Stonehenge: A New Theory

Author(s): Benjamin C. Ray

Source: History of Religions , Feb., 1987, Vol. 26, No. 3 (Feb., 1987), pp. 225-278
Published by: The University of Chicago Press

Stable URL: https://www.jstor.org/stable/1062375

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/1062375?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend
access to History of Religions

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
 Benjamin C. Ray STONEHENGE:
A NEW THEORY

In recent years the study of Stonehenge has reached

impasse, with discussion concentrating unproductively u
of the astronomical features of the monument. Since the mid-1960s
astronomers and surveyors have been interpreting Stonehenge as a
solar and lunar observatory, designed to mark the extreme points of
the solar and lunar cycles for calendrical purposes.' Archaeologists,
however, have challenged these speculations and have insisted that the
monument was primarily ritual and magical in nature.2 Considering
the many views of Stonehenge that have been presented over the
centuries, including the current astronomical theories, one skeptical
archaeologist has astutely observed: "Every age has the Stonehenge it
deserves-or desires."3 This statement contains an important lesson

1 The most noted studies are by Gerald S. Hawkins (with John B. White), Stonehenge
Decoded (New York: Dell Publishing Co., 1965); and Alexander S. Thom, Archibald
Stevenson Thom, and Alexander Strang Thom, "Stonehenge," Journalfor the History
of Astronomy 5 (1974): 71-90, and "Stonehenge as a Possible Lunar Observatory," ibid.,
6 (1975): 19-30; cf. Alexander Thom and Archibald Strang Thom, Megalithic Remains
in Britain and Britany (Oxford: Clarendon Press, 1978), chap. 11.
2 Thorough and sometimes trenchant archaeological criticisms have been made by
R. J. C. Atkinson, "Moonshine over Stonehenge," Antiquity 40 (1966): 212-16, and
"Aspects of the Archaeoastronomy of Stonehenge," in Archaeoastronomy in the Old
World, ed. D. C. Heggie (Cambridge: Cambridge University Press, 1982); and Jacquetta
Hawkes, "God in the Machine," Antiquity 41 (1967): 174-80. For a recent ritual
interpretation, see Aubrey Burl, Rites of the Gods (London: J. M. Dent & Sons, 1981).
3 Hawkes, p. 174.

?1987 by The University of Chicago. All rights reserved.

0018-2710/87/2603-0001$01.00

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
226 Stonehenge

for anyone dealing with the subject. The scantiness of the archaeo-
logical record at Stonehenge means that any attempt to explain the
function of the monument is necessarily interpretive; and in the past
most attempts have tended merely to reflect the cultural biases of their
times. Nevertheless, recent work in archaeology and archaeoastronomy
in Britain indicates that something more than total skepticism is
possible, even though it is clear that the current astronomical inter-
pretations are no longer viable, at least in their present form.
Historians of religion, whose work sometimes relates to prehistoric
phenomena, appear to have ignored the subject of Stonehenge perhaps
because of its increasingly complex astronomical character. To my
knowledge only Mircea Eliade has commented upon the contemporary
situation and offered an opinion. Noting that astronomy has made a
contribution and that the religious significance of the monument is still
disputed, Eliade has proposed that Stonehenge in its earliest form "was
a sanctuary built to insure relations with the ancestors."4 This is not a
new interpretation nor one that has received much recent attention,
although it is especially pertinent in light of contemporary work. As
for the later, more megalithic phases of the monument, Eliade has
nothing specific to say except to state that for "megalithic religions" in
general "the ideas of perenniality and of continuity between life and
death are apprehended through the exaltation of the ancestors as
identified, or associated, with stones."5 Nevertheless, Eliade has pointed
out that comparative research based upon ethnographic examples
might be able to make some progress in the subject of"megalithicism."
And in the case of Stonehenge such research must be concentrated
upon the immediate prehistorical context of southern England.
The principal question about Stonehenge concerns its function,
especially its ritual function, in late Neolithic and early Bronze Age
Britain. This is a subject to which a historian of religions trained in
ritual studies among nonliterate and ancient societies might make a
contribution. British prehistorians themselves have already led the way
by making use of ethnographic comparisons in interpreting Neolithic
social and religious phenomena, including megalithic monuments, and
it is evident that they are usually associated with the cult of the dead.6
Of course, whatever is said about the ritual function of Stonehenge
needs to be formulated in a hypothetical way, as must all interpretive

4 Mircea Eliade, A History of Religious Ideas, trans. Willard R. Trask (Chicago:

University of Chicago Press, 1978), 1:118.
5 Ibid., p. 124.
6 See Colin Renfrew, Before Civilization (New York: Alfred A. Knopf, 1973), and
Approaches to Social Archaeology (Edinburgh: Edinburgh University Press, 1984);
Aubrey Burl, Prehistoric Avebury (New Haven, Conn.: Yale University Press, 1979).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 227

explanations of the monument. Thus the question is, If Stonehenge

had a ritual function, what might it have been? This is the question that
I shall attempt to answer, while at the same time staying as closely as
possible within the limits of the evidence. In passing I shall also
consider why the recent astronomical interpretations of Stonehenge
have achieved a certain popularity and persuasiveness far beyond the
evidence and have become something of a cultural fashion.

THE SOCIAL ARCHAEOLOGY OF NEOLITHIC WESSEX

One of the weaknesses of the astronomical approach to

has been the lack of historical perspective. As Jacquetta
pointed out, "The lack of historical imagination, the ig
historical knowledge, made it possible to identify this monu
an abstract [surveyor's] diagram and analyze it from the po
of wholly modern concepts."7 In recent years the study
prehistory has undergone significant change, which has rad
the understanding of the Neolithic and early Bronze Age, in
origins and development of megalithic monuments. The
of calibrated Carbon-14 dating has led prehistorians to a
invasion-diffusion conception of British Neolithic histo
megalithic monuments of Britain and Europe are now cl
antedate those of the eastern Mediterranean.8 Gone are t
and Mycenaean travelers who were once thought to ha
Europe and Britain with megalithic monuments and bron
This has led to a new estimation of the "barbarians" of
Britain, especially their intellectual and technological abi
seems that they were erecting large-scale monuments o
stone, making basic astromonical observations, and smel
on their own without any help from the eastern Mediterran
has also been a change in the understanding of the
between Britain and the Continent in this period. Rath
invading, brachycephalic "Beaker Folk," whose existence was
sized on the basis of distinctive Beaker pottery and burial r
Beaker complex now seems to be part of an internation
package," stretching from Ireland and Britain to Spain a
which is everywhere blended into local contexts alongsi
facts.9 Thus, the "conquering" Beaker people, who supp

7 Jacquetta Hawkes, "The Proper Study of Mankind," Antiquity 42 (19

8 See Renfrew's Before Civilization for an excellent account of this d
British prehistory, including the dating of megalithic monuments. See a
"Megalithic Monuments," Scientific American 243 (1980): 78-90.
9 See Colin Burgess, The Age of Stonehenge (London: J. M. Dent
pp. 62, 160-61. See also Alasdair Whittle, "Later Neolithic Society in Brit

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
228 Stonehenge

Stonehenge II and III, have all but disappeared from current archaeo-
logical vocabulary.
Given this stituation, a theory of indigenous historical change was
therefore needed to account for developments in technology, society,
and religion in Neolithic Britain. Such a theory has been proposed by
the archaeologist Colin Renfrew for Wessex, those several counties of
southern England in which Stonehenge and other large Neolithic
monuments are located.10 Renfrew's theory is based upon the size,
number, and spatial distribution of Neolithic sites and upon demo-
graphic and sociological considerations, including ethnographic com-
parisons. In southern England the Neolithic period dates from the
development of the first farming communities in 4000 B.C. to the
development of bronze technology in about 2000 B.C., when the
building of megalithic monuments was mostly over. In Wessex the
earliest monuments date from about 4000 to 3500 B.C., and they are not
in fact megalithic. These monuments include a large number of long
barrows, great oblong earthen mounds about 300 feet in length with
wooden burial chambers. In certain areas where large stones were
available, the burial chambers were made of great untooled stones and,
hence, were megalithic constructions. The other type of monument of
this period was the causewayed camp, a large circular enclosure about
650 feet in diameter surrounded by concentric rings of ditches. Later, in
about 3000 B.C., a new type of construction appeared. This was the large
earthen circle, or "henge," a term derived from Stonehenge, although
most of them do not contain stone circles. These large enclosures are
surrounded by a single bank and an interior ditch, with one or more
entrances. There are five such henges in Wessex. Excavation of some,
for example, Durrington Walls, about two miles northeast of Stone-
henge (fig. 1), has revealed the presence of massive circular timber
buildings (fig. 2). Finally, the list of Neolithic monuments in Wessex
includes Silbury Hill, a colossal man-made mound of chalk soil 130
feet high, the largest ancient mound in Europe; Avebury, a great
standing stone circle more than twice the diameter of Stonehenge; the
Dorset Cursus, a six-mile-long pair of straight, parallel ditches south
of Stonehenge (a smaller cursus lies close to Stonehenge); and, of
course, Stonehenge itself, with its unique circle of stones capped by

ment," in Astronomy and Society in Britain during the Period 4000-1500 B.C., ed.
C. L. N. Ruggles and A. W. R. Whittle, British Archaeological Reports, no. 88 (Oxford:
BAR, 1981); and Richard J. Harrison, The Beaker Folk (London: Thames & Hudson,
1980), pp. 159-66.
10 Renfrew, Before Civilization, pp. 228-42; Colin Renfrew, "The Social Archaeology
of Megalithic Monuments," Scientific American 249 (1983): 152-62, and Approaches to
Social Archaeology, chaps. 6, 8.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
 'Is

~~~~~-1 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

le -. . .

00, ~ ~ K~ ~

FIG. 1.-The environs of Stonehe

rington Walls, Woodhenge, the
Taken from R. J. C. Atkinson, S
University Press, for the Royal
1979), map 1 (copy is taken from t

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
230 Stonehenge

AERIAL VIEW FROM

Seclion F D C BA

0 10 20 30 40 S0 60 70 80 90 100
k=q_ L Feet
o 10 20 30
I Il I Metres

FIG. 2.-Large timbe

ed., Recent Work in
Press, 1975), p. 62.

lintels. Indeed, the

organization of sett
control and compe
Noting the spatial d
Renfrew has "mapp

11 Collin Bowen, "Patte

Neolithic to Roman Tim
(Bradford-on-Avon: Mo

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 231

containing a number of long barrows, a single causewayed camp (one

cluster has two), and a single large henge. In Renfrew's view this
arrangement implies modest spatial hierarchy, if each causewayed
camp is regarded as the local center for a small region defined by a
group of long barrows. The later earthen henges were built in these
same five areas. As for differences in scale, it is estimated that a long
barrow took about 5,000-10,000 man-hours to construct and cause-
wayed camps about 40,000-100,000 hours. Much greater labor was
needed to construct the ditch and bank of Durrington Walls. Even if
these manpower estimates are scaled downward somewhat, as has been
recently suggested,12 they still represent significantly different orders
of magnitude. Given the large investment of labor and organization
involved in building the monuments, Renfrew has proposed that by the
late Neolithic times of the third millennium, the five Wessex regions
were organized into five chiefdoms. By "chiefdom" Renfrew means a
society with marked social hierarchy, governed by birth, whose several
subgroups are headed by subordinate chiefs, united around a single
chief who enjoys high prestige and ritual position. Such societies
encourage redistribution of resources, craft specialization, and collec-
tive ceremonies that serve to express and enhance social unity. In
chiefdoms great public works projects, such as megalithic monuments,
are made possible by population size, centralized control of foodstuffs,
and the ceremonial and military solidarity of the society as a whole.
Renfrew's theory is that in the fourth millennium European societies
living close to the Atlantic seaboard began to build megalithic monu-
ments as territorial markers and ritual centers. This, he speculates,
occurred at a time when population pressure upon the land increased
as a result of the agricultural revolution and thus created a need for
expressions of territoriality and ritual solidarity by segmentary soci-
eties. In Wessex such cooperative efforts eventually led to the welding
together of local kinship groups into chiefdoms.
Each of the five Wessex chiefdoms was marked by a cluster of long
barrows (numbering about twenty-two) belonging to the clan chiefs,
and each had a center or "capital," first the causewayed enclosures,
then the enclosures of the large henge type, containing large buildings,
as at Durrington Walls. In this instance, the archaeologist Geoffrey
Wainwright argues, the large building was surrounded by smaller
domestic dwellings and served as the ceremonial center for the sur-
rounding population in the vicinity of Stonehenge. The people lived in
stockaded farmsteads, scattered about the territory, and raised wheat

12 Bill Startin and Richard Bradley, "Some Notes on Work Organization and Society
in Prehistoric Wessex," in C. L. N. Ruggles and A. W. R. Whittle, eds. (n. 9 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
232 Stonehenge

and barley and kept cattle, sheep (or goats), and pigs. The two largest
Wessex monuments, Silbury Hill and the sarsen stone structure of
Stonehenge III, required the greatest efforts, estimated at 18,000,000
and 30,000,000 man-hours, respectively.'3 Renfrew believes that the
organization and labor necessary for these huge projects required even
broader-scale cooperation, something like a federation of chiefdoms
under a paramount chief.
At the time of Stonehenge II and III, in the beginning third
millennium (early Bronze Age), a smaller-scale monument was also
starting to proliferate in the Wessex area. This was the round barrow
whose burial contents, in contrast to the modest collective burials of
the long barrows, indicate a stratified and structured society, emphasiz-
ing the prestige of individual chiefs and the display of wealth. Most of
the barrow groups in the Stonehenge area and many of the individual
barrows lie in prominent positions within sight of the monument
(fig. 1). The third millennium was also a time of agricultural expansion,
population growth, and political consolidation, which lasted until the
middle of the second millennium (the middle Bronze Age), when
worsening climate and population pressure created widespread agri-
cultural disaster and serious population decline. This marked the end of
the megalithic period and of Wessex culture, and the beginning of the
increasingly pastoralist economy and militaristic activity of the late
Bronze Age.
Historically, then, Stonehenge must be fitted into this slowly evolving
scene of larger and more complex building projects and increasingly
larger and more complex forms of political organization. From its
inception, Stonehenge was not the product of an immigrant Mycenaean
architect, as formerly supposed, but of local clans and chieftains,
skilled in mounting great engineering projects (Silbury Hill and Stone-
henge I are roughly contemporary) and in constructing great circular
timber buildings (the structures at Durrington Walls, Woodhenge, and
Stonehenge II and IIIa are roughly contemporary). Renfrew's point
about each of these phases is that monument building and ceremonial
activity were instrumental in creating and maintaining the political and
economic organization of the society.
In circa 3100 B.C. Stonehenge started out as a modest-size henge, 320
feet in diameter, with fifty-six shallow pits, called Aubrey Holes,14
located around its inner circumference (fig. 3). In these respects

13 Startin and Bradley have estimated Stonehenge II and IIIa at 360,000 and 1,750,000
man-hours, respectively.
14 Named after the seventeenth-century antiquarian John Aubrey who discovered
them.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 233

FIG. 3.-Plan of Stonehenge, including elements revealed by excavation.

Taken from Christopher Chippendale, Stonehenge Complete (Ithaca, N.Y.:
Cornell University Press, 1983), p. 9, fig. 4.

Stonehenge resembled similar medium and small-size, single-entrance

henges in Wessex of this period with circular banks and ditches and a
number of pits around the edge. Some of the larger henges, such as
Durrington Walls, Marden, and Mount Pleasant, contained large
timber buildings inside their enclosures. These buildings were made of
concentric circles of wooden posts, supporting, it is assumed, a cone-
shaped roof (fig. 2).15 Atkinson suggests that a smaller circular building
may also have stood at the center of Stonehenge I. Woodhenge, a small
henge near Durrington Walls (fig. 1), less than two miles from Stone-
henge, contains the postholes of an oval building made from concentric
rings of posts. Like Stonehenge, the embankment of Woodhenge has a

15 Geoffrey Wainwright, "Durrington Walls: A Ceremonial Enclosure of the 2nd

Millennium B.C.," Antiquity 42 (1968): 20-26, and "Woodhenges," Scientific American
223 (1970): 30-38.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
234 Stonehenge

single causeway entrance opening toward the northeast in the direction

of the midsummer sunrise (fig. 4).16 An unexcavated henge on nearby
Coneybury Hill is also oriented toward the northeast, an orientation
that is otherwise uncommon for henges with single entrances.17 Stone-
henge I therefore fits in well with neighboring sites of the middle
Neolithic period. The major difference is that Stonehenge I had a pair
of outlying monoliths, the Heel Stone and its mate (stone 97, now
missing), located just outside the entrance to the northeast, and possibly
the Station Stones, located around its circumference and aligned on
the summer solstice sunrise and winter solstice sunset (figs. 3, 5).

STONE CIRCLES AND ARCHAEOASTRONOMY

In addition to the prehistory of Wessex it is also necessary t

Stonehenge in the context of the stone circles of Wessex and
Britain. Stone circles are not easily dated because of the
datable remains associated with them, but it is assumed that
constructed between about 3500 and 1500 B.C. Over nine hundred
circles still exist in the British Isles, and Aubrey Burl has estimated
twice this number may originally have been built. Thanks to Bu
masterly study, The Stone Circles of the British Isles, we have a mu
clearer picture of how Stonehenge fits into the stone circle
medium- to small-size henge context of Wessex and the res
Britain."8 Most of the henges were built in the eastern part of
country, and the stone circles were erected in the stony areas of
west and north. Since circles and henges are roughly contempor
Burl suggests that they both served as ritual centers. The stone circ
of the west may therefore have been the counterparts of the henge
the east. Stonehenge lies along a central zone in which both large sto
circles and large henges exist together, and where circle henges seem
be a combination of the two. While it is possible to suppose that ston
circle construction derived from the earthen henges, this theory m
await further evidence. At Stonehenge, however, the answer is c
the henge of Stonehenge I was built first, about 3100 B.C., and the s
circles were added later. The double bluestone circle of Stoneheng
was constructed about 2100 B.C., and it may have replaced a circ
timber structure, if Atkinson's assumption is correct. This is w

16 M. E. Cunnington, Woodhenge (Devizes: George Simpson, 1927); Euan W. Ma

Science and Society in Prehistoric Britain (London: Paul Elek, 1977), pp. 17
Alexander Thom, Megalithic Sites in Britain (Oxford: Clarendon Press, 1967), p.
17 R. J. C. Atkinson, Stonehenge and Its Environs, Royal Commission on Histo
Monuments (England) (Edinburgh: Edinburgh University Press, 1979), p. x.
18 Aubrey Burl, Stone Circles of the British Isles (New Haven, Conn.: Yale Unive
Press, 1976).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
 History of Religions 235

0 0t 10

, ,\\\\llk\!/ '
x\\x
K
'. \\ 1
K
- x

' t 1 . '- .' "/ J '

~, IS"
,-_/. III//
* .

Britain (London: Paul Elek, 1977), p. 172, fig. 26.

a large henge fifty miles south of Stonehenge, where an open-sided

buildingatthat
happened the had fallen not
Sanctuary, intofar
decay.
fromThere is evidence
Stonehenge. and sarsen
Here large Soiety in Prehis
Britainated
stones wereinplaced
the same
nextway1977), p. 172,This
to the postholes ofhas
twoled Wainwright
concentric to suggest
rings of the
former building. The same seems to have occurred at Mount Pleasant,

happenedt the transformation of timber structures into stonehenge. Here large sa

square or cove of standing stones replaced a large round timber
special [ritual] function for the mustholes of two concentrbuildings .. whic rings o
necessary former building. The same seems to have occurmanently indicatd byMount Ple
circles originated in the same way. This has led Wainwright to sugge
ta large heng e f ifty mi les south of Stonehenge where an open-sid
that the transformation of timber structures into stone indicates "some
squacompleted before of standismantled and replaced by the large round ti
special [ritual] function for the multi-ring buildings ... which made it
building that had fall Stonehen into decay. There is evidence that oterec ston
necessary for their ruins to be permanently indicated by stone set-
tings."]9 The concentric bluestone circle at Stonehenge was not quite
completed before it was dismantled and replaced by the large sarsen
stone circle of Stonehenge III. Later, the bluestones were re-erected in

19 Geoffrey Wainwright, "Religion and Settlement in Essex, 3000-1700 B.C.," in

Fowler, ed. (n. 11 above), p. 68.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
236 Stonehenge

-29 - " -29

+29 __(- -9

FIG. 5. Station rec

1500-1100 B.C. (fi

The pair of outl
orientation (fig. 3
(96) remains, was
like some of the
been known, and
were later constru
astronomical orientations of the other stone circles of the
British Isles. In 1967 Alexander Thom, then retired professor of
engineerieering at Oxford, published the results of his field surveys of over
three hundred stone circles, cairns, and standing stones in England,
Scotland, Wales, and Ireland.20 The accuracy of his field measurements,

20 A. Thom, Megalithic Sites in Britain. See also A. Thom, Megalithic Lunar Observa-
tories (Oxford: Clarendon Press, 1971); and Thom and Thom (n. 1 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
 1,j ~ ~ ~ ~ ~ 1 V

Jft~i4R

4~~~~~~~~~~
''Bluestone Circle (about 60 stones)
Bluestone Horseshoe (1 9 stones)

FIG. 6.-Artistic reconstruction of Stonehenge lllc. Michael Balfour, illustra

rgt 1979 Michael Balfour. Used by permission of Charles Scribner's Sons

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
238 Stonehenge

the precision of his analyses, and the quantity of his data set a new
standard for the young discipline of archaeoastronomy. On the basis of
his findings, which did not initially include Stonehenge, Thom argued
that the megalithic builders made their circles and alignments for the
purpose of marking the solstices, the equinoxes, and nineteen-year
cycle of the moon, and the rising and setting of sixteen bright (first-
magnitude) stars.
Since the solar and lunar cycles examined by Thom and others are
basic to the study of megalithic sites, including Stonehenge, it will be
useful to review them briefly for purposes of clarification.
In the northern hemisphere the midsummer sun rises in the north-
east at a declination21 of +24?, and it appears to rise at the same point
on the horizon for three or four days without moving, that is, to
stand still (solstice) until its rising point appears to move southward.
By September 21, the autumnal equinox, the sun rises exactly east at a
declination of 0?, and by midwinter on December 21 the rising position
of the sun appears to stop again, this time far down on the horizon at
the southeast at a declination of -24?. By the spring equinox, March 21,
the sun rises again due east, and it continues to move day by day nearer
to the midsummer solstice on June 21 at a declination of +24?. The
moon's path lies close to the sun (within 5? either side of it), so th
rising and setting positions of the moon roughly coincide with those of
the sun, though not at the same place and the same time. Like the sun,
the moonrise oscillates between an extreme northeast point at
declination of +29? (24 + 5) and an extreme southwest point at
declination of -29? (-24 -5) and back again to the northeast.
However, due to the regression of the moon's orbit and its perturbation
by the gravity of the earth, the moon's extreme rising positions are no
the same each month. Over a nineteen-year period (actually 18.61
years), the northern moonrise moves from a maximum declination o
+29? to a minimum declination of +19? and back to the maximum
again. The same is true of the southern moonrise whose maximum and
minimum declinations are -29? and -19?, respectively. Thom has
called these maximum and minimum extremes the "major standstill"
and the "minor standstill," and these terms have become part of the
vocabulary of archaeoastronomy. The moon, of course, does not really
stand still but reaches close to the extreme declination with little
noticeable change from month to month for several months. In this
sense a lunar standstill is analogous to a solstice, except that the period
between major standstills is nineteen years, not a single year, as in the

21 Declination is the angular measurement of an astronomical object in reference to its

position north or south of the celestial equator, which is an extension of the earth's
equator.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 239

case of the sun. The interval between the major and minor standstills is
half the nineteen-year cycle or approximately nine and one-half years.
For virtually all ancient societies, the monthly phases of the moon were
significant, especially the first appearance of the new moon and the full
moon; and it is assumed by Thom and others that the full moon was
the phase when the sightings were made. By contrast with the complex-
ities of the moon, the stars move in regular annual cycles, with little
noticeable change over the years in their annual rising and setting
positions on the horizon. Ancient civilizations paid attention to the
rising and setting of both individual bright stars, such as Sirius, and to
groups of stars, such as the Pleiades and the constellations of the
zodiac.
On the basis of his site surveys and astronomical interpretations,
Thom believed that the fundamental motive for the megalithic builders
was scientific, not ritualistic. He argued, with impressive mathematical
precision, that the megalithic sites were astronomical observatories
where alignments were established with an accuracy of y2? or a single
solar diameter. At these sites, Thom supposed that genuine, practical
astronomy was carried out for the purpose of devising an accurate
sixteen-month calendar. The precision of Thom's site surveys also led
him to claim that the Neolithic builders used a standard unit of
measurement, a megalithic yard (2.72 feet), which they employe
every site across the British Isles. Thom also attempted to demonstr
on the basis of his site surveys that the megalithic builders w
interested in exploring the mathematical properties of various g
metrical shapes: circles, egg shapes, ellipses, flattened circles, isoscel
triangles, Pythagorean triangles, and spirals.
Thom's work created an immediate impact upon astronomers b
drew little response from prehistorians. As archaeologist Richa
Atkinson pointed out in 1975, this was because "many prehistor
either ignore the implications of Thom's work, because they do
understand them, or resist them because it is more comfortable to d
so."22 Atkinson himself was prepared to accept that Thom's resu
were not due to chance, and he was willing to raise his estimation of
intellectual and technological abilities of the Neolithic peoples
Britain, in keeping with the implications of calibrated Carbon-
dating.
In recent years, however, Thom's work has been subject to critical
examination.23 Analysis of Thom's data indicates that some of his

22 R. J. C. Atkinson, "Megalithic Astronomy-a Prehistorian's Comments," Journal

for the History of Astronomy 6 (1975): 51.
23 See Evan Hadingham, "The Lunar Observatory Hypothesis at Carnac: A Recon-
sideration," Antiquity 55 (1981): 35-42; D. C. Heggie, "Megalithic Astronomy: High-

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
240 Stonehenge

results were subjectively biased or due to chance or based upon what

Thom calls "indicated foresights," which figure significantly in his later
work. According to Thom, an indicated foresight refers to a prominent
natural feature of the landscape on or near the horizon indicated by a
backsight, that is, a stone or line of stones. An indicated foresight may
be a distant mountain peak, a distant notch in the horizon, the side of a
slope, or, when a sea horizon is involved, a rock far out at sea or the
cliff of an island. By way of clarification, one of Thom's sons and
collaborators, A. S. Thom, points out that "if we stand at a marked
backsight [one or more stones] and make careful observations of the
profile of part of the horizon which turns out to contain a significant
[astronomical] position we can assume that we are at a real observing
point."24 This statement reveals the fundamental ambiguity of Thom's
notion of an indicated foresight and, hence, the subjectivity involved in
the selection process. It turns out to be essentially a matter of designat-
ing features of the landscape as "indicated foresights" on the basis of
their known relation to solar, lunar, and stellar positions for the site in
question. Moreover, in his work Thom never mentions how many
possible natural foresights exist at a site, only the ones for which
there are astronomical alignments. Thus it is impossible to determine
their probability and to decide whether they are intentional or merely
due to chance. For Thom, however, it is the accuracy of the alignments,
not their statistical probability, which is convincing.25 The astronomer
Clive Ruggles's recent study of 189 megalithic sites in Scotland, which
does not make use of indicated foresights, gives little evidence to
support Thom's work. The only alignments Ruggles found were those
for the maximum lunar standstill and the winter solstice, and these

lights and Problems," in Heggie, ed. (n. 2 above), "Highlights and Problems of
Archaeoastronomy," Archaeoastronomy, no. 3 (Journal of the History of Astronomy
[JHA] 12 [1981]): S17-S37, and "Megalithic Lunar Observatories: An Astronomer's
View," Antiquity 46 (1972): 43-48; G. Moir, "Megalithic Science and Some Scottish Site
Plans: Part 1," Antiquity 54 (1980): 37-43, and "Some Archaeological and Astronomical
Objections to Scientific Astronomy in British Prehistory," in Ruggles and Whittle, eds.
(n. 9 above); Jon Patrick, "A Reassessment of Solstitial Observatories at Kintraw and
Ballochroy," in Ruggles and Whittle, eds.; C. L. N. Ruggles, "A Reassessment of the
High Precision Megalithic Lunar Sightlines, Parts 1, 2," Archaeoastronomy, nos. 4, 5
(JHA 13, 14 [1982, 1983]): S21-S40, S1-S33, and "Megalithic Astronomical Sightlines:
Current Reassessment and Future Directions," in Heggie, ed. (n. 2 above), and "A
Critical Examination of the Megalithic Lunar Observatories," in Ruggles and Whittle,
eds. (n. 9 above); Clive Ruggles and Ray Norris, "Megalithic Science and Some Scottish
Site Plans: Part 2," Antiquity 54 (1980): 40-43.
24 Alexander Thom and Alexander Strang Thom, "Astronomical Foresights Used by
Megalithic Man," Archaeoastronomy, no. 2 (JHA 11 [1980]): S90-S91; cf. A. S. Thom,
"Megalithic Lunar Observatories: An Assessment of 42 Lunar Alignments," in Ruggles
and Whittle, eds. (n. 9 above), p. 19.
25 Thom and Thom, "Astronomical Foresights Used by Megalithic Man," p. S94.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 241

alignments were within only 1? or 2? of accuracy. There was no

evidence of alignments for the summer solstice, equinox, other lunar
standstills, or stellar positions, as reported by Thom for Scottish
sites.26 In connection with the study of Scottish recumbant stone
circles, Burl has proposed that "analysis of entire groups might reveal
that, although the monuments do contain alignments, these were
symbolic rather than scientific and were linked with death."27 This, he
notes, is true of earthen and megalithic long barrows, passage graves,
stone cairns, and many stone circles. Stonehenge I with its solstice
alignment and fifty-six burial pits (the Aubrey Holes) clearly fits into
this tradition, and it is worth considering whether the stone circles of
phases II and III had any ritual relation to the dead as well.

ASTRONOMY OF STONEHENGE

As the evaluation of Thom's work continues and the invest

other circles and alignments progresses, it is abundantly clear
study of Stonehenge must be fitted into the developing
archaeoastronomy in the British Isles. Astronomers can n
write about Stonehenge as if it were virtually a unique case.
Although Thom pioneered the recent astronomical study
megalithic sites, he was not responsible for starting the curre
nomical investigation of Stonehenge. This was the work o
most notably Gerald S. Hawkins, a British-born astronomer
then at the Smithsonian Astrophysics Observatory at M.I.T
articles in Nature (1963, 1964) Hawkins asserted the ex
twenty-four solar and lunar alignments at Stonehenge, and he
the theory that the monument could have been used as a ca
predict eclipses. His subsequent book, Stonehenge Decoded
stimulated a surge of popular interest in Stonehenge and prov
most significant material for subsequent debate about its astro
meaning and function. For Hawkins the purpose of Stoneh
threefold: to serve as a solar and lunar observatory for c
purposes, to legitimate priestly power (especially by predictin
and to advance intellectual curiosity about the heavens. How

26 C. L. N. Ruggles, Megalithic Astronomy, British Archaeological Repo

Series, 123 (Oxford: BAR, 1984), pp. 303-10. See also Euan W. MacKie's re
book and his defense of Thom's methods and assumptions ("Megalithic A
Archaeoastronomy 7, nos. 1-4 [1984]: 144-50).
27 Aubrey Burl, "Science or Symbolism: Problems of Archaeoastronomy,"
54 (1980): 191-200, and see also "'By the Light of the Cinerary Moon':
Tombs and the Astronomy of Death," in Ruggles and Whittle, eds. (n. 9
C. L. N. Ruggles and H. A. W. Burl, "A New Study of the Aberdeenshire
Stone Circles, 2: Interpretation," Archaeoastronomy, no. 8 (JHA 15 [1985

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
242 Stonehenge

To Heelstone
+24
4

0 Sunset nrise
Tril-t^ho
V Trilithon S unrise S2
Trilithon
-24 o I 0 >-24
+24--- 0
+29 0 IVIIO---6'19
0 Iv 9
Sunset i

MooonsetTrilthon C - -29
FIG +1k Mtilithon
19 7.Hwkn' CDS
trilitho Moonrise
a Trilithon Q '

-24
--- Sunrise
----* Moonrise n
4- Sunset
-1--- Moonset i

FIG. 7.-Hawkins's trilit

Stonehenge Decoded by
Hawkins and John B. W

now recognized that

possible use as a calcula
agreed that he overesti
involved, although the
that can be confidently
This last point raises t
Hawkins, for example
to the great central t
sunrise axis of the mo
"archways" for viewing
at their extreme poin

28 D. C. Heggie, Megalithi
E. C. Krupp, ed., In Search
Co., 1979), pp. 102-3.
29 Heggie, Megalithic Scie

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 243

sarsens of the outer circle as viewed from the trilithons are about 8? or
sixteen solar or lunar diameters wide. This breadth is hardly precise,
admitting of astronomical "errors" of up to 8? of azimuth, so wide that,
according to one astronomer, "they can hardly be said to indicate solar
or lunar positions at all."30 Atkinson has also argued that Hawkins's
rule of accepting any sightline that falls within +2? of arc (or four solar
diameters) is far too broad, since an alignment may be fixed with a pair
of sticks within limits of five minutes of arc or twenty-four times
greater accuracy.31 Thus it would have been possible for the builders of
Stonehenge to have achieved much smaller errors, if this had been their
intent. Finally, there is the matter of Hawkins's alignments being
attributable entirely to chance. On this question Hawkins greatly
underestimated the probability of his proposed alignments. Subsequent
calculation has shown that there is nothing improbable about ascribing
all of Hawkins's alignments to chance. In fact the thirty-two astro-
nomical alignments that Hawkins found out of the 240 that he
examined are far fewer than the forty-eight alignments that could be
expected from a purely random positioning of stones. The only
alignment on which there is universal agreement is the solstitial
orientation of the main axis.32
Hawkins also seems to have been naive in his efforts to lend
credibility to his interpretations by featuring the use of a large,
7094 computer, the most advanced machine of its day. Thus
referred to his conclusions as the "machine's findings" and por
his role as "little more than a middleman, a means of bringing ma
to monument... or vice versa."33 This was in keeping with the titl
his book, Stonehenge Decoded, which conveyed the misleadin
that interpreting Stonehenge was analogous to "decoding" a tex
that only a machine could do the job properly. The public imp
Hawkins's article in Nature, also titled "Stonehenge Decoded,
worldwide and overwhelmingly positive, and Hawkins himself
dered if the use of a computer influenced his readers. Stylistically
book was an adventure story: "Marvelously exciting. . . delight
read . . . has dash, vibrance, and a simplicity that compel any read
as one newspaper review put it.34
Ten years after Hawkins's initial studies Thom undertook an inv
igation of Stonehenge and agreed that it could have been use
30 Ibid., p. 200; cf. Atkinson, "Aspects of the Archaeoastronomy of Stoneheng
above), p. 113.
31 Atkinson, "Moonshine over Stonehenge" (n. 2 above), pp. 213-14.
32 Heggie, Megalithic Science, pp. 149-51; Atkinson, "Moonshine over Stonehenge,"
p. 214, and "Aspects of the Archaeoastronomy of Stonehenge," pp. 114-15.
33 Hawkins (with White) (n. 1 above), p. 121.
34 Daily Sentinel, Grand Junction, Colorado, as quoted on the back cover of Hawkins,
Stonehenge Decoded.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
244 Stonehenge

solar and lunar observatory. But he held that "no accuracy could have
been possible if the backsights and the foresights were confined to the
monument itself," thus rejecting Hawkins's interpretation.35 To over-
come the problem, Thom treated Stonehenge as he had other stone
circles, pointed out some distant features of the landscape (three
mounds and a knoll) that fitted significant astronomical positions, and
suggested that they might have served as solar and lunar foresights.36
Subsequent investigation by Atkinson has shown that these "foresights"
are all later than Stonehenge.37 In 1966 three large postholes were
discovered during the extension of the visitor's parking lot across the
road. C. A. Newham, an amateur surveyor, proposed several lunar
alignments for these postholes as seen from the center of Stonehenge I
and regarded them as "the most significant 'astronomical' discovery yet
made at Stonehenge."38 Thom also thought they were astronomically
significant. However, the wood material found in two of them dates
from about 4700 B.C., and thus they cannot be connected with the
astronomy of Stonehenge, even at its earliest phase.39
The result is that, except for the main solstitial axis of the monument,
there is no evidence that Stonehenge was constructed for astronomical
purposes. This raises again the question of what constitutes an align-
ment. Atkinson has noted in connection with Hawkins's and Thom's
work that the builders of Stonehenge had the engineering skills to
place stones and posts with considerable precision, and they could
have achieved far greater astronomical accuracy than the site actually
exhibits. Because the alignments attributed to Stonehenge all involve
significant "errors," they are therefore at variance with the known
capabilities of the builders.40 On this point Atkinson is obviously right.
It is therefore necessary to suppose either that the approximate align-
ments are coincidental or that they are intentional and not designed for
astronomical precision.
To explore the latter alternative suggests the possibility that the
alignments may be symbolic rather than scientific and that they may be
related to ritual rather than astronomy. In this connection the archi-
tectural properties of Stonehenge are of major importance. As Burl has

35 Thom, Thom, and Thom, "Stonehenge" (n. 1 above), p. 86.

36 Thom, Thom, and Thom, "Stonehenge as a Possible Lunar Observatory" (n. 1
above).
37 Atkinson, "Aspects of the Archaeoastronomy of Stonehenge," p. 114.
38 Cecil Augustus Newham, "Aspects of the Archaeoastronomy of Stonehenge,"
(Shirenewton, Wales: Moon Publications, 1972), p. 23.
39 Atkinson, "Aspects of the Archaeoastronomy of Stonehenge," p. 107, and review of
Stonehenge and Its Mysteries by Michael Balfor, Archaeoastronomy 5, no. 2 (1982): 33.
40 Atkinson, "Moonshine over Stonehenge" (n. 2 above), pp. 213-14, and "Megalithic
Astronomy-a Prehistorian's Comments" (n. 22 above), p. 50.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 245

perceptively observed, "Had prehistoric man's primary intention been

to design an astronomical monument it is not likely that he would have
constructed a circle. A line of stones for an alignment or a horseshoe of
pillars for calendrical computation would have been more appro-
priate."4' Even the notion of a seasonal calendar is far-fetched. Colin
Burgess makes the obvious point that "the farmers of our [late
Neolithic] period could have observed all they needed to in the heavens
without going to the lengths of setting up stone circles." Thus, he
concludes, "If celestial observation is involved then it is likely to have
been for magico-religious rather than practical reasons."42
Perhaps the most misleading aspect of archeoastronomy in the
British Isles has been the search for scientific precision. The "discovery"
of known alignments in megalithic sites has given the impression of
remarkable astronomical knowledge on the part of the late Neolithic
builders, whereas in most cases the precision involved derives funda-
mentally from the investigator who knows the relevant alignments in
advance and "finds" them in the site. Ruggles has shown that this
characteristic of Thom's work has led to the mistaken view that
megalithic alignments were intentionally precise. This view of Neolit
astronomy as extraordinary has also led to speculation by the arc
ologist Euan MacKie about the existence of an elite, full-time pro
sional priesthood, as an explanation of Thom's results.43 Howev
Burl has noted that, while individual sites sometimes contain e
alignments, others in the vicinity do not. Following Burl's sugges
regarding stone circles and megalithic tombs, it appears that we shou
look for alignments at Stonehenge that are not astronomically specif
and exact but are symbolic and general and linked fundamentally
ritual activity, perhaps concerning the dead.

HISTORY, ARCHITECTURE, AND SYMBOLIC ORIENTATION

The special quality of Stonehenge is of course its architectural form.

No other stone circle in Britain or Europe looks like Stonehenge with
its lintels "hanging" on top of giant stone pillars (fig. 6).44 Its uniqueness
also makes it difficult to interpret. But fortunately something is known
about the construction of Stonehenge during its approximately two
thousand years of use. Any interpretation must take this history into

41 Burl, Stone Circles of the British Isles (n. 18 above), p. 53.

42 Burgess (n. 9 above), p. 343.
43 MacKie, Science and Society in Prehistoric Britain (n. 16 above), pp. 22, 148-49; cf.
Euan W. MacKie, "Wise Men in Antiquity?" in Ruggles and Whittle, eds. (n. 9 above).
44 The second element of the word "Stonehenge" may have meant something "hanging"
or supported in the air (Oxford English Dictionary); hence "Stonehenge" may originally
have meant the "hanging stones," in reference to the lintels.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
246 Stonehenge

account and fit it into the history of Neolithic and early Bronze Age
Britain and of Wessex in particular.
As already noted, Stonehenge began as a modest structure. The
great stone edifice now standing at the site represents the third and last
phase of its history. Originally, Stonehenge (i.e., Stonehenge I) was
only roughly aligned to the summer solstice sunrise, and the initial
question to be asked does not concern the precision of this alignment
but the overall design and purpose of the site in relation to others in the
Wessex area.

STONEHENGE I, CIRCA 3100 B.C.

Richard Atkinson, the principal archaeologist of Stonehenge, has

carefully reconstructed the history of the monument on the basis of
excavations made by himself and others, and he has presented his
results in terms of calibrated Carbon-14 dating.45 According to Atkin-
son, the original site consisted of the circular ditch with internal bank,
the ring of fifty-six Aubrey Holes, the entrance causeway (with its
array of postholes), and stones D and E (now missing) and probably
the newly discovered stonehole (97) close to the Heel Stone (96). The
four so-called Station Stones (91-94; 92 and 94 are missing) and the
Heel Stone may have been erected near the end of this period or at the
beginning of period II (fig. 3).
Like other henges of Wessex, Stonehenge I is defined by a large ring
ditch with a circular embankment and an entrance causeway. However,
most of the large henges have more than one entrance and feature the
ditch inside the bank. In contrast Stonehenge is more like several
smaller henges in Wessex, five at Dorchester-on-Thames and one at
Maumbury, all within forty-five miles of Stonehenge. These late
Neolithic henges range from thirty-three feet to 196 feet in diameter,
and all have single entrances. They also have a number of special pits
spaced evenly around the perimeter. The largest has forty-four, the
smallest has eight. Like the Aubrey Holes, the pits at the Dorchester
henges I and XI form a ring inside the bank of the henge; at the other
henges the pits lie along the bottom of the ditch. Also like the Aubrey
Holes, the pits were dug out and filled in soon afterward. Those at
Dorchester I and XI contained deposits of human cremations added
later at higher levels, and some were accompanied by distinctive
Neolithic bone pins. The same type of burial sequence and cremation

45 R. J. C. Atkinson, Stonehenge (1956, rev. ed., Harmondsworth: Penguin Books,

1979), app. 1, 2. Over the years Atkinson has revised the published dates for Stonehenge
according to different calibration curves. I cite Atkinson's most recent publication,
"Aspects of the Archaeoastronomy of Stonehenge" (n. 2 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 247

deposits were found in most of the excavated Aubrey Hole pits, along
with several of the same bone pins. The Dorchester henges also
contained cremation deposits around the pits and in the ditches, as at
Stonehenge.46 Although only thirty-four of the fifty-six Aubrey Holes
have been excavated, most of them (twenty-seven) contained burials,
and none contained posts or stones.
Hawkins first put forward the idea that the fifty-six Aubrey Holes
were astronomically significant because fifty-six could be seen as a
multiple of the 18.61 year lunar cycle (roughly 19 + 19 + 18), which
would yield an eclipse calendar. By placing three black and three white
stones in different Aubrey Holes and by moving them around the
Aubrey circle, one hole each year, Hawkins proposed that the Aubrey
Holes could be used as an eclipse calculator to predict eclipses in years
when black and white stones fell together into the same holes.47
However, despite the improvements upon this theory by the Royal
Astronomer Sir Fred Hoyle,48 Hawkins's computer theory of the
Aubrey Holes has been shown to be inaccurate after a few cycles, and
Hoyle's theory has been shown to be impossible for the builders of
Stonehenge to have conceived.49 Hawkins also had to admit that, since
not more than half of the eclipses could be seen from Stonehenge, the
calculator could only indicate danger periods when one might occur.
When the number of Aubrey Holes at Stonehenge is considered in
relation to the diverse number of similar pits at the henges at Dor-
chester and Maumbury (8, 10, 13, 14, 44, 56), it is difficult to find any
astronomical pattern in this range of figures; and, as Burl has pointed
out, there is no archaeological reason for singling out Stonehenge from
the group.50 Indeed, it has been noted that cycles other than fifty-six
could have been used for eclipse prediction at Stonehenge.51 When the
pits from these henges are considered as a group, they do not suggest
astronomical relationships but, rather, a consistent form of burial
ritual, perhaps linking the henge to the netherworld of the spirits, as
Atkinson and Burl have proposed. Burl has also stated that the

46 R. J. C. Atkinson, C. M. Piggott, and N. K. Sandars, Excavations at Dorchester,

Oxon. (Oxford: Ashmolean Museum, 1951), chaps. 1, 5, 8; Atkinson, Stonehenge,
pp. 28-29, 156, 171-72.
47 Hawkins (with White) (n. 1 above), chap. 9; Gerald S. Hawkins, Beyond Stonehenge
(New York: Harper & Row, 1973), pp. 301-4.
48 Fred Hoyle, "Stonehenge-an Eclipse Predictor," Nature 211 (1966): 262-76, and
On Stonehenge (London: Heinemann Educational Books, 1977).
49 Heggie, Megalithic Science (n. 28 above), pp. 203-6; Krupp, ed. (n. 28 above),
p. 103.
50 Aubrey Burl, "Holes in the Argument," Archaeoastronomy 4, no. 4 (1981): 22.
51 R. Colton and R. C. Martin, "Eclipse Cycles and Eclipses at Stonehenge," Nature
213 (1967): 476-78, and "Eclipse Predicting at Stonehenge," Nature 221 (1969): 1011-12.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
248 Stonehenge

number of holes was probably arbitrary, but this seems unlikely. Their
careful positioning indicates a definite ritual purpose that could well be
related to kinship groups that built the henge. The evidence of group
efforts can be seen in the construction of the outer ditch. The Stone-
henge ditch, like most Neolithic henges, was not a continuous excava-
tion but a series of quarry holes dug by separate work teams, the holes
being joined together afterward to form a somewhat irregularly shaped
ditch. It has been noted that this technique may be a "form of work
organization appropriate to a segmentary society .. .[and that] the
contribution of each separate group may have been emphasized in the
final form of the monument."52 Thus it would seem reasonable to
suppose that the number of pits around the perimeter of the henge was
related to the number of groups that built it-the larger the henge, the
more groups involved, and the greater number of ritual pits. If the
Aubrey Holes at Stonehenge were made for the purpose of communi-
cating with the world of the dead, then it would be appropriate for
each kinship group involved in building the site to be related to one or
more of the ritual pits that served their interests, a point that Burl has
noted.53 Renfrew's view of the henges as ritual centers of local segmen-
tary societies lends support to this interpretation.
Indeed, Burl has pointed out that fifty-six earthen long barrows have
been discovered in the Salisbury Plain area, none further than eighteen
miles (a two-day walk) from Stonehenge, although he has dismissed
this figure as purely coincidental.54 Of course, it is impossible to know
the precise sociopolitical boundaries of the Salisbury Plain in Neolithic
times. It is also impossible to be certain about the number of long
barrows that were built in this region. For example, two long barrows
near Stonehenge, long since leveled by plowing, were only recently
discovered by aerial photography,55 and these were not included in the
list of barrows that Burl consulted. Nevertheless, the fifty-eight (or
more) barrows in question belong to two adjacent clusters of barrows
in Renfrew's hypothetical map of Neolithic Wessex, consisting of
twenty-nine (now thirty-one) and twenty-seven barrows, respectively.
The point here is that, if Renfrew is right in supposing that the large
henges in these clusters served as ritual and political centers of local
segmentary groups, then Stonehenge I may have been a special "henge
of the dead" serving both clusters at a time of wider sociopolitical
unity. This possibility, when taken together with other evidence,

52 Startin and Bradley (n. 12 above), p. 293.

53 Burl, Rites of the Gods (n. 2 above), p. 105.
54 Ibid., p. 105.
55 Atkinson, Stonehenge and Its Environs (n. 17 above), p. ix.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 249

strongly suggests (regardless of the exact numbers involved) that the

function of the Aubrey Holes was not fundamentally astronomical but
ritual and sociological in nature.
Along the circumference of the Aubrey circle lie the four Station
Stones (fig. 3, 91-94). Atkinson originally thought that the Stations
belonged to the last period of construction because their diagonals
cross at the center of the sarsen circle, but he subsequently revised his
opinion and has dated them from the end of period I or from the early
part of period II. All that can be definitely said is that they were set up
after the ditch, embankment, and Aubrey Holes because they are
positioned in relation to these earthworks. Whether they belong to late
period I or early period II, their layout and positioning have been
subject to much astronomical discussion. Both Hawkins and Newham
claimed that the Stations were used as sightlines and that they formed
a rectangular pattern.56 The short sides (93-94, 91-92) point toward
the midsummer sunrise and midwinter sunset; the long sides (91-94,
93-92) point toward the moonrise and moonset at the extreme points
of the moon's nineteen-year cycle (fig. 5). The rectangular relationship
between the solar and lunar alignments at this latitude is an unusual
phenomenon. If it had been significant to the builders of Stonehenge, it
has been argued, then this would explain why the monument was built
where it was.57 A difference of thirty miles north or south would
change the latitude enough to transform the Station rectangle into a
parallelogram. In 1978, after locating the exact positions of the missing
Station Stones (92 and 94), Atkinson was able to confirm that the
layout of the Stations was almost perfectly rectangular, with adjacent
sides and the diagonals forming two nearly perfect Pythagorean tri-
angles. Atkinson also confirmed that the short sides were aligned
roughly on the midsummer sunrise and the midwinter sunset and the
long sides on the most southerly rising and the most northerly setting
of the moon. However, in this connection Atkinson has pointed out
that given the actual horizon levels at Stonehenge the latitude of exact
rectangularity of solar and lunar alignments lies more than eighty miles
to the south, that is, somewhere in the English Channel! Thus it is
unlikely that both exact rectangularity and astronomical alignment
were a consideration.58 It is also clear, in any case, that the location of
Stonehenge had been fixed, probably because of the presence of other

56 Hawkins (with White) (n. 1 above), p. 107; C. A. Newham, "Stonehenge-a

Neolithic Observatory," Nature 211 (1966): 456-58.
57 G. Charriere, "Stonehenge: Rythmes architecturaux et orientation," Bulletin de la
Soci&et Prehistorique FranCaise 58 (1961): 276-79.
58 Atkinson, "Aspects of the Archaeoastronomy of Stonehenge" (n. 2 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
250 Stonehenge

monuments already in the area (the long barrows and the cursus), and
that the ditch, bank, and Aubrey Holes had been constructed long
before the Station Stones were erected.
Two of the Station Stones (94 and 92) are surrounded by circular
ditches, and this strongly suggests that they were used as solar fore-
sights. These ditches are similar to the one around the Heel Stone,
which is the principal solar foresight of the monument. The two other
Station Stones logically serve as backsights. Of course, the relatively
short distances between the Stations do not permit much astronomical
accuracy, which would indicate that accuracy was not a relevant
consideration of the builders. A more relevant consideration seems to
have been the positioning of the Stations among the burial pits of the
Aubrey circle. Here the Stations are set out in perfect symmetry.
Starting with Aubrey Holes 56 and 28, which lie along the midsummer
sunrise axis of the monument, the Stations are set out at a distance of
ten holes either side of the axis holes, with seven holes separating the
proximate pairs of stones (fig. 3). This arrangement places the Stations
symmetrically on either side of the solstice axis. The western pair
(93-94) is aligned with the summer solstice sunrise, and the eastern
pair (91-92) is aligned with the winter solstice sunset. Looking across
the axis, the northern pair (91-92) is aligned with the maximum winter
full moonset, the southern pair (93-92) with the maximum summer full
moonrise. Thus the location of the Stations among the Aubrey Holes is
precise enough to give the impression that the relationship between
them was an important consideration. This would also suggest that the
Aubrey Holes were visible and still in use when the Stations were laid
out and that the Stations therefore belonged to period I. The Stations
are evenly placed among the Aubrey Holes, and the astronomical
sightlines connecting them may be seen only from positions on the
Aubrey circle. Thus the Station Stones appear to link the solar and
lunar cycles to the burial pits of the Aubrey Hole setting, bringing the
dead into conjunction with the sun and the moon.
There is reason, however, to question whether the lunar alignments
were intentional. If the Station Stones were set up as solar alignments
parallel to the main axis, then the lunar alignments would follow
fortuitously because of the latitude of the site. Moreover, while the
solar alignments would repeat annually, the lunar alignments would
not repeat for nineteen yers. This is a very long ritual interval, over half
a life-time in the Neolithic period.59 Nevertheless, long-term ritual
cycles are known among contemporary nonliterate peoples. For exam-
ple, the Dogon of central Mali in west Africa celebrate a sixty-year

59 Burgess (n. 9 above), p. 162.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 251

cycle in connection with the passage of a generation of men.60 In

Neolithic times, when life expectancy for the majority of men was
twenty-five years, a nineteen-year lunar ceremony might well have had
the same significance as a rite of generational transition. Thus, while
the most obvious function of the Station Stones is solstitial, their lunar
orientations may also have played a role.
Standing beyond the Aubrey circle and outside the causeway
entrance is the celebrated Heel Stone (96). Much has been made of the
fact that the Heel Stone stands noticeably off center, to the right of the
causeway entrance and the main axis of the site. At the time of
Stonehenge I (ca. 3100 B.c.) the midsummer sun, when seen from the
center of the Aubrey circle, did not rise over the Heel Stone but to the
left of it by approximately 2y/2. Today, due to the tilt of the earth's axis
(and the forward tilt of the Heel Stone itself), the sun appears to rise
more directly over the Heel Stone. But in 3100 B.C. the sun rose to the
west of it by nearly five solar diameters, an obvious discrepancy.61
Recently, the imprint of a large stone (97) was found to the left of the
Heel Stone, and it has been suggested by M. W. Pitts that the missing
stone was the mate of the Heel Stone.62 If this stone was standing next
to the Heel Stone in period I, the sun would have risen on alignment
between the two stones, and they would have served as portals,
flanking the solstitial axis. Atkinson believes that the missing stone 97
may have been removed in period II and the Heel Stone erected at this
time.63 But it seems more likely that Stonehenge I originally had a pair
of outliers, stones 97 and Heel Stone (96), as flankers of the solstitial
axis. This arrangement is consistent with the existence of another pair
of flanker stones, holes D and E (fig. 3), which stood astride the

60 M. Griaule, Masques Dogon (Paris: Musee de l'Homme, 1938).

61 Krupp, ed. (n. 28 above), p. 122. Questions concerning the alignment of the Heel
Stone, the angle of its tilt, the height of the horizon and the viewer, and the part of the
sun that was observed have been dealt with at length by Hawkins (n. 1 above, pp. 62-64)
and R. J. C. Atkinson ("Some New Measurements on Stonehenge," Nature 275 [1978]:
50-52, "The Aubrey Holes and the Avenue," Archaeoastronomy 6 [1983]: 132-33,
and "Aspects of the Archaeoastronomy of Stonehenge"). I follow Atkinson's view that
the axis of the site is based upon the sun's upper limb at sunrise, or "first gleam," at the
time of the summer solstice.
62 M. W. Pitts, "Stones, Pits and Stonehenge," Nature 290 (1982): 46-47, and "The
Discovery of a New Stone at Stonehenge," Archaeoastronomy 4, no. 2 (1982): 17-21.
Interestingly, Hawkins agrees with this theory, despite his former claim that sunrise over
the Heel Stone was the primary alignment (Gerald S. Hawkins, Mindsteps to the
Cosmos [New York: Harper & Row, 1983], pp. 79-80). Burl also agrees that the original
alignment was formed by the Heel Stone and its mate, standing like the portal stones of a
megalithic tomb (Rites of the Gods [n. 1 above], p. 104). Alternatively, the Heel Stone
and its mate may have served as the entrance pillars of a large ceremonial building, like
the one at Durrington Walls (fig. 2).
63 Atkinson, "Aspects of the Archaeoastronomy of Stonehenge" (n. 2 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
252 Stonehenge

causeway entrance. This arrangement would also mean that the monu-
ment originally had two axes, the causeway axis for pedestrian traffic
and the solstice axis for the midsummer sunrise. A similar configuration
is found at nearby Woodhenge, where that causeway axis of the
monument is oriented slightly to the left (west) of the summer solstice
sunrise axis of the building (fig. 4). When the mate to the Heel Stone
was removed, the Heel Stone was left standing to the right of the
solstice alignment. The result of this offset was that each year the sun
rose over the Heel Stone nine or ten days before the solstice, then rose
to the west and appeared to stop for three or four days, and afterward
returned to rise over the Heel Stone a few days later. Hoyle has
indicated that this situation would actually allow for an exact deter-
mination of the solstice date by simple interpolation, but this type of
positive offset is not present at enough alignments at Stonehenge to be
significant as a regular astronomical procedure. Hawkins noticed,
however, that, whenever the full moon rises over the Heel Stone at the
time of the winter solstice, an eclipse of the moon follows because the
sun and the moon are then on the same path. In this case the moon
rises 1 /2? to the right of the solstitial line, due to the nearness of the
moon to the earth, and thus it rises exactly over the top of the Heel
Stone. But only half of the eclipses indicated in this fashion would be
visible from Stonehenge, so the Heel Stone would not be a reliable
indicator, only a crude danger signal, as Hawkins noted. Lunar
eclipses that occurred at other times of the year would not, however, be
signaled in this way at all; and there is no reason to assume that the
builders would have been interested only in solstitial eclipses, as
Hawkins assumes. The fact is that any alignment that is solstitial will
automatically serve as a full moon eclipse signal because eclipses occur
only when the sun and moon are in the same plane. Since it is agreed
that the alignment of the causeway and Heel Stone is fundamentally
solstitial, then the lunar eclipse function of the Heel Stone is essentially
fortuitous.64
There is, however, another possible explanation. The Heel Stone
may have been left in its offset position to indicate the beginning and
ending of a ritual period around the time of the summer solstice.
During this period the sun would have passed over the Heel Stone for
several days before and after the solstice. Of course, there is no evidence

64 Discussion of eclipse prediction at Stonehenge may be found in Hawkins (with

White), Stonehenge Decoded (n. 1 above), pp. 138-39; and J. H. Robinson, "Sunrise
and Moonrise at Stonehenge," Nature 225 (1970): 1236-37, and "The Solstice Eclipses of
Stonehenge II," Archaeoastronomy 6, nos. 1-4 (1983): 124-31; Colton and Martin,
"Eclipse Predicting at Stonehenge" (n. 51 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 253

for this theory, except that the eccentric position of the Heel Stone was
kept throughout later periods when other stones were removed and
shifted around, thus indicating that its position was intentional and,
undoubtedly, functional as well. By contrast, the Station Stones are
not aligned with a positive offset; hence the ritual offset hypothesis
applies only to the Heel Stone of this period. This arrangement seems
appropriate because the summer solstice axis is clearly the primary
orientation of the monument and the one that would theoretically
receive the most ritual attention, perhaps lasting for a period of several
days. A similar solstice "interval" exists at the magnificent passage
grave at Newgrange, Ireland, constructed in about 3200 B.C. The "roof
box" over the entrance was designed so that the midwinter sunrise
shines down the tomb's passage at the time of the solstice and about a
week before and a week after the event.65
To the left of the Heel Stone there is a row of four postholes three
feet in diameter and four feet deep (fig. 3, postholes A). Atkinson
suggests that these holes may have held the uprights of a triple gateway
into the monument, since a line from the center of Stonehenge I
through the center of the entrance causeway bisects this row and is
perpendicular to it. This interpretation seems appropriate, especially in
light of the timber entrance structures at nearby Durrington Walls.
The location of this gateway to the left of the solstice portal stones
appears to support the idea of there being two axes, one solar and one
pedestrian, leading into the monument.
In front of the entrance causeway there are fifty-four small postholes,
about a foot in diameter, set in a somewhat regular pattern four to five
feet apart, forming rows running both across and parallel to the axis of
the causeway (fig. 3). The posts that once stood in these holes filled the
area between the two axes of the monument and obviously restricted
access to the enclosure. Compared to the entrance fences at Durrington
Walls, these postholes appear too numerous and haphazardly placed
to form a restrictive fence or palisade. They have been interpreted
astronomically, but not convincingly, for the postholes are too irregu-
larly placed for astronomical purposes.66 It is clear, however, that the
causeway's orientation toward both the summer solstice sunrise and
the winter moonrise standstill means that the cluster of posts that
stood in these holes received the first light of the rising sun and of the

65 J. Patrick, "Midwinter Sunrise at Newgrange," Nature 249 (1974): 517-19. See also
Michael J. O'Kelly, Newgrange: Archaeology, Art and Legend (London: Thames &
Hudson, 1982), pp. 123-24.
66 Atkinson, "Megalithic Astronomy-a Prehistorian's Comments" (n. 22 above), and
"Aspects of the Archaeoastronomy of Stonehenge" (n. 2 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
254 Stonehenge

full moon at the time of their longest duration in the sky. The posts
might, therefore, have served a symbolic purpose, perhaps holding
aloft the clan insignia or ancestor symbols to the light of the rising sun
and moon at their strongest moments. Clearly, the people who used the
site had to file past this array of posts to enter the enclosure, and it
would have been dramatically appropriate to mark the entrance with a
facade of posts bearing the groups' lineage symbols, facing the direction
of major solar and lunar events. Although there is no evidence for this
interpretation, it is at least consistent with the sociological features of
the site. It is perhaps coincidental that the number of causeway posts
(fifty-three) approximates the number of cremation deposits that have
been found at the site (fifty-five) and the number of Aubrey Holes
(fifty-six), and there may be a relationship here. Whatever it is, the
prominent position of these postholes in front of the entrance to the
site seems to indicate that a sociological interpretation is appropriate.

STONEHENGE II, CIRCA 2150 B.C.

Succeeding generations used Stonehenge I for about 700 years until

there was a desire for a more elaborate monument, involving far
greater efforts than the original one. The centerpiece of this project was
a double circle of tall, spotted dolerite stones, called "bluestones"
because of their bluish tint. The sole source of these stones is the
Prescelly Mountains in Wales, some 240 miles over mountain, vall
sea, river, and plain to Stonehenge. No less than eighty stones, so
weighing up to four tons, were transported mostly by raft or dugou
canoe over sea and river. There is evidence that these stones may have
been in use at some other site in Wessex several centuries before they
were taken to Stonehenge II. The new plan, based on these stone
called for the widening of the entrance causeway by twenty-six feet
the eastern side, the making of an entrance Avenue, marked by paral
ditches (580 yards long), and the aligning of the Avenue on
midsummer sunrise. Stone 97 was removed and its mate, the He
Stone, was encircled by a small ditch, as were the two Station ston
probably to indicate their special function as solar foresights. T
so-called Altar Stone, a micaceous, green-tinged stone, longer th
any of the bluestones, may also have been set up at this time on t
axial line opposite the entrance of the concentric circle.
Concentric circles, like the double bluestone circle, are rare in t
British Isles, and a group of them exists in the Wessex area. Burl
surmised that the most probable origin of the concentric idea i
Wessex is the circular timber building.67 Such buildings consisted

67 Burl, Stone Circles of the British Isles (n. 18 above), pp. 309-10. Burl also specula
that the original timber building at Stonehenge I may have been a "mortuary house

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 255

concentric rings of posts supporting a conical roof. The concentric

stone setting of the Sanctuary, near Avebury, twenty miles north of
Stonehenge, is known to have replaced a multiple-ring, timber struc-
ture. The double bluestone circle at Stonehenge may therefore have
replaced a circular timber building, as Atkinson has proposed.
The important feature of this new stone-pillared construction was its
grand entranceway. Two parallel, in-lying rows of five pillars each
formed the entrance, whose axis was accurately aligned on the mid-
summer sunrise. The pillars formed a massive gateway through which
the first light of the sun shone precisely at the time of the summer
solstice. The axis of this alignment extended outward through the
entrance of the circle, across the eastern side of the causeway, along the
center of the Avenue, and past the left side of the Heel Stone. Although
Atkinson has suggested that the Avenue was built as a processional
way for hauling the bluestones into the enclosure, the fact that the
Avenue architecturally extends the solstice axis of the monument
toward the horizon may also have been a consideration. Two large
stones were also placed along the center line of the Avenue (in holes B
and C), thereby marking the alignment of the Avenue with the solstice
sunrise and the entrance of the bluestone circle.
At this stage the intent of the builders seems sufficiently clear, even
though they never quite finished the circle. The new Avenue and the
new concentric circle were both centered upon the first gleam of the
summer solstice sunrise. When this occurred, the first rays of the sun
would have shown through the multipillared entrance to the tall Altar
Stone and reflected off the tiny fragments of mica covering its smoothed
surface. Atkinson reports some "slight and tantalizing" evidence for an
inner stone setting of an undetermined plan inside the bluestone circle.
Whatever this might be, there is no architectural clue to the rituals
performed at the summer solstice except their obvious coordination
with the sunrise. Clearly, the stone-pillared "building" was designed to
give dramatic effect to this event. As the first gleam of sunlight
appeared to the left of the Heel Stone, it passed through the entrance
of the double circle and hit the Altar Stone and illuminated its mica-
flecked surface. As dawn brightened, in an epoch that was less cloudy
and more temperate than today's, the sun climbed upward at an angle
to the right toward the Heel Stone and eventually passed high over the
top. Thus, at midsummer it could be said that the sun was first seen to
enter the "building" or "house" and then to rise majestically into the
sky over the Heel Stone.

the doorway open to the northeast and the midsummer sunrise that shone on the
decaying corpses inside" (Rites of the Gods [n. 2 above], p. 102).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
256 Stonehenge

The same effect was also apparent at Woodhenge, two miles away
next to Durrington Walls. Carbon-14 dating indicates that Woodhenge
was built about 2290-2190 B.C., a little before or perhaps simulta-
neously with Stonehenge II. It consists of six concentric rings of
postholes, set out in a somewhat oval plan, surrounded by an earthen
henge, with a causeway entrance opening to the northeast (fig. 4). The
causeway entrance is oriented slightly to the west of the solstice line,
as at Stonehenge, while the axis of the oval post structure is aligned
exactly upon the midsummer sun at first gleam, like the bluestone
circle in Stonehenge II.68 But here the entrance is not on the same line
and corresponds with the causeway gap in the earthwork. This means
that the building at Woodhenge had two entrances: one for pedestrian
traffic through the causeway, the other for the midsummer sunrise
through a narrow gap in the outside wall of the structure. It is assumed
that Woodhenge was partly roofed over and that the central ellipse
formed an inner court (170 feet across) that was open to the sky. If the
building was roofed in this fashion, the first rays of the midsummer sun
would have shone dramatically through a gleaming gap in the outer
wall into the central oval of the building. Woodhenge, then, may have
served as a model for Stonehenge II or vice versa. A final point of
similarity is the grave burials on the solstice line at both monuments.
The single grave at Woodhenge contained the skeleton of a child, about
three and a half years old, with its skull split in half before burial. The
single grave on the axis at Stonehenge was disturbed in antiquity and
the bones were mixed up; it contained objects of various dates, includ-
ing Bronze Age (Beaker) pottery, that link it to Stonehenge II or III.
The relatively large quantity of animal bones, flint flakes, and
pottery found at Woodhenge, mainly in the postholes, has been taken
to indicate its domestic character. In this connection MacKie has
suggested that the building served as a residential and ceremonial
structure for the members of a priesthood associated with the large
building at Durrington Walls.69 By contrast, the absence of any food or
other domestic debris at Stonehenge indicates that it was never a place
of habitation. So striking is the lack of refuse at the site (except for a
few deliberately buried antler picks, stone mauls, and pieces of pottery)
that it is possible to assume that the site was carefully scoured of refuse
throughout the centuries of its use. This might signify that Stonehenge
was a place of extreme sanctity. The single entrance of the bluestone
circle aligned to the solstice sunrise (along the widened causeway and

68 Cunnington (n. 16 above); Thom, Megalithic Sites in Britain (n. 11 above),

pp. 73-74.
69 MacKie, Science and Society in Prehistoric Britain (n. 43 above), p. 175.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 257

new Avenue) indicates that the circle may have been intended exclu-
sively for ritual purposes, perhaps as a great "house" of solar forces.

STONEHENGE III, 2000-1500 B.C.

The double bluestone circle was dismantled before it was completed in

order to make way for the great sarsen structure. This involved the
hauling (on rollers or sledges) of no less than eighty-one enormous
sandstone stones, called "sarsens,"70 averaging eighteen feet and
weighing twenty-six tons, a distance of twenty miles from the Marl-
borough Downs near Avebury. This phase of construction occurred in
three basic stages: (a) the erection of the five sarsen trilithons, the outer
sarsen circle, and the "Slaughter Stone" and its companion (the
missing stone in hole E); (b) the digging of the Y and Z holes, and the
erection of an elliptical bluestone setting of twenty-two stones within
the sarsen horseshoe (later dismantled); (c) the erection of the bluestone
circle, consisting of approximately sixty stones, inside the sarsen circle,
and the bluestone horseshoe of nineteen stones within the trilithon
horseshoe (fig. 6). In circa 1100 B.C. (Atkinson's period IV) the Avenue
was extended about four miles to the bank of the Avon River (fig. 1).
This represents the last datable work associated with the monument.
The new sarsen structure can be seen as an evolution of the ideas
behind the double bluestone circle. Although the sarsen structure i
unique, it includes some of the elements of the bluestone concentric
the concentric plan, the solstice axis, the building-like appearance, and
approximately the same number of stones. If the double bluestone
circle was intended to represent a building-like structure, this concep-
tion was magnificently realized in the new sarsen edifice. As Atkinson
points out, the woodworking methods used on the sarsen stone
indicate that "the sarsen structure must surely be related in some way
to the framework of a circular wooden building" (fig. 4).71 All th
evidence points to the fact that the intended form and visual impact of
the monument was architectural, not astronomical.
First it should be noted how unsuitable the plan was for precise
astronomical observation. We have already seen that the sightlines
within the monument are too short for astronomical accuracy. Recog-
nition of this fact led Thom to look for possible distant foresights
none of which have proven to be related to the monument. The circular
shape is also unnecessary; indeed, it is confusing, for it contains to
many stones and too many possible sightlines. The sarsen stones are, of

70 The word "sarsen" is usually regarded as a corruption of "Saracen," meaning

foreigner (Atkinson, Stonehenge [n. 45 above], p. 36).
71 Atkinson, Stonehenge, p. 177.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
258 Stonehenge

course, too tall to be seen over, so the solar and lunar risings and
settings have to be seen through the gaps between them, and these are
too broad for astronomical precision. The placing of the lintels over the
sarsens and the careful (and laborious) leveling of them was also
unnecessary for observational purposes,72 as was the dressing and
smoothing of the inside of the sarsens. For astronomical purposes, a
far more suitable plan would have been the construction of a few long
rows of widely spaced, medium-size stones radiating outward from a
central hub and aimed at various points on the horizon. Such a plan
would also have been far easier to build.
Instead, Stonehenge III expresses the qualities of architecture, and it
expresses them so magnificently that it was supposed to have been built
by a representative of Mycenean civilization. The architectural vision
of Stonehenge III makes sense only if it was intended to imitate in
imperishable stone a great timber house with lintels serving as the roof
plates. The attachment of the lintels to the uprights by mortise and
tenon technique, the joining of the lintels to each other by tongue and
groove method, the shaping of the lintels to fit the curve of the circle,
and the leveling of the lintels and the uprights, all imitate methods that
were presumably used on the great timber buildings whose remains
have only recently come to light (fig. 8). Finally, the deliberate convex
shaping of the sarsen uprights so that their sides would appear from a
distance to be straight73 (a method known as entasis from the columns
of Greek temples) indicates a keen desire for the structure to appear to
be a massive, straight-columned building (fig. 6).
The intent to create a building-like appearance may also account for
the unexplained digging of the concentric rings of Y and Z holes
outside the sarsen circle. These rings contain thirty holes (one hole is
missing from the unfinished Z ring), each placed in line with the thirty
sarsen uprights. These two rings complete the six rings of Stonehenge,
a pattern that resembles the six-ring posthole configuration of the
buildings at Durrington Walls and Woodhenge (figs. 2, 4). Indeed, the
irregularity of the Y and Z rings at Stonehenge resembles the irregu-
larity of the two outer rings of postholes at Woodhenge; in both cases

72 It has been claimed that the lintels served as a "walk-way" and that certain
depressions in the tops were made to hold markers for the purpose of lunar sightings
(R. F. Brinckerhoff, "Astronomically Oriented Markings on Stonehenge," Nature 263
[1976]: 465). But natural weathering has produced many regular-looking holes in the
tops of the lintels, and the proposed sightlines require the central trilithon to have been
partly fallen in prehistoric times (as it is today), which is unlikely (R. J. C. Atkinson,
"Interpreting Stonehenge," Nature 265 [1977]: 11, and "Aspects of the Archaeoastronomy
of Stonehenge," pp. 113-14; see a photograph in Hawkins, Mindsteps to the Cosmos
[n. 62 above, p. 84], which shows the lintels, markers, and holes).
73 Atkinson, Stonehenge (n. 16 above), p. 37.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 259

/1-11"- --_ Dished slightly

Tenon

UPRIGHT

FIG. 8. -Joints connecting lintels

Thom, and A. S. Thom, "Stonehen
5 (1974): 73, fig. 3. (Science Histo
Lane, Chalfont St. Giles, Bucks,

the interior structures were er

measurement from the center
six-ring design, some of the s
henge II should have been pla
great sarsen circle. Such a plan
Z holes were excavated, until
abandoned the project. None
and Z holes; instead, token ch
bottom of the holes. What m
placing the bluestone circles
obscured the impressive mass
umns. Structurally of course th
Stonehenge was not intended to
placed inside instead, around
Approximately sixty blueston
within the sarsen circle, and
pattern inside the trilithon h
tained six concentric rings lik
Woodhenge. As M. E. Cunnin
stones at Stonehenge closely ma

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
260 Stonehenge

four inner rings of posts at Woodhenge.74 The desire for a six-ring

structure at Stonehenge may therefore explain why the Y and Z holes
were dug, although only symbolic deposits of bluestones were placed in
them, in order to make up the two outer rings of the six-ring "building."
The structure was of course open to the sky, and the vault of the sky
served symbolically as its roof. The sky-roof idea was dramatically
conveyed by the unobscured wall-like face of the sarsen circle with its
lintel roof plates. Seen from a distance, Stonehenge stands like a
circular hub in the midst of the vast Salisbury Plain, which was as vast
and open 5,000 years ago as it is now. In this context the viewer's eyes
are naturally drawn to the monument because it is the primary point of
focus within the gently rolling downs. As Ralph Waldo Emerson put it,
"On the broad Downs, under the gray sky . . . nothing but Stonehenge
.. [was visible] in the wide expanse."75 The very setting implies that
there is some connection between the great stone circle on the plain
and the vault of sky above.
This brings Stonehenge into line with what we know about the
architecture of temples all over the world. Their purpose is to link
together different realms, the transcendent with the terrestrial, the
divine with the human, the living with the dead. The fact that Stone-
henge, like other temples, has an axial orientation toward a transcen-
dent symbol, in this case the sun, confirms its ritual significance. For
purely astronomical purposes, no alignment need be emphasized over
another. The careful smoothing of the inside surfaces of the sarsen
circle and the trilithon horseshoe indicates that the ritual center of the
monument was located inside. Further indication of ritual focus is the
fact that the trilithons increase in height toward the central trilithon
(III), which stands astride the solstice axis. Because this trilithon is
finished on both sides, Atkinson has suggested that the viewing point
of the summer solstice sunrise may have been behind it, so that the gap
between the trilithon uprights served as a great solar portal or archway
at the time of the sunrise. The fact that the winter solstice sunset occurs
in the reverse direction almost exactly on the same axis could not have
been missed, and it is assumed that this event was observed from the
other side of the great trilithon.
So far, Hawkins is the only scholar to have suggested that the other
trilithons (I, II, IV, V) also functioned as celestial archways (fig. 7); and
for this he has been severely criticized. The reason is partly because the
precise placement of some of the trilithons is unknown, partly due to a

74 Cunnington (n. 16 above), pp. 21, 104-5.

75 Ralph Waldo Emerson, English Traits, in The Complete Works of Ralph Waldo
Emerson (Boston: Houghton Mifflin Co., 1883), 5:276.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 261

few fallen, missing, or re-erected stones, but chiefly because the gaps in
the outer sarsen circle through which alignments occur are at least 8?
or sixteen solar diameters wide, a width that is astronomically useless.
However, from a symbolic and architectural perspective, the align-
ments are dramatic and obvious, and for ritual purposes the "errors"
are unimportant. The Hopi, for example, are content to work with a
winter solstice error of up to three days.76 The trilithon archways are
approximately aligned upon all the risings and settings of the sun at the
extreme points of its cycle and upon a few of the risings and settings of
the full moon at its extreme points. Like the central trilithon, the other
trilithons could therefore be said to link these significant celestial
events with the inner space of the monument. Moreover, as Hawkins
points out, the trilithons repeat all the alignments of the older Station
Stone setting.
However, it is questionable whether all the trilithon alignments were
actually intended. Thom's carefully made plan, based on his 1975
survey, shows that the main consideration in setting out the trilithons
was geometrical and solstitial (fig. 9).77 They were not placed in a circle
or a square but in an elliptical pattern whose axis corresponds with the
midsummer sunrise. As Atkinson indicates, the trilithons were set up
before the sarsen circle was in place, for they are much too large to
have been manuevered into position after the circle was built. This
means that the trilithon archways could have been set squarely toward
any intended astronomical target because there was no interfering
structure. However, only two of the archways face directly toward
astronomical positions. The archway of trilithon I faces the midwinter
sunrise, while the archway of trilithon III faces both the midsummer
sunrise and the midwinter sunset. The other archways could easily
have been aimed directly at other significant positions, such as the
midsummer sunset and certain lunar extremes, but they stand askew,
due to their position in the elliptical plan, and do not face the proposed
positions very exactly. As Thom's plan shows (and Atkinson agrees),
the elliptical character of the trilithon setting was deliberate, and it
resembles the elliptical arrangement of the large inner posts of Wood-
henge whose axis is similarly aligned. The northern and southern pairs
of trilithons (I and V, II and IV) stand symmetrically on either side of
the center of the sarsen circle (a line running perpendicular to and
bisecting the axis), while the great trilithon stands somewhat inside the

76 Moir, "Some Archaeological and Astronomical Objections to Scientific Astronomy

in British Prehistory" (n. 23 above), p. 230. See S. C. McCluskey, "The Astronomy of the
Hopi Indians," Journal of Historical Astronomy 8 (1977): 174-95.
77 Thom, Thom, and Thom, "Stonehenge" (n. 1 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
262 Stonehenge

\ ,, :> --"r.'- '-

/ / \ :./ /
-t -?/ ',%
. // I / , '. / / " \N
4s

" -o

w ' ?'
'.Aubrey cen
I s ; e I ? -- - ^ re
/ , '

\ ( " \ l / 7

f., ,

,\ '

i~" /

10 0 'O 20 30 t
t., . r t l I i

J Stone which has been re-erected

3 Stone which has never been disturbed :'"" Post hole or stone hole

j Stone which has been 'straightened' :.', .: Stone now below ground

FIG. 9.-The Thors' diagram: the geometry of Stonehenge IIIc. T

from A. Thom, A. S. Thom, and A. S. Thom, "Stonehenge," Journalfo
History of Astronomy 5 (1974): 75, fig. 5. (Science History Publication
Halfpenny Furze, Mill Lane, Chalfont St. Giles, Bucks, HP8 4NR, U.K

ellipse so that all the trilithons are approximately the same dis
from the center.
Hawkins has considered the alignments of the trilithons as a
standing structure and produced a table of results for solar and
extremes, showing a range of error from 0.1? to 2.9?. These alignm
run between the trilithons themselves. The alignments differ, how
in almost every instance from those Hawkins proposes for the tril

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 263

within the sarsen circle. This has caused him to wonder "whether the
construction of Stonehenge III was according to a single plan or
followed a series of second thoughts."78 In fact, most of the alignments
are blocked by the stones of the later sarsen circle. However, as
Atkinson points out, the sarsen circle had to have been measured and
laid out before the trilithons were erected. Thus the circle was planned
as an integral part of the structure, and it would be illogical to assume
that the builders deliberately blocked the sightlines originally estab-
lished. Moreover, if Hawkins's alignments for the trilithons as a free-
standing structure were to make sense, the trilithons should stand more
squarely opposite each other, instead of in their ellipitical arrangement,
and they should be angled more directly toward their foresights, which
they are not.
Atkinson indicates that, in order to avoid the accumulation of errors
of spacing due to the irregular widths of the sarsens, the centers of the
stones were set at equal intervals of ten and a half feet around the
circle. This meant that the width of the gaps varied unpredictably, and
the exact location of the gaps in the sarsen circle could not have been
known until the stones were set into position. Nevertheless, the builders
did achieve perfect symmetry in placing the sarsens opposite the
trilithons; hence the positional relation between the trilithons and the
outer sarsen circle is exactly symmetrical. The eastern pair of trilithons
(I and II) stand opposite the same sarsens as the western pair (IV and
V) on the other side of the axis, and the central trilithon stands
opposite the sarsens that stand on either side of the axis (fig. 9).
However, the trilithon alignments that Hawkins proposes are
asymmetrical, except for the central trilithon, and they occur through
different sarsen gaps in each instance, as Jacquetta Hawkes points out:
"The picture of all the trilithons being aligned through the peristyle
[the outer sarsen circle] on the maximum risings and settings sounds
impressive until you see that one is simply the opposite of the sunrise
alignment, while in the other four the positional relation between the
trilithon gap and peristyle gap is different in every alignment."79
Hawkins is also incorrect in saying that "your view is constricted by
the narrowness of the archways. You cannot look down lines which
would point to no meaningful sun or moon position; you are forced to
look through paired archways towards those inevitable sun-moon posi-
tions."80 It is, in fact, possible to sight from trilithon V through the gap

78 G. S. Hawkins, "Astronomical Alignments in Britain, Egypt, and Peru," Philo-

sophical Transactions of the Royal Society of London A 276 (1974): 162.
79 Hawkes, "God in the Machine" (n. 2 above), pp. 178-79.
80 Hawkins (with White), Stonehenge Decoded (n. I above), p. 116.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
264 Stonehenge

between stones 24 and 25 (both missing), although the azimuth of this

alignment points to no meaningful astronomical position (figs. 9, 7).
Nevertheless, it seems reasonable to suppose that the trilithon
alignments proposed by Hawkins were visible to the builders of
Stonehenge III. Hawkins's photographs in Stonehenge Decoded show
dramatic views of the midwinter sunrise through trilithon I and of the
midsummer sunrise and midwinter sunset next to the upright of
trilithon III. Surely, the midsummer sunset through trilithon V was
also noticed, although this occurred at an angle. Indeed, quite a few
sunsets would have been visible through trilithon V, both before and
after the summer solstice. The same is true of Hawkins's "moonrise"
and "moonset" trilithons (II and IV). Given the wide angle of vision
through these archways, a number of moonrises around the major and
minor standstill of the summer full moon would have been visible, as
would a number of moonsets around the major winter standstill. What
this situation implies, if anything, is unclear. These alignments are too
general for astronomical purposes, and while some or all of them may
have been important symbolically, it is just as likely that they were not.
All the indications are that the five trilithons were erected primarily
for architectural and ritual reasons. They were set up in order to
produce the same elliptical arrangement as in the inner court at
Woodhenge, constructed about the same time and aligned on the
summer solstice axis. The other astronomical alignments are coinci-
dental, and some may in fact have been blocked by the stones of the
bluestone circle, which stood between the trilithons and the sarsen
circle. The only deliberate gaps in the bluestone circle occur near the
solstice axis, where some of the stones have been spaced further apart.
This means that the trilithon horseshoe and the inner bluestone
horseshoe served mainly an architectural function, defining the inner
ritual space of the monument around the solstice axis.
Possible indications of the lunar significance of the monument may
lie in the number of uprights in the sarsen circle and the number of
bluestones in the bluestone horseshoe. Newham has pointed out that
the thirty sarsen circle stones might represent the number of days of the
synodic month, the period from, say, full moon to full moon, which is
actually twenty-nine and one-half days. By placing a marker next to a
stone each day a tally could be kept, and it would be possible to know
the number of days since the last full moon or the days before the next
one was due. This is certainly consistent with the fact that the monthly
lunar phases were important phenomena to most primitive and ancient
societies, although the lunar phases are usually identified by simple
observation. Newham has also suggested that the nineteen stones in the

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 265

bluestone circle might stand for the nineteen-year lunar cycle and
have served as a means of keeping track of it.81 However, it is well to
keep in mind the results of Burl's study of over three hundred stone
circles, namely, that the number of stones in the circles does not seem
to be related to astronomical calculation of any kind.82

A RITUAL INTERPRETATION

In 1957 Atkinson wrote that "one thing upon which it is agr

it [Stonehenge] is primarily a 'temple', a structure in w
possible for man to establish contact and communication
mundane forces or beings."83 Yet, apart from his comments
henge I, Atkinson has had little to say about the ritual signi
the monument, especially Stonehenge III, which he beli
unique so that it is "hopeless to interpret it in terms o
ritual."84 From 1964 onward the study of the astronomical p
of Stonehenge obscured the question of its ritual significanc
the temple-like features that once seemed so obvious. Th
discussion has laid the foundations for looking again at St
a ritual monument.
It also needs to be noted that a ritual interpretation is not entirely a
hypothetical matter and that there are some sound reasons for thinking
that Stonehenge was in fact a ritual center. First, the solar orientation
of the site conforms to the common practice of orienting ritual sites
toward transcendent powers. Second, the cremation deposits in the
Aubrey Holes indicate that Stonehenge I was originally a place where
relationships with the dead were important. Third, the relatively few
henges with outside ditches, like Stonehenge, are regarded as cere-
monial monuments. Fourth, the structures of Stonehenge I, II, and III
are all nonpractical, that is, there is no evidence to indicate that they
served any domestic, military, agricultural, or economic purpose.
Fifth, the lack of any domestic debris is unusual and indicative of the
sanctity of the site, as is the fact that no prehistoric agriculture was ever
practiced around it.85 Sixth, the estimated 30 million man-hours
required to build Stonehenge III suggests that it was a structure of the
highest value, of "ultimate concern," to the society as a whole. Seventh,
the fact that the largest concentration of late Neolithic and early
Bronze Age round barrows in Wessex lies in the area of Stonehenge is

81 Newham, The Astronomical Significance of Stonehenge (n. 38 above), pp. 29-30.

82 Burl, Stone Circles of the British Isles (n. 18 above), pp. 374-75.
83 Atkinson, Stonehenge (n. 45 above), pp. 169-70.
84 Ibid., p. 177.
85 Bowen, "Pattern and Interpretation" (n. 11 above), p. 45.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
266 Stonehenge

suggestive of a ritual relationship with the dead. Finally, the stones

themselves are indicative of a ritual purpose, for in Wessex and the rest
of Neolithic and early Bronze Age Britain stone was used only for
ritual structures, especially for those related to the dead.
In considering the ritual function of the site, it will be useful to take
up previous theories: first, that it was a Druid temple; second, a temple
to the sun god; and third, a memorial to the dead.
In the seventeenth century John Aubrey suggested that Stonehenge
might have been built by the Druids. Given the state of archaeological
knowledge at the time, this was a reasonable idea, more reasonable
than that of Aubrey's contemporary, the architect Inigo Jones, who
thought Stonehenge to be a Roman temple built according to the
Tuscan order. Aubrey's Druid-temple notion was later taken up and
expanded by the eighteenth-century antiquarian William Stukeley,
and it has been repeated by later writers and remained a popular
theory until today. However, the Celtic society in which the Druid
priests flourished came into existence in Britain only after 300 B.C. So it
could not have been built by the Druids, and there is no evidence that
the Druids, who worshiped in forest groves, took any interest in it.86
Students of Stonehenge have also been intrigued, since the mid-
eighteenth century, with a passage in the writings of Diodorus of
Sicily, a Greek writer of the first century B.C. In his Library of History
Diodorus refers to the History of the Hyperboreans by the Greek
historian Hecataeus of Abdera, who lived in the fourth century B.C.
"Hecataeus and certain others," writes Diodorus, "say that in the
regions beyond the lands of the Celts [i.e., Gaul] there lies in the ocean
an island .... The following legend is told concerning it: Leto [mother
of Apollo] was born on this island, and for that reason Apollo is
honoured among them above all other Gods.... And there is also on
the island both a magnificent sacred precinct of Apollo and a notable
temple which is adorned with many votive offerings and is spherical in
shape." It is possible, but not certain, that the island was Britain and
that the temple was Stonehenge. Diodorus goes on to write that

they say also that the moon, as viewed from this island, appears to be but a
little distance from the earth and to have upon it prominences, like those of the
earth, which are visible to the eye. The account is also given that the [sun,
moon?] god visits the island every nineteen years, the period in which the
return of the stars to the same place in the heavens is accomplished; and for
this reason the nineteen-year period is called by the Greeks the "year of
Meton". At the time of the appearance of the god he both plays on the cithara

86 Atkinson, Stonehenge, p. 180.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 267

and dances continuously the night through from the vernal equinox until the
rising of the Pleiades.87

Hawkins has made much of this passage because of its reference to

the Metonic cycle, the nineteen-year lunar cycle discovered in the fifth
century B.C. by the Greek astronomer Meton. Hawkins thought that
this might refer to eclipses of the moon or the sun that occur in
approximately a nineteen-year cycle. However, only half of these
eclipses are visible from Stonehenge, and Hawkins's method of predict-
ing them by using the Aubrey circle as a calculator has proven
unacceptable. As Hoyle himself indicated, such a method would show
only how we could have gone about the problem.
Nevertheless, if Hecataeus's account, as reported by Diodorus, refers
to Stonehenge, we might assume that it was regarded in the fourth
century B.C. as a temple of the sun god. This, however, would be 800
years after the last available date for Stonehenge, and belief and
practice would undoubtedly have changed by then. It is also unlikely
that Stonehenge was functioning much after 1100 B.C., for Wessex
society changed completely after 1400 B.C. and the whole area went
into decline. Hecataeus's account also says nothing about the obvious
summer solstice cycle and refers only to a spring ceremony held in
connection with the equinox. It is therefore difficult to accept the
account as reliable. Indeed, Hecataeus's entire History of the Hyper-
boreans, which is no longer extant, is generally regarded by classical
scholars as a fictional portrayal of a utopian society,88 and Diodorus
himself refers to Hecataeus's accounts of the Hyperboreans ("people
beyond the north winds") as legends or myths (mythologia). Hence, it
is questionable whether Hecataeus's description of the temple of
Apollo had any historical foundation.
Finally, Stonehenge III has been regarded as a memorial to the
dead. This, in fact, is the earliest interpretation on record. In about
1136 Geoffrey of Monmouth wrote that Stonehenge was built by
Aurelius Ambrosius, King of the Britons, with the help of Merlin the
magician, as a memorial to 460 British nobles, slain by a Saxon chief.89
Eliade's theory, mentioned at the beginning of this article, that "mega-
lithicism" is related to ancestor worship, is a current version of this
interpretation. One difficulty with this view is that the Aubrey Holes of

87 Diodorus Sicilus, II, Loeb Classical Library (Cambridge, Mass.: Harvard Uni-
versity Press, 1935), pp. 35-41.
88 Paulys Realencyclopadie der Classischen Altertumswissenschaft (1912; reprint,
Stuttgart: Alfred Druckenmuller Verlag, 1970), 8, pt. 2:2756.
89 L. V. Grinsell, Legendary History and Folklore of Stonehenge, West Country
Folklore, no. 9 (Guernsey: Toucan Press, 1975).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
268 Stonehenge

period I had been abandoned by periods II and III, and there are only
two rather meager Bronze Age burials associated with the bluestone
and sarsen structures. One of these is an inhumation recently found in
a pit dug into the ditch and associated with late period I or possibly
with period II.90 The skeleton was that of a young man, with one of the
arrow heads found among the bones still stuck into a rib. The other
burial is one that lies across the axis, as at Woodhenge, which may
have been a dedicatory deposit associated with the construction of the
stone circles of periods II or III. Another problem is that the funerary
monuments built during the period of Stonehenge II and III were
round barrows, not stone circles. Thus Stonehenge III cannot be
regarded as a "tomb" associated with the "dying" midwinter sun god91
or with the Bronze Age burials around it, for the form of Stonehenge
III is not sepulchral. Rather, as I have tried to show, Stonehenge III
was constructed to look like one of the great ceremonial buildings of
the period, like those at nearby Woodhenge and Durrington Walls.
However, Eliade and other scholars have noted that Stonehenge lies
in a "field of funeral barrows," which implies some sort of association
between the monument and these tombs. The barrows in the area were
built between about 2000 B.C. and 1400 B.C., the same period as
Stonehenge II and III. According to archaeologist Paul Ashbee,
Stonehenge II was "most probably the product of Wessex chieftains
whose barrows throng its vicinity.... It is thus hardly unlikely that
they are unconnected, and they should be considered as a group or
complex, rather than as a number of isolated monuments, as they have
been up to now."92 But, on the face of it, there appears to be nothing
except a general spatial relationship between them, such as that
between a church and the graves that surround it in a church yard; no
specific ritual linkage is apparent. There are, of course, other clusters
of round barrows in Wessex, so that the area around Stonehenge is not
unique in this respect, although it does have the highest concentration
of them. To be sure, most of the barrows lie on elevated ground within
sight of the monument, but many also appear to be intentionally
clustered around several Neolithic long barrows in the vicinity that
predate Stonehenge I. Given the number of early Bronze Age "Celtic"

90 R. J. C. Atkinson, "Recent Excavations at Stonehenge," Antiquity 52 (1978): 235-36,

and Stonehenge (n. 45 above), p. 215; Burgess (n. 9 above), p. 330; Christopher
Chippendale, Stonehenge Complete (Ithaca, N.Y.: Cornell University Press, 1983),
p. 215. The head of the skeleton was lying toward the north.
91 R. S. Newall, Stonehenge (1959; reprint, Edinburgh: H.M.S.O., 1967), pp. 15-16.
A similar interpretation appears in the 1977 edition of Newall's Stonehenge, pp. 21-22.
92 Paul Ashbee, The Bronze Age Round Barrow in Britain (London: Pheonix House,
1960), p. 133. Compare Harrison (n. 9 above), p. 96.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 269

fields in the immediate area, it is apparent that the round barrows

would have stood within sight of the settlements of the period, and this
might also account for their presence. Therefore, on the surface at
least, the evidence for a specific ritual relationship between the barrows
and Stonehenge II and III appears to be crucially vague.
However, additional evidence that is highly suggestive lies beneath
the ground. Chippings of bluestone and of sarsen have been found
in the material of some of the excavated barrows.93 While this evi-
dence is not conclusive (the chips could have come from stones other
than those at Stonehenge, though in the case of the bluestone chips this
is unlikely), the presence of these distinctive stone chips greatly
increases the likelihood that there was a direct ritual relationship
between the monument and the barrows. Hence, the appropriateness
of the observation that Stonehenge "has attracted these barrow-groups
like a magnet."94
The solution to this problem, I believe, lies in the history of the
monument, that is, in its different configurations and in their relation
to the changing sociopolitical circumstances and burial practices of
late Neolithic and early Bronze Age Wessex. On the assumption that
ritual function relates to architectural form and that the history of the
site is consistent, the several configurations of Stonehenge do, in fact,
suggest a single, underlying ritual function: the periodic coordination
of solar and (possibly) lunar cycles with the remains of the dead. This, I
believe, is the fundamental ritual theme uniting the different phases of
the monument's history.
As we have seen, Stonehenge I consisted basically of the ditch, bank,
Aubrey Holes, the Heel Stone and its mate, and the Station Stones.
Like most of the Neolithic monuments in Wessex, Stonehenge I was
related to the dead. Architecturally, its primary function was to enclose
a ring of burial pits. Atkinson has emphasized the fundamentally
symbolic nature of these ritual pits, indicating that they were made to
sanctify the site and to link it to the underworld of the dead. As in the
case of the long barrows, which contain the bone deposits of many
generations, it is now thought that Neolithic burial sites served a
general ritual purpose, perhaps as the locus for a range of religious
rituals relating the community as a whole to its ancestors in addition to
the recently dead. In the case of Stonehenge I it is obvious that
something more than funerary ritual was going on. First of all, the
burial deposits in the pits consisted of only small amounts of cremated

93 L. V. Grinsell, Stonehenge Barrow Groups (Salisbury: Salisbury & South Wiltshire

Museum, 1978), p. 14.
94 Ibid., p. 5.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
270 Stonehenge

remains, implying that they were dedicatory or token representations.

Second, they were laid out around the solar axis of the site, and the
solar and (possibly) lunar Station Stones were symmetrically placed
among the Aubrey Hole pits and the burial deposits. From this
arrangement the general nature of the ritual can be inferred: it was an
attempt to coordinate the solar and (possibly) lunar cycles with the
spirits of the dead represented by the token cremations. It can also be
safely assumed that the rituals benefited the living community in some
important way, otherwise the collective efforts that went into building
Stonehenge I would have no justification.
To say more than this, however, is to be very speculative. We know
of course from nonliterate and ancient societies and from folk customs
in Europe and the British Isles, especially among the Celts (whose
ancestors may date from Neolithic times), that the sun was regarded as
a life-giving power. It would be reasonable to suppose, then, that the
sun was regarded as a sacred body whose apparent cyclical rhythm, its
seasonal strengthening and weakening, had a positive magical effect
upon the life of human beings. In Neolithic times the effective channel
of these solar (and possibly lunar) forces seems to have been the
superior realm of the dead. Most of the Neolithic monuments in
Britain were associated with the dead, including the long barrows in
Wessex, which are oriented toward the rising sun, within the 80? arc of
its summer and winter extremes.95 What this relationship was is
unknown, but the function of Stonehenge I seems to have been that of
periodically activating it, undoubtedly for the welfare of the living
community. The implicit periodicity of the rites indicates that their
virtue lay in repetition, hence in some sort of collective "renewing"
function.
Stonehenge I seems to have played this role for about 1,000 years
until major modifications were introduced. The assumption that the
site was in continuous use for this period of time is not justified by
anything except the fact that antler picks were found in the holes of the
new bluestone concentric dating 1,000 years after the beginning of
Stonehenge I. However, it is a reasonable assumption in the context of
other Wessex monuments, one example being the chambered long
mounds used by successive generations for 1,000 years.
Stonehenge II may or may not have represented a new departure in
design, but it was a new departure in materials. If a circular timber
sanctuary did stand at the center of Stonehenge I, the building of the
double bluestone circle was an attempt to reproduce this building in
permanent materials, using a type of stone that had never been seen at
Stonehenge before. The bluestones may also have been regarded as
95 Burl, "'By the Light of the Cinerary Moon'" (n. 27 above), p. 245.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 271

sacred stones, coming from the Prescelly Mountains, whose cloud-

capped summit may have been looked upon as a sacred realm, as
Atkinson supposes.96 The new concentric plan was intended to focus
the whole monument upon the midsummer sunrise, including the
entrance causeway, which had formerly been oriented to the left of the
solar axis. The new Avenue extended this alignment toward the
horizon by about 580 yards and provided a long processional way into
the site along the sunrise path. The multipillared entrance to the circle
also gave the impression that the first light of the midsummer sun was
entering a house along a sacred pathway. The accuracy of this align-
ment indicates that the timing of the ceremonies with the solstice
sunrise was crucial. This suggests that there was something important
about the relationship between the sun at this time and the type of
ritual that was performed. The nature of this relationship can only be
guessed at, but it clearly would have been of vital importance, worthy
of the enormous social and economic effort that went into the monu-
ment, and one that required an impressively bounded ritual space.
Although the sunken pits of the Aubrey Holes were undoubtedly still
visible at this time, Atkinson indicates that they were no longer used
for burial purposes, for the holes that have been examined thus far
have not yielded any material dating after the period of Stonehenge I.
In fact Atkinson believes that the period of Stonehenge II and III may
have seen a shift away from the dead toward a sky-oriented religion
due to the new influences from Europe associated with the introduction
of Beaker pottery and other imported artifacts at this time.97 The
beginning of the Bronze Age is also marked by significant changes in
burial practices, from collective burial in long mounds to the separate
burial of individuals of wealth and rank in round barrows. This
strongly indicates a change toward a more hierarchical and centralized
society, with an unprecedented emphasis upon important figures
Stonehenge lies in an area where the concentration of round barrows is
the highest in all of Wessex. Certainly Burgess is right in stating that
these graves are "redolent of chiefs and dynasties" and that the picture
that emerges, of sporadic rich graves amid a mass of simpler graves, is
very much in keeping with the concept of a structured society.9
Undoubtedly, the tall and stout man with the golden chest plate, fine
daggers, mace, lance, and shield who was buried in Bush Barrow, le
than a mile southwest of Stonehenge, was a paramount ruler of th
region.99

96 Atkinson, Stonehenge (n. 45 above), pp. 175-76.

97 Ibid., p. 173.
98 Burgess (n. 9 above), p. 166.
99 Ashbee (n. 92 above), p. 76.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
272 Stonehenge

Thus the bluestone circle of Stonehenge II coincided with the

beginnings of a hierarchical political system under a central authority.
Whether we call it a paramount chiefship, governing a federation of
chiefs, as in Renfrew's model, or a kingship, as Atkinson and others
have supposed, is irrelevant. In Renfrew's view, Wessex gives ample
evidence of a shift to some form of structured society and centralized
authority. Elsewhere in the world the change from segmentary, lineage-
based systems to centralized authority involved the establishment of
new religious forms focused upon the ruler, his ancestors, and universal
symbols of religious power, superseding the territorial cults and ances-
tor shrines of the segmentary chiefs. In Stonehenge II the solar
orientation of the site was retained and enhanced by the new Avenue
and bluestone structure. The religious shift, then, does not seem to
have been a shift from the dead to the sky, as Atkinson proposes but,
rather, from the ancestors of the old order, whose cremated bones lay
in the Aubrey Holes, to the more recently departed leaders of the new
order, who were buried in the round barrows. Hence the many round
barrows associated with Stonehenge II and III may be regarded as an
extension of the Aubrey Hole system of Stonehenge I, accommodating
the new form of burial that was adopted in the later period. The new
orientation of Stonehenge would reflect the shift from segmentary
leadership of clan-based figures to the centralized leadership of high-
status individuals, as Renfrew proposes for Wessex at this time. The
plan of Stonehenge II would clearly fit into the early phase of such a
development, with its disregard for burial deposits of the Aubrey Hole
circle, and focus upon the solar alignment linked to a new megalithic
ceremonial house in the center.
The dismantling of the bluestone circle before its completion and the
undertaking of the prodigious sarsen structure seems indicative of a
sudden surge in political organization associated with the consolidation
of a new centralized system of religiopolitical authority. Burl has
drawn attention to the absence of stone monuments in the Stonehenge
area of the Salisbury Plain prior to Stonehenge III. He speculates that
the hauling of the sarsens from the Marborough Downs, some twenty-
five miles north of Stonehenge, where the great Avebury circle of
sarsens already stood along with a number of megalithic tombs, may
reflect a conquest of this area by the builders of Stonehenge III or
perhaps a joining together of two great tribal regions under the
leadership of the Stonehenge authorities.1??
While the bluestone circle may have been intended to be building-like,
it did not have lintels and was obviously smaller than the great timber
ceremonial structures of the late Neolithic and early Bronze Age. The

100 Burl, Prehistoric Avebury (n. 6 above), p. 242.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 273

large, central timbers of neighboring Woodhenge have been estimated

to have stood twenty-four feet high.101 If the bluestone double circle
was intended to reflect the status and authority of a new institution of
paramount chiefship, this structure was not a very grand edifice. At
this time all the bluestones were rough and unshaped, and most of
them stood less than six feet tall. Although the bluestone concentric
did express the refocusing of the monument in the hands of a new
political system, the new order was obviously capable of something
much grander in scale.
Stonehenge III was therefore begun, and its initial sarsen phase (the
outer circle and inner trilithon horseshoe) took well over a decade to
construct. When completed, it was the most spectacular megalithic
monument in all of Wessex, and, for that matter, in all of the British
Isles and Europe. To be sure, the greatest sarsen circle of all stood in
Avebury eighteen miles to the north. It was twice the diameter of
Stonehenge and contained two inner stone circles somewhat larger
than Stonehenge. But Stonehenge could boast a circle of far greater
height, with stone lintels giving it a distinctive building-like appearance.
The sophistication in engineering skills and stone craftsmanship were
plain to see and attested to a more powerful cultural and political
system.
If Stonehenge III was built by the powers of paramount chiefship, as
Renfrew suggests, then it seems reasonable to assume that the ritual
function of the monument was primarily related to this institution. It is
possible, of course, to imagine a variety of ritual functions. But this
would require making additional assumptions about the nature of the
socioreligious system, and it is preferable to make as few as possible.
The most suitable theory is therefore one that would link the function
of the monument to the institution that built it; and all the evidence
points to that institution's being the paramount rulership.
The fact that Stonehenge III resembles a great ceremonial building,
not a tomb or barrow, might suggest that it served only the living ruler
rather than his deceased predecessors. If so, the periodicity implied in
the monument's solstitial alignment indicates that it might have been
used at regular intervals during the ruler's reign, perhaps for the
purpose of annual legitimation and renewal. Such a ceremony existed
among the Swazi of southern Africa. Here the ruler was annually
"strengthened" by rites performed in conjunction with the summer
solstice and full moon, and the solidarity of the society as a whole was
also emphasized.102 Thus the intent to make Stonehenge look like a

101 Cunnington (n. 16 above), p. 23.

102 Hilda Kuper, An African Aristocracy (1974; reprint, London: Oxford University
Press, 1969).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
274 Stonehenge

building or a house may have been more specifically a desire to make it

look like a palace, the ritual house of the central religiopolitical figure
of Wessex civilization, like the large building at Durrington Walls.
Wainwright believes that this building was the ceremonial center of a
large complex of domestic buildings within the henge, and, given the
large amount of pig and cattle bones found at the site and open roof
design of the building, it could well have been a place of public
assembly and feasting associated with a ruling figure.
But the weight of the evidence is against interpreting Stonehenge
exclusively in terms of the living ruler. Virtually all megalithic monu-
ments that have been excavated in the British Isles and Europe are
associated in some way with the dead. In the case of Stonehenge III it
seems reasonable to assume, in light of all the circumstantial evidence,
that the Bronze Age barrows that surround it probably had a direct
ritual association with the monument. This would place the round
barrows in historical continuity with the monument's original purpose,
that of joining together the solar and (possibly) lunar cycles with the
remains of the dead. To assume otherwise, namely, that the barrows
were not integrally related to the monument and that the monument
was reoriented in the Bronze Age to serve only the living rulers, would
be to propose a major discontinuity in the history of the site for which
there is no evidence.
Although it is possible that agricultural fertility was involved in
such rites, this could not have been their primary purpose. The summer
solstice occurs well after the growing season begins, and the winter
solstice occurs well after the harvest is finished. Equinox alignments
would indicate agricultural purposes far more clearly than solstitial
ones. Burl has also envisioned shamanistic ceremonies at stone circles
for the purpose of animal and human fertility.103 While this might be
appropriate for small and egalitarian groups, such as those associated
with Neolithic Scottish circles of Burl's studies, it is not appropriate for
the Wessex context. Shamanism is an individualized form of priest-
hood, involving personal client relationships, and it is associated with
segmentary societies. In a hierarchical, structured society, such as that
of early Bronze Age Wessex, public rites are concerned with corporate
institutional matters: the fertility of the land and the people, seasonal
transitions, the purification of the society, the legitimation and renewal
of sacred chiefs and kings, and the cult of the deceased rulers. If we
agree with Renfrew that the building of the great Wessex monuments
was a major act of social unification, then the rituals performed at such
sites would have had a broad-based social function.

103 Burl, Stone Circles of the British Isles (n. 18 above), pp. 86-87.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 275

Here it is interesting to consider the role of Stonehenge's nearby

counterpart, Woodhenge. This solstice-aligned building, surrounded
by a henge, was located just outside the embankment of Durrington
Walls, close to the eastern entrance (fig. 1). Could this building have
been a ritual center for use at the time of the solstices for the ruler who
lived within the great henge? If so, Stonehenge might have been its
counterpart where solstitial rites were devoted to the spirits of the
deceased rulers. Their graves may also have been intentionally arrayed
around Stonehenge like the smaller dwellings that Wainwright believes
stood in the henge together with the large ceremonial building. Thus, it
is possible to see Stonehenge as the ceremonial center of a great
"henge" formed by the surrounding barrows, with an opening toward
the northeast, where few barrows were built (fig. 1).
At this point it will be illuminating to look more closely at the Swazi
ceremonies mentioned above. What is suggestive here are the symbolic
and ritual relationships between the solstitial sun, and phases of the
moon, the king, the New Year, and the solidarity of the society as a
whole.
The Swazi Incwala ceremony was performed at the time of the
summer solstice that marked the beginning of the year, and it was the
time when the mystic powers of the king reached their weakest state.
According to the Swazi, the Incwala aimed at "strengthening the
kingship," at "showing the kingship," and "to make stand the nation."
The timing of the ceremonies was crucial. In the 1930s, when the
anthropologist Hilda Kuper witnessed the rites, she noted that as the
time grew near, "throughout the whole country there [was] ... one
main topic of conversation: 'When will the moon and the sun be
right?'"104 Every morning aged councillors stood in the royal cattle
byre, which faced east, and watched the sun rise and judged its position
by certain fixed landmarks. Every evening they discussed the size of the
moon and the position of the stars. The ceremonies had to begin the
day after the solstice when the sun had reached its "hut" and was
"resting," that is, on the solstice. For the ceremony to end before the
sun started its return journey to the north, or for it to have begun
afterward, would have been a national calamity, requiring special
rituals to rectify. If the rituals were wrongly timed, the king would not
have been strong enough to bear the development of the coming year,
and he could not have withstood the kings of other nations who had
the "strength of the sun" with them. The Swazi also believed that the
phases of the moon were related to human development and to the life
cycle. When the moon was full, human beings on whom it was

104 Kuper, p. 201.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
276 Stonehenge

operative would be healthy; when it was dark or waning a person

would be weak and puny. A ceremony that gave someone new status
was always held when the moon was growing full: whereas a ceremony
which temporarily separated someone from society was held at the
time when the moon was dark. The Incwala began when the moon was
dark, for this was when the king in his weakened state withdrew into
his enclosure to be doctored and strengthened by medicines prepared
by his people. Since the dark or waning moon was rarely coincidental
with the December solstice, the Swazi chose to hold the ceremony on
the dark moon before December 22 and had the ceremony end after
the sun had reached its southernmost point. At this time the king was
identified with the sun, and the moon was said "to race the sun." The
ceremony was performed mainly by the king's royal relatives (who
governed the districts of the kingdom), the king's councillors, and the
king's regiment of warriors. In the course of the ceremony the king was
treated as though he had lost his power and had become humiliated
before his people. The king drank potent medicine to renew his
strength and spat it out "to break the old year and prepare for the
new." First he spat to the east, then to the west, and the crowd shouted,
"He stabs it," meaning that the king had ended the old year and had
strengthened the earth. After an interval of fourteen days, when the
songs of the ceremonies were sung around the country, the main
ceremonies began at the capital on the day after the full moon. The
climax occurred when the king was reborn and revitalized. He displayed
his power in a fearsome dance before his warriors, and he hurled at
them a specially treated gourd symbolizing the past year. Afterward
the king was led away and his costume removed. He entered his hut for
the night and cohabited with his first ritual wife. He spent the next day
in seclusion, and the cheeks of his face were painted white in imitation
of the full moon. On the final day all the ritual implements containing
the filth and pollution of the king and the people were burned in a great
fire. The king was cleansed and the drops of water falling from his
body were said to make the rains come. The dirt and pollution of the
old year were thus exchanged for the rain and renewal of the new. As
one Swazi put it, "The nation's life, soul, and well-being hang in the
faith and belief that the rebirth, rejuvenation, and purification of the
king ushers in a new life, added virtue, and strength and national
unity." Another said, "We see we are all Swazi, and we are joined
against outside foes."105 Although there was an agricultural, first-fruits
aspect to the Incwala, the main harvest was held much later. Hence
Kuper points out that the Incwala was not an agricultural ritual: "The

105 Ibid., pp. 210, 224.

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
History of Religions 277

time of the Incwala as a ceremony of the kingship and of new strength

is regulated by the natural phenomena, moon and sun, which link
mystically with the stages of human destinies, and not by agricultural
routine, which people themselves largely control."'06
This is not to say, of course, that Stonehenge II and III functioned in
the same fashion. Ethnographic comparisons can be misleading if too
much is made of their similarities and differences. Yet such comparisons
are useful if they help in the formulation of hypotheses.107 In this case
the question is, How might solstice ceremonies, linked to the graves of,
chiefly, ancestors, play a major ritual function in the chiefdom society?
The Swazi rites suggest that the solar and (possibly) lunar alignments
of Stonehenge II and III could have been part of an annual renewal
ceremony aimed at the paramount chief himself or at the institution of
rulership in general, through the powerful spirits of the ancestors. In
this case the Wessex chiefs would have erected Stonehenge as an
expression of the celestial and ancestral foundations of the chiefship and
as the ritual instrument for maintaining the vital connections between
these sacred forces. When the chiefs died, they were buried close to the
site of Stonehenge's splendor because they participated in (and perhaps
mediated) its cosmic renewing action like the original ancestors whose
bones lay buried in and around the Aubrey Holes of Stonehenge I.
Thus the proximity to Stonehenge III of so many rich barrows seems
to confirm the central role of the nonument in the ruling structures of
the early Bronze Age period.

CONCLUSION

One might well wonder why such a relatively simple theo

been proposed before. Of course its general import is not ent
for it is basically in agreement with the memorial-to-the-
pretation proposed by Geoffrey of Monmouth in the twelfth
Since it is likely that Geoffrey drew upon extant folk tra
interpretation may date far back in time. L. V. Grinsell has n
the number of slain nobles (460) may also have represented th
of round barrows there were thought to have been in the
Stonehenge and believed to been the nobles' graves."0
conviction that there was a relation between Stonehenge
surrounding barrows may be an ancient one. In this essa
brought together the evidence for this idea with the evidenc

106 Ibid., p. 224.

107 I. F. Thorpe, "Ethnoastronomy: Its Patterns and Archaeological Impl
Ruggles and Whittle, eds. (n. 9 above).
108 Grinsell, Stonehenge Barrow Groups (n. 93 above).

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms
278 Stonehenge

original function of Stonehenge and suggested a unifying ritual inter-

pretation involving the institution of chiefship and the astronomical
orientations of the monument. Until recently, the relationship between
Stonehenge I, II, and III and its neighboring monuments had not
received enough attention for such a theory to be proposed. The
astronomical study of other megalithic sites was also insufficiently
developed to provide a useful context for interpreting the astronomical
features of Stonehenge. If the key to the sociopolitical order of the
early Bronze Age was the institution of paramount chiefship, as
Renfrew believes, then the ritual function of Stonehenge II and III can
be reasonably supposed to have served this institution, perhaps being
focused upon the person of the paramount ruler himself. To suppose
otherwise would entail postulating other institutional arrangements for
which there is at present no evidence. Stonehenge II and III may
therefore be seen as a continuation of the original purpose of the
monument built by the Neolithic system of segmentary clan leadership.
Given the enormous public effort involved in both the Neolithic and
Bronze Age epochs, it may also be assumed that the Stonehenge rites
were supposed to secure the welfare of the society as a whole, not just
that of its leaders, living or dead. The concentration of Bronze Age
round barrows around Stonehenge thus appears to fit in with the
historic purpose of the monument and to explain why it was main-
tained and rebuilt through the centuries.

University of Virginia

This content downloaded from

203.144.88.1 on Mon, 22 Apr 2024 03:34:12 +00:00
All use subject to https://about.jstor.org/terms