museum.
Pukui, Mary Kawena, ed. ‘Ölelo No‘eau: Hawaiian Proverbs and Poetical Sayings. Bernice
Pauahi Bishop Museum Special Publication 71. Honolulu: Bishop Museum Press,
1983.
Pukui, Mary Kawena, and C. Curtis. Pikoi and Other Legends of the Island of Hawai‘i.
Reprint, Honolulu: Kamehameha Schools Press, 1949.
Pukui, Mary Kawena, and Samuel H. Elbert. Hawaiian Dictionary. Honolulu: University
of Hawai‘i Press, 1975.
Pukui, Mary Kawena, Samuel H. Elbert, and Esther T. Mookini. Place Names of Hawaii.
Honolulu: University of Hawai‘i Press, 1984.
Pukui, Mary Kawena, and Laura Green. “Nä Ali‘i Ho‘oluhi o Ka‘ü, or The Despotic Chiefs
of Ka‘ü.” In Folktales of Hawai‘i: He Mau Ka‘ao Hawai‘i. Honolulu: Bishop Museum
Press, 1995.
Pukui, Mary Kawena, and E. S. Craighill Handy. The Polynesian Family System in Ka-‘u,
Hawai‘i. Tokyo and Rutland, Vt.: Charles E. Tuttle, 1976.
Pukui, Mary Kawena, and Alfons Korns, The Echo of Our Song. Honolulu: University of
Hawai‘i Press, 1973.
Pukui, Mary Kawena, ed. ‘Ölelo No‘eau: Hawaiian Proverbs and Poetical Sayings. Bernice
Pauahi Bishop Museum Special Publication 71. Honolulu: Bishop Museum Press,
1983.
Puna Hui ‘Ohana. “Assessment of Geothermal Impact on Aboriginal Hawaiians.” Wash-
ington, D.C.: Department of Energy, 1982.
Ralston, Caroline. “Hawaii, 1778–1854: Some Aspects of Maka‘äinana Response to Rapid
Cultural Change.” Journal of Pacific History 29, no. 1 ( January 1984).
Remy, Jules. Contributions of a Venerable Savage to the Ancient History of the Hawaiian
Islands. Translated by William T. Brigham. Boston: Press of A. A. Kingman, 1868.
Reyes, B. J. “With Little Fanfare, Kahoolawe Island Returns to Hawaiian Control.”
Associated Press. November 12, 2003. http://zwire.com /site /news.cfm?newsid=
10504278&BRD=1817&PAG=461&dept_id=222077&rfi=6 (accessed December 30,
2005).

361
bibliography

Riley, Thomas. “Wet and Dry in a Hawaiian Valley: The Archaeology of an Agricultural
System.” Ph.D. dissertation, University of Hawai‘i at Mänoa, 1973.
Rose, Roger. Reconciling the Past: Two Basketry, Ka‘ai, and the Legendary Liloa and Lonoika-
makahiki. Honolulu: Bishop Museum Press, 1992.
Sahlins, Marshall. Historical Metaphors and Mythical Realities: Structure in the Early History
of the Sandwich Islands Kingdom. Ann Arbor: University of Michigan Press, 1981.
Schmitt, Robert. Demographic Statistics of Hawaii, 1778–1965. Honolulu: University of
Hawai‘i Press, 1968.
———. The Missionary Census of Hawai‘i. Pacific Anthropological Records 20. Honolulu:
Bishop Museum Press, 1973.
Shepard, F. P., G. A. MacDonald, and D. C. Cox. The Tsunami of April 1, 1946. Berkeley
and Los Angeles: University of California Press, 1950.
Silva, Carro. Kahoolawe Cultural Study, Part 1: Historical Documentation. Prepared for the
Pacific Division, Naval Facilities Engineering Command, U.S. Navy, Pearl Harbor,
Honolulu, April 1983.
Skolmen, Roger. “Hawaii’s Forest Products Industry.” Paper presented at the 18th Annual
Hawaii Forestry Conference, November 18–19, 1976, Honolulu.
Stannard, David. Before the Horror. Honolulu: University of Hawai‘i Press, 1989.
Stokes, John, F. J. Heiau of the Island of Hawai‘i: A Historic Survey of Native Hawaiian Tem-
ple Sites. Honolulu: Bishop Museum Press, 1991.
Summers, Catherine. Molokai: A Site Survey. Pacific Anthropological Records 14. Hono-
lulu: Bishop Museum Press, 1971.
Thrum, Thomas G. Hawaiian Folk Tales. Chicago: McClurg, 1921.
———. More Hawaiian Folk Tales. Chicago: McClurg, 1923.
———. Thrum’s Hawaiian Almanac and Annual. Honolulu: Honolulu Star-Bulletin, 1895.
Trust for Public Land with Bay Pacific Consulting. “East Maui Resource Inventory.”
Honolulu: Rivers, Trails, and Conservation Assistance Program, National Park Ser-
vice, 1998.
Uemoto, Shuzo. Nana I Na Loea Hula: Look to the Hula Resources. With narratives by Hula
Resources. Honolulu: Kalihi-Palama Culture and Arts Society, 1997.
University of Hawai‘i at Mänoa. Department of Urban and Regional Planning. “Waipi‘o
Valley: Towards Community Planning and Ahupua‘a Management.” Technical report.
Fall 1999.
Walker, W. M. “Archaeology of Maui.” Unpublished manuscript, 1931. Bernice Pauahi
Bishop Museum.
Westervelt, W. D. Legends of Ma-ui. Reprint, New York: AMS Press, 1979.
Wilkes, Charles. Narrative of the United States Exploring Expedition During the Years 1838,
1839, 1840, 1841, 1842. 5 vols. Philadelphia, 1849.

362
bibliography

Willis, Koko, and Pali Jae Lee. Tales from the Night Rainbow. Honolulu: Night Rainbow
Publishing, Native Books, 1990.
Yamamura, Douglas. “A Study of Some of the Factors in the Education of the Child of
Hawaiian Ancestry in Hana, Maui.” Master’s thesis, University of Hawai‘i at Mänoa,
1941.

Government Documents

Bush, George. Memorandum on the Kaho‘olawe, Hawaii, Weapons Range. October 22,
1990. http:// bushlibrary.tamu.edu /research/papers/1990/90102203.html (accessed
December 30, 2005).
County of Hawai‘i. “The General Plan Hawai‘i County.” November 1989.
Hawaiian Kingdom. Board of Health. Hansen’s Disease in Hawaii. 1866.
Hawai‘i. Department of Hawaiian Home Lands. Kalama‘ula Development Plan. Honolulu:
DHHL, 1983.
Hawai‘i. Revised Statutes, Chapter 6K.
Hawai‘i. State Archives. Boundary Commission Hawai‘i, 3rd & 4th Circuits.
———. Foreign Office and Executive File. 1841–1850.
———. Governors’ Proclamations. 1911.
———. Governors’ Proclamations. 1926–1930.
———. Interior Department Letters. 1850–1860.
———. Land Matters. Public Land Petitions File. 1845–1855.
———. Laws. 1864–1865.
———. Legislative Petitions File. 1841–1850.
Hawaii (Territory). Governor. Governors’ Proclamations. 1926–1930.
———. Report of the Governor’s Advisory Committee on Leprosy in Hawai‘i. October 1930.
———. Report of the Hawaiian Homes Commission to the Legislature of Hawaii, Regular Ses-
sion. 1923–1931.
Hawaii (Territory). Office of the Commissioner of Public Lands. Indices of Awards Made
by the Board of Commissioners to Quiet Land Titles in the Hawaiian Islands. Honolulu: Star-
Bulletin Press, 1929.
Kaho‘olawe Island Conveyance Commission. Kaho‘olawe: Restoring a Cultural Treasure.
Final report of the Kaho‘olawe Island Conveyance Commission to the Congress of
the United States, March 31, 1993, Maui.
Maui Visitors Bureau. “Hana Visitors Guide.” 1998.
Pele Defense Fund v. Estate of James Campbell (Civil no. 89-089).
Republic of Hawaii. Report of the General Superintendent of the Census, 1896. Honolulu:
Hawaiian Star Press, 1897.

363
bibliography

Rice v. Cayetano 528 US 495 (2000) Docket no. 98-818.


State of Hawaii. Department of Business and Economic Development and Tourism. State
of Hawai‘i Counties and Districts, 1991.
———. University of Hawaii Land Study Bureau. “Preliminary Survey of Resource
Development and Rehabilitaiton Opportunites in Waipio Valley, Island of Hawaii.”
Special Study Series. L.S.B. Report no. 1. January 1960.
U.S. Bureau of the Census. Fifteenth Census of the United States, 1930: Population, Second
Series, Hawai‘i, Composition and Characteristics of the Population and Unemployment.
Washington, D.C.: Government Printing Office, 1931.
———. Fifteenth Census of the United States, 1930: Occupation Statistics, Hawai‘i. Washing-
ton, D.C.: Government Printing Office, 1932.
U.S. Congress. Act of June 20, 1938 (52 Stat. 781 et seq.).
———. Congressional Record. 107th Congress, 1st sess., April 6, 2001.
———. House. Committee on the Territories. A Bill to Amend the Hawaiian Homes Com-
mission Act, 1920, approved July 9, 1921, As Amended by Act of February 3, 1923, H.R.
6989, 70th Congress, 1st sess.
———. House. Report No. 243, “Intervention of United States Government in Affairs of
Foreign Friendly Governments.” 53rd Congress, 2d sess., December 21, 1893. Wash-
ington, D.C.: Government Printing Office, 1893.
———. Senate. Committee on Foreign Relations. “Hawaiian Islands.” In Report of the
Committee on Foreign Relations with Accompanying Testimony and Executive Documents
Transmitted to Congress from January 1, 1893 to March 19, 1894. Vols. 1–2. Washington,
D.C.: Government Printing Office, 1894.
———. Senate. Committee on Foreign Relations. Report No. 227, “Report from the
Committee on Foreign Relations and Appendix in Relation to the Hawaiian Islands,
February 26, 1894.” 53d Congress, 2d sess. Washington, D.C.: Government Printing
Office, 1894.
U.S. Department of Interior, 1993–1998. “Federal Indian Policies.” June 12, 1998.
U.S. Department of Interior and Department of Justice. “From Mauka to Makai: The
River of Justice Must Flow Freely; Report on the Reconciliation Process between the
Federal Government and Native Hawaiians.” Washington, D.C., October 23, 2000.
———. “1993–1998 Federal Indian Policies.” June 12, 1998.
U.S. Department of State. Papers Relating to the Mission of James H. Blount, United States
Commissioner to the Hawaiian Islands. Washington, D.C.: Government Printing Office,
1893.
U.S. Supreme Court, no. 98-818, February 23, 2000.

364
 index 

adzes, 24, 25, 27; of Maui, 24 Captain Cook, 30, 98, 196, 331n. 51, 337n. 34
Ahupü, 263, 266 celebrations, family and holiday, 77, 171, 238,
ahupua‘a, 20–21, 27, 35–36, 56, 59, 81–85, 93, 240
106, 127–128, 153–154, 157, 282, 288, chants (oli), 8, 13–15, 18, 22–23, 25, 31, 82,
290–291, 293, 320n. 16, 321n. 24, 322n. 277, 283, 340n. 24; creation, 13, 264;
28, 328n. 97, 335n. 144, 338nn. 49, 50, Häna, 83, 96; Kaho‘olawe, 253–257, 260,
341n. 39; awarded in Mähele, 101–102, 276, 285, 320n. 13, 335n. 6, 351n. 51,
159–161, 163, 165–166, 173, 180, 187, 352n. 10; leprosy/ Kalawao, 202; Maka-
190; description, 20, 26–27, 282; of Häna, hiki, 270–271, 273; Moloka‘i, 192, 223;
112, 115, 118, 121 Pele, 6; Puna, 144, 145–147, 152, 183,
‘Ai Noa, 31, 272 293–294, 319n. 7
Akua, 8, 13, 23–24, 31, 76, 191, 254–255, 260, childhood activities, 45, 292; Häna, 107–108,
272–273, 286; aloha ke akua, 2. See also 110, 116, 120, 125, 127, 141–142, 332n.
Kanaloa 76; Kalawao nursery, 224–225; Moloka‘i,
ali‘i, 25–30, 32, 37, 53, 95, 99, 101, 151, 194, 214–215, 222, 223, 230, 232, 234, 237,
197–198, 308, 325n. 28 240–241; Puna, 163, 171–172, 184;
aloha, 83, 86, 187, 220, 226, 268–270, 273, Waipi‘o, 62–66, 71–72, 77
283, 286, 299, 304; aloha ‘äina, 2, 20, 193, Chinese, 9, 42; Häna, 102–105, 112–114, 117,
219, 250, 255, 264–265, 268, 277, 281, 140, 142; Moloka‘i, 218; Puna, 169, 172;
283; aloha ‘äina, aloha ke akua, aloha Waipi‘o, 59–67, 69, 71, 73–74, 77–78
kekähi i kekähi, 2; Hui Aloha ‘Äina, 166 churches: Häna, 100, 106–107, 116; Kalawao,
Aluli, Noa Emmett, 252, 268, 270–271, 296, 224; Moloka‘i, 240; Puna, 158, 172–173;
304, 319n. 4, 349n. 35, 352n. 10 Waipi‘o, 60, 74
American Indian Religious Freedom Act, 265 coffee, 40; Häna, 124; Moloka‘i, 201, 216;
Ana, Keoni, 101 Puna, 159, 166–168, 170, 182; Waipi‘o,
ancestral memory knowledge, 5–6, 15, 18, 59, 64, 71, 79
127, 249, 255, 264, 283, 289, 295, 352n. 1 colonization period, 23
archaeology, 23, 320n. 17, 324n. 1, 325nn. 26, Constitution, 1840, 31–32, 35, 201; Bayone,
30, 329nn. 5, 16, 19, 22, 330nn. 32, 46, 42; U.S., 279–280
336nn. 10, 12, 21, 337n. 40, 343n. 67 Crown lands, 37–39, 42, 227, 278; Häna, 138;
‘aumakua, 6, 13–15, 17, 21, 23, 31, 61, 75, 87, Moloka‘i, 198; Puna, 159, 161, 166
88, 109, 187, 221, 292 cultivation, 12, 16, 21, 24–25, 269, 278, 283,
290–291, 301, 330n. 26; Häna, 91,
Bill of Rights, 1839, 31–32 110–111, 113, 115–116, 120, 127, 129,
boat transportation, 104–105, 114, 122–123, 137–138; Moloka‘i, 200–201, 210,
128, 177, 206, 208, 211, 217, 219, 226, 231–232, 245–246; Puna, 167, 168,
252, 302 185–186; Waipi‘o, 52, 59, 60, 68, 81–82

365
index

cultural kïpuka, 4, 12, 15, 17, 19–22, 46, 48, 134, 138; types of fish, 109–111, 133;
62, 82, 142, 178, 187, 248, 250, 281, Waipi‘o, 51, 55, 59, 63–66, 70, 73, 75–76,
286–288, 290, 293–294, 296, 319n. 1, 79–80
320n. 16; definition/features, 8; locations, fishponds, 23, 25, 28, 45, 54, 260, 282, 290,
12 320n. 16, 341n. 36; Häna, 87, 89, 96, 103,
134; Moloka‘i, 193–194, 196, 199–200,
Damien, Father Kalawao, 203, 224, 340n. 27; 208, 210, 221–222, 243–245; Puna, 161;
in Puna, 158, 172 Waipi‘o, 59
Davis, Isaac, 30, 55, 159, 161; Kale, 159 floods, 17; Häna, 94, 109, 117; Kaho‘olawe,
developmental period, 23–24, 29; Häna, 256; Puna, 178; volcanic, 149; Waipi‘o,
90–94; Puna, 147–149 50–51, 61, 70, 78–81
diseases, 10, 29, 30, 100, 141, 163, 196, foreigners’ rights, 31–32, 38–39, 197; land
197–198, 339n. 12; Hansen’s disease, 10, awards, 102, 198
201–205, 207, 223–227, 340nn. 20, 22, 25; frogs: Puko‘o, 206; Waipi‘o, 63
plant, 179, 228, 239, 264
government, 10, 16, 31–33, 39, 41, 42, 47, 97,
East End Policy Statement, 243 196–197, 225, 279, 284, 293–294, 298,
East Maui Irrigation Company, 112, 114, 130, 304, 308, 350n. 41; Hawaiians in, 44, 46
137–139, 141 government land, 20, 34, 36–38, 42, 101,
Edith Kanaka‘ole Foundation, 82, 270, 272, 278, 337n. 43; Moloka‘i, 198, 202, 203,
276, 281, 288, 349nn. 31, 33, 34; Aunty 227–228, 235; Puna, 159, 161, 166–167,
Edith Kanaka‘ole, 253, 270, 272 173, 305; Waipi‘o, 73, 79
Ellis, Reverend William Puna, 155–157; government roads: Häna, 117; maintenance
Waipi‘o, 55–56, 325n. 22, 336n. 14, crews, 9–10, 117; prison labor, 117; Puna,
337n. 37 164, 176; Waipi‘o, 73

Fagan, Mr. and Mrs. Paul, 125, 329n. 17 Hakau, 53, 152
ferns, 7, 71, 92, 95, 110, 112, 133, 161, 180, Hakioawa, 3, 261, 264, 266, 271, 273, 303
187, 216; pulu industry, 40, 161, 201 Hälau o Kekuhi, 272
fire clan, 144, 147–148; Waipi‘o, 64, 70, 72; Hälawa, Moloka‘i, 198–199, 200–203, 206,
with wind, 85 212, 217–218, 221, 223, 242, 257, 342n.
fishing, 8, 9, 12, 22, 30, 40–41, 44–45, 52–53, 46, 344n. 92
253–254, 341n. 38, 344n. 90, 351n. 50; Hale o Lïloa, 53–56, 287
Häna, 85, 88–91, 93, 95, 103–105, Haleakalä, 84, 86–87, 90, 109, 120–121,
108–110, 112, 115–116, 118–121, 124, 126–128, 267, 272, 288, 335n. 144
127–128, 133–135; of the islands, 24, 87; Häloa, 13
Kaho‘olawe, 260–264, 266, 268–270, 281, Häna, 2, 8, 9–12, 20–21, 51, 54–55, 83–117,
303; methods, 12–19, 70–71, 76, 93, 288–292
111–112, 124, 134, 170–171, 278, 282, hanai, 15, 62; Häna, 101, 106; Moloka‘i, 225;
290; Moloka‘i, 27, 193, 196–197, 203, Puna, 163, 171
205–208, 210, 212, 214–217, 219–220, Handy, E. S. Craighill, 12, 30, 62, 114, 117,
222–223, 225, 227, 241, 243–244, 119, 121, 170, 210, 218, 294, 320nn. 10,
246–247, 297–301; Puna, 15, 157, 12, 321nn. 21, 25, 322n. 33, 325nn. 14,
160–161, 163–164, 170–172, 175–177, 27, 35, 329n. 9, 330nn. 24, 26, 30, 36, 44,
180–184, 186, 296; spawning, 87, 94, 333nn. 100, 108, 111, 334nn. 116, 120,

366
index

121, 125, 132, 336n. 17, 338n. 57, 341n. Imakakoloa, 154
41, 342n. 43, 343n. 69 impact: cultural and social, 22, 43, 46, 57–58,
Hawaiian Homelands, 204, 207, 216, 227–244, 100, 125–126, 138, 173, 181, 188,
296–298; homesteading, 160, 166, 345nn. 196–197, 245, 275, 283, 288, 291–295,
102, 104, 109, 346nn. 115, 116, 121, 124, 297–298, 334n. 126, 339n. 70
125, 129, 135, 347nn. 136, 142, 144, 147, indigenous people, 23–24, 33, 42, 52, 272,
149, 151 279–280, 284, 350n. 43, 351n. 1
Hawaiian language, 24, 38, 43, 159, 262, 322n. irrigation, 25, 28, 43, 50, 283, 290–291;
39; punished for speaking, 223, 344n. 91; Häna, 103, 112, 114, 130–131, 136–138;
revival, 249, 278 Moloka‘i, 228; Waipi‘o, 52, 59–61, 77, 82
healing plants, 14–15, 31, 71, 76, 108, 132,
180, 185, 208, 211 Japanese submarine, 124
heiau, 25, 52–53, 90, 97, 99, 152, 154, 169, Judd Commission on Hawaiian Homelands,
221, 257–258, 261, 264, 291, 303, 325nn. 1930, 239–240
17, 20; Paka‘alana, 52–53, 55–56; Waha- Judd Committee on Hansen’s disease, 225–226
‘ula, 150–151, 153, 168–169, 336n. 17.
See also Pi‘ilani, High Chief of Maui Ka ‘Ohana O Kalae, 270, 351n. 50
Helm, George, 249–250, 252, 265, 268, 271, Ka‘ahumanu, High Chiefess Regent, 31, 56,
276, 296 101–102, 287
hïhïwai, 109–110, 129–130, 132, 211, 215–217 ka‘ai, 51, 53, 56
Hi‘iaka, 145–146, 272 Ka‘eokulani, High Chief of Kaua‘i, 54–55, 99,
ho‘ailona, 17, 282, 292 195–196
Höküle‘a, 258, 277, 348n. 13 Kahaulelio, 208, 262–263, 341n. 38, 349n. 24
honey, 205 Kahawali, 149–150
Honokanai‘a, 263, 303 Kahekili, High Chief of Maui, 54–55, 97–99,
Honukanaenae, 263 102, 193, 195, 264, 321n. 22
Ho‘ohokukalani, 13, 256 Kahikinui, 10, 12, 23, 83, 85, 105, 259–260,
ho‘okupu, 28, 183, 220, 236, 271–273, 302 288
Hui Ala Loa, 268–269, 277, 296 Kaho‘olawe, 1–4, 14, 20, 22–23, 27, 30, 38, 48,
Hui Ala Nui o Makena, 269 90, 126, 142, 194, 196, 249–285, 4n. 319,
Hui Aloha ‘Aina, 166 13n. 320, 22n. 321; Kaho‘olawe Island
Hui Mälama o Mo‘omomi, 298 Conveyance Commission, 275–276,
Hui o Kuapä, 245 348nn. 2, 3, 13, 349nn. 28, 31, 35, 36, 37,
hula, 31, 89, 143, 145–146, 149–150, 180, 202, 351nn. 49, 51, 352n. 10; Kaho‘olawe Oli
221, 266, 271–273, 277, 293, 349n. 32 Kühohonu o Kaho‘olawe, 13–14, 254, 260;
Hulihia chants, 6–7; Kua Loloa Kea‘au i Ka Kaho‘olawe Reserve Commission vision,
Nähelehele, 6–7 281, 301–302
hunting, 8, 16, 18, 103, 105, 108, 111, 115, kahuna, 25, 51, 61, 75, 94, 151, 154, 191–193,
128–130, 135, 137, 160, 163, 180–185, 221, 254, 258–259, 271; anä‘anä, 75; lä‘au
188, 208, 210, 223, 243, 245–247, lapa‘au, 31, 108
290–291, 296–298, 301 Kalaeoka‘ilio, battle of, 98
Kalaipahoa, 193
I.L.W.U. (International Longshoremen and Kalama, Queen Hakaleleponi (Hazalelepono),
Warehousemen’s Union), 125 56, 59, 159
‘Imaikalani, Chief of Ka‘ü, 152, 153–154 Kalama‘ula, 192, 199, 207, 228, 229–232, 235,

367
index

242, 244, 341n. 34, 342n. 50, 343n. 72, Kanaloa, 13–14, 23, 51, 96, 260–262; chief,
344n. 92, 347n. 158 96; fishpond, 23, 87, 260–264; Kaho‘o-
Kalaniana‘ole, Prince Jonah Kühi‘ö, 38, 227, lawe, 249, 252, 253–277, 281, 283–286,
230, 233, 236, 239, 241, 323n. 49 301–304, 320n. 13, 349nn. 31, 34, 351nn.
Kalaniopu‘u, High Chief, 30, 98–99, 154, 264, 48, 51, 352n. 10; springs, 23, 51, 87, 139
321n. 22 Kanapou, 257, 263, 303
Kalapana, 10–11, 156, 158, 163–164, 167, Käne, 88, 221, 264
169–173, 176–180, 186; chief, 96; Kala- Kanehoa, James Young, 101
pana rights, 173, 175–176, 294, 336n. 10, Känehunamoku, 88
337nn. 39, 43, 44, 338nn. 58, 59, 60, 62, Kaoanaeha, Mary, 56–57
346n. 126 kapu, 13, 55, 151, 210, 223, 255, 272; conser-
Kalawao-Kalaupapa, 10–11, 199, 201–204, vation practices, 14–16, 182
207, 212, 215, 217–219, 223–227, 242, Ka‘ü Puna Mäkaha Kümäkaha, 152–153
340nn. 20, 25, 27 Ka‘uiki, 83, 87–88, 95–99, 102
Kamakau, Samuel, 23, 53–54, 95, 100, Kaupö, 10, 11, 20, 23, 83–85, 87, 90, 92, 94,
150–154, 160, 259–260, 294, 321nn. 17, 97–98, 100–102, 105–106, 110–112,
19, 20, 322nn. 28, 36, 38, 325nn. 14, 15, 119–123, 126, 128, 134, 260, 288, 289,
24, 330nn. 35, 39, 45, 48, 331nn. 52, 53, 328n. 4, 329n. 23, 333n. 96, 334n. 126
54, 60, 336nn. 16, 19, 337nn. 26, 31, Kauwa, Julia Alapai, 101
339n. 2, 348n. 16, 349n. 18 Kea‘au, 6, 142, 153, 156–157, 159, 163–164,
Kamämalu, Victoria, 101, 159, 198 167, 179–180, 184, 338n. 49
Kamapua‘a, 148, 336n. 11 Kealaakapüpü, 194
Kamehameha I, 26, 29, 30–33, 52–54, 57, Kealaikahiki, 258–260, 262–263, 267–268, 303
97–99, 139, 141, 151, 155, 159, 164, 193, Ke‘anae-Wailuanui, 2–4, 20–21, 91, 97,
195–196, 198, 255 112–113, 115, 125, 127–142, 288–291,
Kamehameha II, Liholiho, 31, 56, 97, 151, 333n. 110, 335nn. 144, 160
195, 198, 308–309 Keawema‘uhili, 54, 155
Kamehameha III, Kauikeaouli, 32–33, 35, Ke‘elikolani, Princess Ruth, 56, 159, 161
37–38, 56, 97, 159, 195, 197, 308 Kekauonohi, Miriam, 101, 159, 198
Kamehameha IV, Alexander Liholiho, 56, 159 Kekuhaupi‘o, 98–99, 325nn. 19, 21, 331nn.
Kamehameha V, Lot, 56, 159, 201, 232, 240 50, 52, 54, 55, 339n. 2
Kamehameha Schools, 227, 308–309, 324, Keli‘imaika‘i, High Chief, 57, 99, 152–153
324n. 11, 325nn. 19, 21, 331nn. 50, 52, Keohokalole, 101, 159
54, 55, 339n. 2 Ke‘öpüolani, 56, 195
Kamehameha‘ailü‘au, 220, 223 Keouaküahu‘ula, 54, 99, 155, 195
Kamehamehanui, 97–98 Kepüwaha‘ula, 54, 57, 99, 195
Kamohoali‘i, 96, 257, 274 Kihaapi‘ilani, 25, 96–97, 194
Kanahele, Edward, 5–6, 270, 273, 284, 291, Kihapü, 51, 53
352n. 5 Kila, 52–53
Kanahele, Pualani Kanaka‘ole, 143, 145–146, Kïlauea, 156, 158, 168–170, 178, 270, 336n.
273, 285, 293, 319n. 7, 322n. 37, 335nn. 5, 12
6, 351nn. 49, 51 kioea, 221
Kana‘ina, High Chief Charles, 159, 165 Kïpahulu, 10, 11, 83, 85–90, 92, 94, 98–102,
kanaka, 27, 29, 31–32, 37, 40; ‘öiwi, 160 105–106, 111, 117–121, 126, 140, 288–289,
Kanaka‘ole, Edith, 253, 270, 272 329nn. 12, 13, 15, 332nn. 91, 94

368
index

kïpuka, 1, 4, 7–9, 12, 15–17, 19–22, 46, 48, local people, 44–45, 47–48, 126, 249, 269, 278,
250, 281, 286–288, 290, 293–294, 296, 284–285, 288, 320n. 16
319n. 1, 320n. 16, 341n. 39; Häna, 142; lökähi, 12–17
Moloka‘i, 248; Puna, 178, 187 Lonoikamakahiki, 51, 56, 287, 324n. 8
Kïwala‘ö, 154–155, 195 lumber, 113, 238; export, 167–169
ko‘a, 89–90, 112, 246, 261, 263–264, 282, Lunalilo, William Charles, 102, 159, 165,
351n. 50 198, 203
Kohemälamalama o Kanaloa, 249, 253–255,
257, 274; Protect Kaho‘olawe Fund, 268 Machado, Colette, 268, 270, 296
kökua, 248 Mahealani, 76
konohiki, 26–28, 30, 33, 34–37, 39, 52, 101, Mähele, 33–39; Häna, 100–101, 323nn. 45, 47,
198 325n. 28, 331n. 62, 337n. 43; Moloka‘i,
kua‘äina, 1, 2–6, 8–9, 12, 19–22, 35, 43–44, 197–199, 318; Puna, 158–166; Waipi‘o,
48, 249–250, 252, 266, 268, 270, 281, 283, 56–57. See also kuleana; land tenure
286–287, 289–290, 292, 296–298, 300, maka‘äinana, 25–28, 31, 38, 101, 151,
302; definition, 2–4; Häna, 85, 93, 103, 198–199, 305, 308, 321n. 23
110–111, 113–114, 117–118, 121, Makahiki, 150, 266–267, 270, 272–274, 286,
126–129, 131, 139–142; Moloka‘i, 191, 302–303, 322n. 39. See also Lonoikamaka-
197, 208–212, 243, 248; practices, 12, hiki
14–18, 19; Puna, 143–144, 160, 163, 165, mana, 6, 23, 144, 187, 255, 260, 271, 274, 285;
169–170, 172, 174–176, 180–188; Waipi‘o, taro, 218
50, 55–59, 62, 65, 78–80, 82 Matsuoka, Jon, 182, 247, 292, 293, 295, 297,
Kühe‘eia, 266 329n. 22, 339n. 71, 347n. 160
kuleana (ancestral lands), 15, 20, 269; Häna, Maui, demigod, 24, 51, 87
100–101, 119; Kuleana Act, 37–39, 56; menehune, 90, 94, 109, 221
Moloka‘i, 198–199, 203, 207–208, 210, migrations, 24, 150–155, 178, 268, 320n. 17,
318; Puna, 158–160, 305; rights, 33–34, 324n. 11
37; tax, 40; Waipi‘o, 56, 63–65, 74, 79, 81 military, 26, 29–30, 42–45, 78, 101, 125, 159,
Kumulipo, 13, 320n. 11 176, 178, 202–203, 249–250, 253, 263,
Kü‘ula, 14, 75, 89–90, 170, 222, 261–262; 265–266, 269, 275–277, 283, 302
Hina, 24 Minerbi, Luciano, 247, 287, 292, 297, 347n.
160
La‘amaikahiki, 25, 52, 259 missionaries, 8, 31, 41, 56, 74, 100, 155–158,
Lahilahi, Gina, 159 197, 200–201, 294
land tenure, 26–27, 29, 33, 65. See also Mitchell, Harry Künihi, cover, 1–4, 20, 83,
kuleana; Mähele 104, 108–109, 113, 253–254, 260,
Lanikaula, 192, 221 266–267, 286, 288, 301, 330n. 25, 331nn.
laulima, 28, 114, 268 66, 71, 74, 332nn. 76, 79, 83, 88, 90, 92,
Leleiohoku, 56, 159, 198 333nn. 95, 97, 102, 104, 349n. 27
leprosy (Hansen’s disease), 10, 163, 201–204, Moa‘ulaiki, 257–258, 303
207, 223–226, 340nn. 20, 22, 25, 345n. 97 Mo‘ikeha, 25, 51–52, 151, 259
livelihoods, 12, 19, 22, 40, 43, 270, 278, 294, Moloka‘i, 5, 8–12, 16, 19–21, 23–25, 27, 30,
300; Häna, 85, 115, 125, 130; Moloka‘i, 51–52, 88–90, 98, 114, 141, 158, 164,
196, 212, 214, 216, 243; Puna, 144, 163, 191–248, 249–250, 252, 257–260,
174–175; Waipi‘o, 55, 65–66 266–269, 271, 277, 284, 287, 293–294,

369
index

295–301, 308–309, 319nn. 4, 5, 6, 320n. 9, 322n. 29, 328n. 1, 330n. 37, 335n. 1, 336n.
321n. 22, 352n. 8 9, 337nn. 23, 36
mo‘o, 24 oli, 6, 13, 254, 260, 272
mo‘olelo, 2, 4, 10, 13, 16, 19, 23, 48, 62, olonä, 91, 163–164, 171, 211, 262
94–95, 127, 147, 150, 169–173, 191, Olopana, 51
193–194, 253, 260, 266–267, 284, ‘o‘opu, 70; Häna, 109–110, 129–132; Moloka‘i,
286–289, 293–294, 301, 304, 319nn. 1, 6, 208, 211, 216–219; Puna, 154
321nn. 19, 23, 322n. 28, 324n. 12, 336n. ‘opae, 70; Häna, 91, 109–110, 129–132,
11, 340n. 31, 347n. 154, 348n. 7, 352n. 6 136–137, 208, 211, 216–218, 221
Mua ha‘i küpuna Kahualele, 266, 285 ‘opihi, 18, 95, 120, 128–130, 133–134, 136,
170, 177, 184, 217–218, 262–263
Nahi‘ena‘ena, 56, 159 original Hawaiians, 12, 14, 23, 38, 42, 90, 115,
Nähiku, 10–11, 43, 91, 97, 115–116, 124, 128, 117, 127, 135, 152, 220, 241, 253, 289,
132, 140, 335n. 144 320n. 17, 323n. 44, 324n. 11, 350n. 43
Nanaue, 51, 88
Native Hawaiian population, 9–12, 20, 41–42, Pa‘ao, 25, 52, 150–151, 192, 336n. 17
348n. 4; contact, 30; decline, 30, 44, Padilla, Hokulani Holt, 273
322nn. 30, 32; Häna, 85–87, 91, 94, 100, Pä Hula Ka ‘Ie‘Ie, 266, 273–274
102, 105, 116–117, 125, 127; Kaho‘o- Pahupu, Daniel, 210, 319n. 5, 341n. 34, 342n.
lawe / Kanaloa, 264, 285; Moloka‘i, 44
196–197, 199, 201, 204, 207, 218, 224, Pala‘au-Ho‘olehua, 10–11, 204, 207, 216,
244, 246, 339n. 12; precontact, 23, 25; 231–232, 234–235, 237–238, 241, 267,
Puna, 57, 155–156, 158–160, 163, 167, 297, 299, 342n. 50, 345n. 92, 346nn. 119,
169, 178–179, 186; Waipi‘o, 56–59, 62, 127
79 participatory mapping, 181–182, 247, 290,
navigation: way-finding, 251, 253, 255, 295–297
258–259, 277, 303 Pele Defense Fund, 187–188, 189, 270, 295,
Nawahiokalaniopu‘u, Joseph, 166 339n. 72
night marchers (huaka‘ipö), 222 Pelehonuamea, 6–7, 21, 24, 31, 90, 96,
1946 tidal wave, 21, 45–46, 334n. 139; Häna, 143–150, 160, 169, 183, 186–190,
123–125; Moloka‘i, 218, 242; Waipi‘o, 66, 256–257, 261, 264, 270, 272, 277, 293
79–80, 82 Pelekunu, 202, 203, 206, 212–216, 218, 243,
nonconforming subdivisions, 177–179 342nn. 46, 57, 347n. 157
Nu‘u, 87–88, 94, 102–103, 112, 121–124, pepeiao, 40, 201
126–127, 288–289 petitions, 327n. 73, 338n. 49; 1845, 197–199,
308; Puna, 160, 305; Waipi‘o, 80
‘ohana, 5, 14–19, 24–28, 33, 35, 37, 40, 46, 61, Pi‘ilani, High Chief of Maui, 95–96; Pi‘ilani-
93, 106–109, 114, 127–128, 150, 156, 163, hale, 95, 140; trail, 135
171, 180–182, 185–186, 191, 203, 210, 219, pineapples, 46; Häna, 117, 119; Ho‘olehua
223, 226–227, 241, 248, 253, 265–268, 270, homesteaders, 201, 204, 207, 228, 230,
272–274, 276, 281, 283–285, 288–291, 232–234, 237–241, 243–244, 247; Puna,
295–296, 302–304, 319n. 4, 339n. 71; 167–168
Puna Hui ‘Ohana, 181, 339n. 70 place-names 14, 18, 220–221, 242, 246, 253,
Ola‘a, 157, 160, 163–169, 179, 307 257, 263, 283, 293, 319n. 6, 349n. 28
‘ölelo no‘eau, 22, 95, 143, 293–294, 321n. 26, plantations, 8–9, 29, 39–40, 43–46, 56, 59,

370
index

157–158, 278, 288; Hamakua, 59, 78; Pu‘uloa (Pearl Harbor), 43–44, 78, 125, 242,
Häna, 88, 103, 106, 116–119, 124–125, 348n. 10, 349n. 28
142; Moloka‘i, 201, 204, 228, 232–233,
247–248; Puna, 167–169 rains, named, 5, 14, 85–87, 95, 143, 254–255,
Pöhaku Ahu ‘Aikupele Kapili o Keaweiki, 254, 267, 283, 299
258 ranches, 8–9, 20, 40, 43–44, 290; Häna, 105,
Pöhaku Kuhi Ke‘e i Kahiki, 127 119, 121–123, 125–126, 130; Kaho‘olawe,
poi, 9, 26, 40, 58, 61, 63–67, 69, 71, 73, 79, 251, 303; Moloka‘i, 201, 204–205, 207,
91–92, 95, 104–105, 112, 119, 120, 228, 243, 245, 247–248, 297–300, 340nn.
139–141, 156, 171, 176, 186, 212, 214, 31, 32; Puna, 161, 167, 169, 292
220, 225, 232; mills, 141, 225 rights: access, 26–27, 31–39, 47, 128, 135–137,
politics and Native Hawaiians, 4, 8, 29, 32, 187–188, 190, 208, 282, 295–296, 352n. 7;
44–45, 48, 73, 152, 172, 201, 224, fishing, 164, 176; foreigners’, 31, 39;
249–251, 265, 268, 274, 278, 280, 284–285, Native Hawaiian land, 2, 31, 34–39, 208,
296, 302, 36n. 322, 336nn. 10, 20, 337nn. 278, 323nn. 44, 48; Native Hawaiian polit-
25, 28, 30, 35 ical, 31–34, 42, 249, 265, 268, 279–280;
private property, 32–33, 36–38, 101, 112, 160, water, 39, 138, 269
197. See also kuleana; land tenure; Mähele Reciprocity Treaty, 1876 and 1887, 40–42
Protect Kaho‘olawe ‘Ohana, 265–274, renaissance, Native Hawaiian, 4, 20, 48, 248,
276–277, 281, 283–285, 296, 303, 4n. 319 249, 251, 271, 276–278, 350n. 39
Puapualenalena, 51, 53 rubber plantation, 115–116, 124, 167–168
Puko‘o, 200, 206, 208, 210, 222, 242, 340n.
33, 341nn. 36, 39, 343nn. 61, 63, 344n. 82 Saiki, Congresswoman Patricia, 274–275, 285
Pukui, Mary Kawena, 2, 5, 12, 30, 62, 109, sandalwood trade, 30, 132, 164–165
122, 147, 155, 206, 210, 213, 216, 218–220, schooling, 15, 40, 43, 45, 62–63, 77–78, 82,
223, 231, 234, 242, 289, 296–297, 319nn. 257, 348n. 4
2, 5, 6, 320nn. 10, 12, 321nn. 19, 21, 25, schools, 284, 288, 296, 303; hälau hula, 271,
26, 322nn. 28, 29, 33, 325nn. 14, 27, 35, 277; Häna, 100, 103, 107–108 116,
328nn. 1, 4, 329nn. 4, 9, 12, 13, 14, 15, 17, 120–121, 127, 134, 136, 142, 332n. 77,
330nn. 24, 27, 28, 36, 37, 44, 331n. 69, 342n. 56; Moloka‘i, 203, 214–215, 217,
332nn. 87, 89, 91, 93, 94, 95, 333nn. 96, 223, 226–227, 237, 240, 344n. 91; naviga-
98, 99, 109, 334nn. 120, 122, 131, 335n. 1, tion, 257–258; Puna, 159, 164, 172,
336nn. 9, 10, 14, 337nn. 23, 36, 338n. 57, 176–177, 179; Waipi‘o, 63, 66, 77–78, 82.
340nn. 24, 32, 33, 341nn. 34, 38, 342nn. See also Hälau o Kekuhi; Pünana Leo
42, 44, 45, 47, 50, 56, 59, 343nn. 61, 62, sisal, 205–206, 212
63, 64, 71, 72, 74, 75, 76, 78, 79, 80, 81, springs, 14, 23, 39, 51, 87–88, 91–92, 120,
344nn. 82, 83, 84, 85, 86, 87, 88, 89, 90, 138–139, 142, 143, 148, 164–165,
91, 92, 119, 346nn. 126, 127, 128, 133, 168–169, 184, 200, 207, 220, 228, 232,
347nn. 146, 153 264, 272, 282, 291–292, 341n. 39
pulu, 40, 161, 201 stewardship principles, 5, 24, 26, 33, 193,
Puna, 6–7, 9, 12, 21, 54, 57, 96, 98, 143–190, 265–266, 281–284, 286, 288, 295, 299–301
194, 198, 297, 305, 306, 329n. 22, 353n. 7 stores, 8–9, 45; Waipi‘o, 60, 63, 71–72, 74,
Pünana Leo, 278 107, 116–117, 126, 137, 169–172, 214,
püpü, 70 219, 225, 232, 241, 247
Pu‘u Pe‘elua, 222 subsistence activities, 9, 12, 14–19, 22, 24–25,

371
index

28, 41, 44, 270, 278, 286–287; definition, wahi pana, 5–6, 8, 14, 18, 95, 143, 255–257,
16; Häna, 85, 105, 107, 109–112, 114, 117, 281, 291, 351n. 48
125, 127–129, 132, 289–292; Kaho‘olawe, Waikauikalä‘au, 92
253–254; Moloka‘i, 196, 199–200, 202, Wailau, 89, 198, 202–203, 206, 212, 216–218,
205, 207, 212, 218, 241, 243–248, 277, 221, 342n. 46
297–298, 300–301, 347n. 160, 348n. 161; Waipi‘o Mano Wai, 49–50, 82, 287
Puna, 156, 160, 163–164, 169–170, 172, Wao Kele o Puna, 188–190, 277, 295, 337n.
175–177, 180–184, 186–187, 294–296; 40, 352n. 7
Waipi‘o, 69–72, 82, 288 water, 8, 23, 26–28, 39, 43, 49–51, 54, 59–60,
63–67, 69–71, 75–76, 80–82, 83, 87–88,
taro, 2, 4, 9, 12–14, 21, 25–26, 35, 43, 45, 264; 90–92, 95, 97, 99, 109–110, 112–115,
278, 319n. 3, 320n. 16, 325n. 37, 326n. 47; 118–121, 128–129, 131–132, 134–135,
Haloa, 13; Häna, 91–92, 95, 103–104, 110, 137–141, 144–145, 148, 152, 157, 165,
112–120, 127, 129, 133, 137–142, 269, 171, 178–179, 182, 196, 200, 202, 204,
288, 290–291; Moloka‘i, 196–197, 200, 207–208, 210–212, 215, 219–221, 223,
206–208, 210–212, 218, 225, 242, 248; 226, 228, 232, 234,–235, 245, 248,
Puna, 157–158, 163, 170–171, 176–177, 256–257, 260, 263, 266, 269, 275,
182, 185–186; Waipi‘o, 49–50, 52–55, 277–278, 281–283, 287–288, 296,
58–61, 64–67, 69–70, 72–73, 76, 78–82, 298–299, 320n. 16, 323n. 51, 334n. 126,
287–288 341n. 39, 342n. 46, 351n. 48
taxes, 16, 40, 43, 47, 61, 160–161, 163, 197, weaving, 71, 112, 133, 171–172, 180, 300
201, 234, 294, 306, 309 Wilcox Rebellion, 1889, 42
thatched grass: churches, 158; houses 59, 61, Wilkes Expedition, 158, 337n. 41
112, 119–121, 158, 172, 270 Wilson, Jennie, 215, 342n. 56
tourism, 41, 46–48, 79, 125–126, 249, 284, winds, named, 5, 14; Häna, 83, 85–87, 121,
300 143, 220, 254, 283, 345n. 92; wind gourd,
turtles, 75, 215, 221, 245–246, 263 51, 299
World War I, 61, 69, 205, 215, 217
‘Umialïloa, 25, 49, 52–53, 152–153, 324n. 13; World War II, 21, 44–45, 73, 78, 82, 106, 124,
Ahu a ‘Umi, 25 141, 176, 207, 242, 249, 251

Wä Kahiko, 23, 321nn. 17, 19 Young, John, 30, 55, 57

372

You might also like