0% found this document useful (0 votes)
21 views

Metamorphoses of Science Fiction

Darko Suvin's 'Metamorphoses of Science Fiction' explores the poetics and history of science fiction as a significant literary genre. The book argues for the importance of studying science fiction beyond its commercial aspects, emphasizing its cognitive potential and sociological significance. Suvin presents a dual approach, combining theoretical insights with historical analysis to highlight the genre's evolution and its relationship with societal changes.

Uploaded by

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

Metamorphoses of Science Fiction

Darko Suvin's 'Metamorphoses of Science Fiction' explores the poetics and history of science fiction as a significant literary genre. The book argues for the importance of studying science fiction beyond its commercial aspects, emphasizing its cognitive potential and sociological significance. Suvin presents a dual approach, combining theoretical insights with historical analysis to highlight the genre's evolution and its relationship with societal changes.

Uploaded by

Carlos
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 328

METAMORPHOSES

OF SCIENCE FICTION

On the Poetics and History


of a Literary Genre

Darko Suvin

New Haven and London Yale University Press

1979
To friends and comrades from Science-Fiction Studies
Marc, Fred, Ursula, Dale, Patrick, Bob, and Don:
they helped.

Copyright C 1979 by Yale University. Library of Congress Cataloging in Publication


Data
Al! rights reserved. This book may not be
reproduced, in whole or in part, in any form Surin, Darko, 1980-
(beyond that copying permitted by Sections Metamorphoses of science fiction.
] 07 and 108 of the U.S. Copyright Law and Bibliography
except by reviewers for the public press), Includes index.
without written permission from the publishers. 1. Science fiction—Hitory and criticism,
Printed in the United Stales of America by 1. Title.
Murray Printing Company, West ford, PNS448.S45SB97 809.3'876 7 8-6265
Massachusetts. ISBN 0-3004)2250-6
Published in Great Britain, Lu rope, Africa, and
Aria (except Japan) by Yale University Press,
Ltd., London. Distributed in Australia and New
Zealand by Book & Film Services, Artarmcn.
N.S.W., Australia: and in Japan by Harper 9c
Row, Publishers, Tokyo Office.
Contents

Preface vii

Acknowledgments xvii

I: POETICS
1 Estrangement and Cognition 3
2 SF and the Genological Jungle 16
3 Defining the Literary Genre of Utopia: Some Histori­
cal Semantics, Some Genology, a Proposal, and a Plea 37
4 SF and the Novum 63

II: HISTORY
Introduction to Older SF History 87
5 The Alternative Island 90
6 The Shift to Anticipation: Radical Rhapsody and
Romantic Recoil 115
7 Liberalism Mutes the Anticipation: The Space-
Binding Machines 145
8 Anticipating the Sunburst: Dream, Vision—or Night­
mare? 170
Introduction to Newer SF History 205
9 Wells as the Turning Point of the SF Tradition 208
10 The Time Machine versus Utopia as Structural Models
for SF 222
11 Russian SF and Its Utopian Tradition 243
12 Karel ¿apek, Or the Aliens Amongst Us 270

Bibliography 285

Index of Names 311


Les choses pourraient être autrement. . . .
Raymond Ruyer

I remember that in Spain during the civil war I had a revelation of


“the other man." . . . The Spanish dream was broken and defiled
later, ... the faces I saw have become as they were before they were
transformed by that elated sureness. ... But the memory will never
leave me. Anyone who has looked Hope in the face will never
forget it. He will search for it everywhere he goes, among all kinds
of men. And he will dream of finding it again someday, some­
where, perhaps among those closest to him. In every man there is a
possibility of his being—or, to be more exact, of his becoming once
again—another man.
Octavio Paz

What is at issue is not (merely) relating the works of literary art to


the historical context of their origin, but representing the time of
interpretation (i.e. our time) in the time of their genesis. Thus
literature becomes an organon of history. . . .
Walter Benjamin

I have been a sorehead occupant of a file drawer labelled "science


fiction" ever since [my first novel], and I would like out, particularly
since so many critics regularly mistake the drawer for a urinal.
Kun Vonnegut, Jr. ,

But then are we in order when we are most out of order.


William Shakespeare
Preface

1. A justification for paying serious attention to science fiction


may by now be necessary only for other literary critics and
scholars. Still, the question of why a book addressing itself to
science fiction, and furthermore one that does not deal with what
is admittedly the focal point of the genre and the convergence
point of this book too—modern English-language SF—has to be
faced briefly. I cannot even begin here to discuss the reasons for
studying paraliterature—the popular, “low,” or plebeian literary
production of various times, particularly since the Industrial
Revolution. Let me merely note that a discipline which refuses to
take into account 90 percent or more of what constitutes its
domain seems to me not only to have large zones of blindness
but also to run serious risks of distorted vision in the small zone
it focuses on (so-called high lit.). The noncanonic, repressed twin
of Literature which, for want of another name, one calls Para­
literature is (for belter or worse) the literature that is really
read—as opposed to most literature taught in schools. Within it,
SF is one of the largest genres, and to my mind the most interest­
ing and cognitively most significant one. This is not at all to say
that an average SF text is “good,” that is, aesthetically significant.
On the contrary, 90 or even 95 percent of SF production is
strictly perishable stuff, produced in viewr of instant obsolescence
for the publisher’s profit and the writer’s acquisition of other
perishable commodities. But even this 90 or 95 percent is highly
significant from the sociological point of view, since it is read by
the young generation, the university graduates, and other key
strata of contemporary society, and is thus only less important
than the 5 to 10 percent of SF that it aesthetically significant: in
our days the writings of Lem, Le Guin, Dick, Disch, Delany, the
Strugatsky brothers, Jeury, Aldiss, Ballard, and others.
2.1 If the interest for this genre has flowed out from the ex­
pansion and influence of that category of commercial publication
called “science fiction,” this interest should not be confined to the
last 50 (or even 100) years, a period in which the SF production
has been predominantly determined and strongly inflected by
the capitalist market with its alienating and degrading tenden-
vti
VU1 PREFACE

cies. On the contrary, the no doubt very important empirical


realities of SF must—if we are to pass any value-judgment on
them—be obstinately confronted with the as important historical
potentialities of the genre. These potentialities are necessarily in
part speculative, but no more so than any reasonable hypothesis
based on observable facts and probable laws. For laws, we can
use the characteristics inherent in this genre and its generic telos,
determined by homology with the characteristics of other, more
developed genres, be they the psychological novel or the fairy
tale. For facts, we can use the best productions in this genre, as
redefined by these laws in a proper, spiralling hermeneutic feed­
back. In this perspective, SF should not be seen (as 1 will argue at
length in the theoretical part of this book) in terms of science,
the future, or any other element of its potentially unlimited
thematic field. Rather, it should be defined as a fictional tale
determined by the hegemonic literary device of a locus and/or
dramatis personae that (1) are radically or at least significantly different
from the empirical times, places, and characters of “mimetic” or
“naturalist” fiction, but (2) are nonetheless—to the extent that SF
differs from other “fantastic” genres, that is, ensembles of fic­
tional tales without empirical validation—simultaneously per­
ceived as not impossible within the cognitive (cosmological and
anthropological) norms of the author’s epoch. Basically, SF is a
developed oxymoron, a realistic irreality, with humanized
nonhumans, this-worldly Other Worlds, and so forth. Which
means that it is—potentially—the space of a potent estrangement,
validated by the pathos and prestige of the basic cognitive norms
of our times.
As always, theoretical delimitation calls for a historical corpus
(and vice versa): if one does refuse the one-dimensionality of
most current commercial SF, then I do not see any logical possi­
bility of a delimitation that would not be at least akin to the one
in the preceding paragraph. In that case, all the not impossible
other worlds and voyages thereto, from Lucian and More on,
must be included into SF on the same basis as Verne and Wells.
Thus it is not for directly ideological reasons but for formal ones
that I have included utopias (and marvelous voyages) in SF. Ad­
mittedly, formal principles are themselves ultimately bound to
given conceptual horizons and are in that sense "ideological," but
in an indirect or mediated way (by way of the notion of “lateral
PREFACE ix

possibilities,” which is touched on in the first three chapters); and


to my mind this makes all the difference.
2.2 The general working hypothesis in this book has been that
the history of SF is the result of two conflicting tendencies. A
potential cognitive tendency, quite evident in all the significant
writers dealt with (More, Lucian, Cyrano, Swift, Verne, Wells,
Capek, etc.), is allied to the rise of subversive social classes and
their development of more sophisticated productive forces and
cognitions. However, an opposed tendency toward mystifying
escapism dominates in second-rate SF and shows even in the
masters (the statics of More or Swift, the catastrophism of
Mary Shelley and Wells, the positivism of Verne), formed as it is
by the practical and cognitive limitations of fiction steeped in the
alienation of class society and in particular by the stagnation of a
whilom subversive class. Such is the case with the bourgeoisie
which, in its decline from the leader of a general plebeian pro­
gression to an isolated exploiter of nature and people, comes to
treat productivity and cognition not as fundamentals of creative­
ness but as means of profit, and reverts in the process to all kinds
of mystifications, from theism to astrology. The ascendancy of a
cognitive approach makes for its fertile blend with the ludic
pleasure of estrangement. Obversely, the predominance ol anti-
cognitive impulses degrades estrangement to a formal, surface
sensationalism that first shocks the bourgeois but then rejoins
him—already very apparent in Verne and the mediocre Wells.
In the first case, estrangement is a creative approach, an organon
(as Bloch said of utopia) for exploring the novum, but in the
second case it is an opium for the people: if one should not
forget that opiates may be necessary for momentary relief from
great pain, one should not forget either the venerable adage
corruptio optimi pessima.
All such considerations should, no doubt, also be the start­
ing point for any serious examination of English-language SF
of the last 50 years. I have tried my hand at such examination
elsewhere;1 and I hope to have here at least indicated why this

1. The reader is respectfully referred to my articles “The Significant Context of SF,”


‘‘P. K. Dick's Opus: Artifice as Refuge and World View," and "Parables of De-Aliena lion:
Le Guin's Widdershins Dance," in R. D. Mullen and Darko Suvin, eds. (Bibliography I};
"The Science Fiction .Novel in 1969,” in James Blish, ed,, Nebula Award Stories Five (New-
York, 1972); and, in spite of its title, "Stanislaw Lem und das mitteleuropäische Bewusst-
PREFACE

is, in my opinion, impossible without a previous theoretico-


historical reflection such as is attempted—but certainly not
concluded—in the present book. The preceding very brief and
partial sketch of the book’s rationale may provide some first
arguments as to why and how the lack of such a reflection makes
for what one must regretfully call the one-dimensionality of
much SF criticism—measured not so much by a priori ideological
criteria (which can nonetheless, just as the ethical and formal
criteria, sene as a first alarm bell) as by the practical criteria of
the stupefying consequences for the matter at hand, contempo­
rary SF. Hence my belief in the necessity of reculer pour mieux
sauter, of beginning from the beginnings.
3. It remains to sketch a few considerations about the shape of
the book at hand: how did it come about, why does it contain two
kinds of essays, theoretical and historical, and why do they deal
with what they do rather than with other possible matters?
3.1 Habent sua fata UbeUr. and though the historical vicissitudes
of this book do not make it better or worse, they may help to
explain it. For the book has been gestating (on and off) for about
20 years. A first sketch of its historical part—no doubt crude and
excessively lacunary, but in its attitudes, subdivisions, and choice
of subjects a "baby figure of the mass to come"—was published as
“The Sailor on the Mast” in the Zagreb monthly N’a3e teme (Our
Themes) in 1958. Edited by Ivo Bojanic fresh from the chairman­
ship of the Zagreb student union, this was an exciting periodical
which brought together and largely revealed the young post­
revolutionary generation of philosophers, social scientists, and
cultural and Iiterarv critics arising from Yugoslavia’s dual anti­
capitalist and anti-Stalinist struggle; even the somewhat pathetic
title of my article, taken from an essay of Zamyatin’s as charac­
terizing the role of modern literature in general and SF in par­
ticular, testifies to the excitement of that moment ¡n time and
space. In more academically mediated ways, a similar orientation
and pathos informed at least the most significant contributions
which blended a “warm" Marxist attention to historical specificity
with the Formalist attention to material forms of textuality in
Umjetnost rijeci (The Verbal Art). This quarterly was published at

setn der Science-fiction," in Werner Berthel. ed., Inset-Abnanach auf das Jahr 1976: Stanis­
law Lem . . . (Frankfurt. 1976).
PREFACE XJ

the Faculty of Philosophy in Zagreb, where I began teaching in


1959, and in it a first approach to the theoretical part of this
book appeared in 1963 as “SF and Utopianism.” I will not re­
count the further stages of gestation, which led through a
number of essays, published in Yugoslav periodicals, on major
figures in SF history, from Lucian to Morris, to my history-cum-
anihologv Od Lukijana do Lunjika (From Lucian to the Lunik, Za­
greb, 1965) and to English-language publications from 1969 on,
except to say that though all these texts were completely rewrit­
ten in the 1970s the main orientation has, I trust, remained
constant. It can perhaps be followed in the epigraphs to this
book.
3.2 As to the book’s coverage, I am acutely aware how diffi­
cult it is to present a fully convincing case about practically any­
thing, much less a whole historical tradition, within the scope of an
individual’s effort. True, in the last decades much scholarship
and criticism of the first order has been devoted both to the
major works in this tradition and to the general theory of culture
and fiction that underlies my approach, and (even when I had to
differ with established opinion in particular fields) I have
thought it possible to deal with both the historical and theoretical
part in the spirit of was mich nicht umbringt, macht mich starker'.
a spirit of summary, connection, and incorporation into whatever
newr concepts I had arrived at. Nonetheless, the only proper,
truly modern approach to all such matters would have been an
organized collective effort by a team with identical horizons that
included not only people who would know much better the vari­
ous literary times and places dealt with, but also an economist,
a psychologist, a philosopher; and in the case of SF also a
politicologist, a historian of science and technology, and so forth.
Indeed, an individual overview of a whole historical process,
however one slices it up, has by now become an impossibility for
any single scholar (for a socialist it is, furthermore, a conscious
contradictio in adiecto). Even this work of mine would have been
quite impossible without the generous material help of collective
bodies and moral and intellectual help of a number of friendly
colleagues, mentioned in the acknowledgements and dedication.
But this poor individual approach to or variant of teamwork has
still left me with the necessity of a number of compromises and
maneuvers. One is a certain rhetorical boldness and foreshorten-
xii PREFACE

ing, especially in the theoretical chapters, without which nothing


of interest could have been said at all. Another is a not merely
linear but also spiral progression of the argument in the theoret­
ical section of the book, where matters already touched upon are
looked at again in a wider context and (I hope) with better
results—as, for example, the discussions of naturalistic versus
estranged genres or of extrapolation. A third is the adoption of a
systematic—but implicit rather than explicit—convergence to­
ward the locus from which any reflection on modern SF must
start out, the contemporary mass production, as a principle of
unity. One or two other stratagems will be touched on below; let
me mention here only one further problem (not peculiar, of
course, to this book) posed by culture being at any moment a
conflictual unity of the historically concrete particular and the
equally historically given possibilities of the general. A cultural
process such as a literary tradition can therefore only be grasped
with help of the general: the horizons and tendencies of a cul­
tural locus—its historical semantics, to begin with—have to be
used to interpret any production of that time, place, and society.
Yet, paradoxically, no such general approach can afford to
forget that, while in the antinomic class culture the atomized
particular—the actual text as isolated monad—is stunted, the
undifferentiated general—the theoretical approach as pure
ideation—is dehumanized. Thus, for a vivifying and meaningful
result, a (hopefully) wise interaction between these two poles is
constantly needed. This is why in my opinion further valid
theoretical elaboration will have to pass through investigations
into a social theory of literature—a long, arduous, and expensive
pursuit obviously impossible without teamwork. It is to be hoped
that such work might be carried out around the journal Science-
Fiction Studies in Montreal.2

2. The "Sociology of SF" issue of Sctence-Fiction Studies (No. 13 [1977]), reprinted in R.


D. Mullen and Darko Suvin, eds., Science-Fiction Studies: Selected Articles on Science Fiction
1976-1977 (published in 1978 by Gregg Press/G. K. Hall, Boston) could be considered
as a prelude to such investigations. A general basis for them is given by the items anno­
tated in Marc Angenot's bibliography of the sociology of literature in that issue and in
the book. In particular. Fredric Jameson's Marxism and Form (Princeton, 1972) and Ray­
mond Williams’s Marxism and Literature (Oxford. 1977) provide both excellent summa­
tions and significant developments ol a social theory of literature and culture. Though
the latter title appeared too late for this book. I am methodologically much indebted to
both of these critics, as well as—obviously—to quite a number of others either mentioned
PREFACE xiii

3.3 The historical part is arranged chronologically, from


More to Wells with a retrospective from More to at least some kin­
dred earlier forms in the Middle Ages and antiquity, and with two
forays from Wells into twentieth-century domains, forays which
are if not chronologically then logically anterior co English-
language SF of the 1940s and 50s. But as 1 have said, this book is
an essay in definition, appreciation, and evaluation; although it
takes a historical view and attempts to identify key epochs of
achievement and development, it is not—nor could it within a
reasonable compass even approach—detailed literary history.
Even where it most closely approaches an overview, I found it
inevitable to focus at some length on a small number of tales,
trusting that the reader will infer their representative status from
my analyses rather than from cumbersome special metadiscus-
sions of reasons for it. More, Rabelais, Bacon, Campanella,
Cyrano, Swift, Mary Shelley, Verne, Bellamy, Morris, Twain,
Wells, and Zamyatin are names that may be expected to appear
in a study of the SF tradition as delimited in my theoretical part,
and 1 have included them. Obversely, however, the nature of the
field demanded also that such “intensive” representativeness or
vertical cross-cutting be in a number of cases accompanied by
horizontal cross-cutting or “extensive” representativeness. Some
analytico-synthetical assumptions about the equivalence of these
two types of representativeness or cutting, difference from as­
sumptions classically used for “high lit.,” are perhaps hidden in
this methodology', but the book does not explicitly delve into this.
A number of further points about the historical section will be
indicated in the introductions to “Newer History" and “Older
History.”
In particular, 1 am painfully conscious that ideal circumstances
of time, resources, teamwork, and the like might have mitigated
my curtness on subjects, authors, even works that in their rich­
ness deserve by themselves to be (and have often been) dealt with
at book length. This holds true for almost all major authors or
works glanced at in this book (and indeed for a number of minor

in this introduction, in the notes and bibliography to the book, in my introduction to the
"Sociology of SF' issue cited above, or left uninentioned on the assumption that it was
better to run the risk of some readers not noticing my debts than to swamp other readers
with unnecessary references to them.
xiv PREFACE

ones); perhaps the two most obvious ones might be Blake and
The Tempest. I am here not referring primarily to my possibly
heretical approach to some works: after much soul-searching I
find myself, for example, much less favorably disposed toward
New Atlantis and much more critical of The Tempest than tile
present majority (though not unanimous) opinion, much more
appreciative of Cyrano, and guilty of theoretical imperialism by
annexing to SF not only the fictional utopia but also Gulliver’s
Travels. An accusation of subversiveness or heresy would, how­
ever, be quite congruous to my heretic subject matter, the genre
showing how “things could be different,” and thus in a way a
great compliment. What I have in mind is rather what seems
prima facie the sheer inadequacy of dealing with Blake or The
Tempest in a few pages each. Admitted: and the accompanying
Bibliography is at any rate an indication that any sins were pre­
meditated. But what I wanted to provide a first argument and
sketch for was not the central thing that could be said about a
number of cultural phenomena in abstracto, but how each of
those representative phenomena in some significant—central or
eccentric—aspects arises out of, Hows into, or otherwise contrib­
utes to my purposes—the tradition I am sketching and arguing
for; and on that basis I would like to think I shall be judged. I
am attempting, in other words, to apply a by now fairly classical
procedure—in the wake of the concepts of Eliot, Lukács, Auer­
bach, Bloch, and Brecht, possibly best formulated by the latter's
friend Benjamin in writings such as that from which one of the
epigraphs to this book was taken—to a new (but only seemingly
newfangled) subject. Furthermore, as different from the “high”
or elite tradition of most (though not all) Eliot and Lukács, pos­
sibly the Russian Formalists, Brecht, Bloch, and Benjamin have
by now taught us that a tradition is not necessarily—or even that
any healthy tradition is necessarily not—only a canonic or “high
lit.” one. On the contrary, a healthy tradition is a diachronic
texture of what Raymond Williams has lately called the domi­
nant, the emergent, and the residual in any cultural synchrony
(for a synchronic moment is simply an analytic convenience for
marking a process, a pause between systole and diastole rather
than Faust's beautiful cardiac arrest). Nearest to SF, I still vividly
remember the revelation that, in my student days, Empson’s
sometimes perhaps perverse but always beautifully bold Some
PREFACE xv

Versions of Pastoral was to me, to the point that I seriously consid­


ered calling this book, alliteratively, Some Versions of SF—a title
that might still be appropriate. However, the pleasing blend of
protean for mal-cum-substantial process identified by the (Lucre­
tian rather than Ovidian) metaphor of metamorphosis kept re­
curring in my typewriter so often that I finally, as a materialist
should, surrendered to my matter, hominum deorumque genitrix.
4. No doubt one could envisage various ways of supplement­
ing this book: not only chronologically but also in depth. Thus the
history of SF could—once we had it—usefully be discussed as a
history7 of some invariant problems in its storytelling: plot,
chronotopes, the inextricable link between socielas hominum and
socielas rerum, and so on. This study is confined to working out
the rationale and the main lines of an SF tradition. It has the
ambition to be pathbreaking, and the limitation of pathbreakers:
once the path has been chosen, one cannot glance right or left
very often. But on the whole, I would hope the book might show
the interested reader that side by side with the “canonic” genres
there is a great number of works of fiction which have so far
been neglected, or considered solely as nonfiction (history of
ideas, umnediated ideology, etc.), or considered as marginal
aberrations from another properly canonic tradition; and that, as
a consequence, the specific pleasure which a reader has the right
to expect and find in them has been too largely ignored. In a
way, this book is an exercise in cleansing our perception; and
w hat wfe could perceive is this material dialectics of human his­
tory and its possibilities: “the development of the five senses is a
labor of the whole previous history of the world.”3 Never was it
so necessary as it is today to ilumínate this history of ours, with
its works and dreams, triumphs and servitudes. The formalized
daydreams of science fiction can be claimed as a privileged pars
pro toto or vibration of this history.

3. Karl Marx, '‘Private Properly and Communism," in Loyd D. Easton and Kurt
Guddat. eds., Writings of the Young Marx on Philosophy and Society (Garden City, NY,
1967), p. 309,
Acknowledgments

Various versions and parts of this work have benefited from


the financial assistance of the University of Massachusetts at
Amherst, the Humanities Research fund at McGill University,
and especially of several Canada Council research grants. The
bulk of the book was prepared during the 1973-74 sabbatical
leave granted me by McGill University and partly supported by a
Canada Council Leave Fellowship. My sincere thanks go to all
these institutions, whose encouragement was not merely mate­
rial, as well as to the main libraries whose staff facilitated my
research: the McLennan Library at McGill University (and in
particular its Inter-Library Loans department, unfailingly help­
ful above and beyond the call of duty), the New York Public
Library, the Library of Congress, the libraries of Yale, Indiana,
and Cambridge universities, the British Library in London, the
national Libraries at Paris and Florence, and the National Uni­
versity Library at Zagreb.
Earlier reasonably recognizable versions of various chapters—
as a rule significantly changed or enlarged in the present
book—were first published in the following periodicals: chapter
I in College English No. 3 (1972) (Copyright © 1972 by the Na­
tional Council of Teachers of English. Reprinted by permission),
and in a differing version in Foundation No. 2 (1972); chapter 2
in Genre No. 3 (1973); chapter 3 in Studies in the Literary Imagina­
tion No. 2 (1973); the first part of chapter 5 and chapter 6 in
Science-Fiction Studies No. 4 (1974) and No. 10 (1976); the first
part of chapter 7 in Clio No. 1 (1974); chapter 9 in The Minnesota
Review No. 4 (1975); chapter 10 in Comparative Literature Studies
No. 4 (1973) and in a differing version in Strumenti critici No. 18
(1972); and chapter 11 in The Modern Language Review No. 1
(1971). My thanks are due to the editors of all these periodicals
for permission to use the above materials.
i have incurred debts of gratitude larger than usual, I suspect.
First and foremost, to the friends to whom the book is dedicated,
and to whose names that of Nena should (as always: come prima,
piu di prima) be joined. Second, to many other people who have
KVii
xviii ACKNOWLEDGMENTS

encouraged my work, from whom I have surely learned and


absorbed much, but who are too numerous to mention here. I
would at least like to single out my McGill students, whom I have
slyly induced to teach me more often than they suspected; the
Department of English chaired by Donald F. Theall and later by
Peter Ohlin, who allowed and indeed incited me to teach SF to
those students; and my friends and colleagues, the late Jim Blish,
Mike Bristol, Mike Holquist, Charles Le Guin, and Judy Merrill
in North America, as well as—much earlier—Ivo Bojanic, Ivan
V. Lalic, Milan Miric, and Zdenko Skreb in Yugoslavia. I am also
indebted to Barbara L. Campbell for the typing and indexing,
and especially to Ellen Graham and Lynn Walterick, my editors
at Yale University Press.
As for the reasons, factors, thrones, dominations, and powers
that prevented me from writing a better book, I hint at some in
my preface . . . : and to spell them out in a kind of “Anti­
Acknowledgements” might have been quite illuminating, but it
would be too long.
D. S.
Montreal, April 1978
I

POETICS
1

Estrangement and Cognition

1. Science Fiction As Fiction (Estrangement)


1.1. The importance of science fiction (SF) in our time is on
the increase. First, there are strong indications that its popularity
in the leading industrial nations (United States, USSR, United
Kingdom, Japan) has risen sharply over the last 100 years, de­
spite all the local and short-range fluctuations. SF has particu­
larly affected such key strata or groups of modern society as
college graduates, young writers, and the avant-garde of general
readers appreciative of new sets of values. This is a significant
cultural effect which goes beyond any merely quantitative cen­
sus. Second, if one takes the minimal generic difference of
SF the presence of a narrative novum (the dramatis personae
and/or their context) significantly different from what is the
norm in “naturalistic" or empiricist fiction, it will be found that
SF has an interesting and close kinship with other literary sub­
genres that flourished at different times and places of literary
history: the classical and medieval “fortunate island" story, the
“fabulous voyage” story from antiquity on, the Renaissance and
Baroque “utopia" and “planetar}' novel,” the Enlightenment
“state [political] novel,” the modern “anticipation” and “anti-
utopia.” Moreover, although SF shares with myth, fantasy, fairy
tale, and pastoral an opposition to naturalistic or empiricist liter-

The first version of this essay emerged from a lecture given in Spring 1968 in J. M.
Holquist’s seminar on fantastic literature in the Yale University Slavic Languages and
Literatures Department. 1 have derived much profit from discussions with him. with the
late Jacques Ehrmann, my L’Mass colleague David Porter, and my McGill colleagues
Irwin and Myrna Gopnik, over and above a number of persons mentioned in my general
acknowledgements. The final version owes much to Stanislaw Lem's Fantastyka i fu-
turolagui (see Bibliography I), which considerably emboldened me in further pursuits
within this protean field, even where 1 differed from some of Lem’s emphases and
conclusions. Notes to all chapters are supplemented by the bibliographic sections to be
found at the end of the book.

3
4 ESTRANGEMENT AND COGNITION

ary genres, it differs very significantly in approach and social


function from such adjoining non-naturalistic or metaempirical
genres. Both these complementary aspects, the sociological and
the methodological, are being vigorously debated by writers and
critics in several countries, evidence of lively interest in a genre
that should undergo scholarly discussion too.
In this chapter, I will argue for an understanding of SF as the
literature of cognitive estrangement. This definition seems to possess
the unique advantage of rendering justice to a literary tradition
which is coherent through the ages and within itself, yet distinct
from nonfictional utopianism, from naturalistic literature, and
from other non-naturalistic fiction. It thus makes it possible to
lay the basis for a coherent poetics of SF.
7.2. I want to begin by postulating a spectrum or spread of
literary subject matter which extends from the ideal extreme of
exact recreation of the author’s empirical environment1 to exclu­
sive interest in a strange newness, a novum. From the 18th to the
20th centuries, the literary mainstream of our civilization has
been nearer to the first of these two extremes. However, at the
beginnings of a literature, the concern with a domestication of
the amazing is very strong. Early tale-tellers relate amazing voy­
ages into the next valley, where they found dog-headed people,

J. A benefit of discussing the seemingly peripheral subject of "science fiction" is that


one has to go back to first principles, one cannot really assume them as given. One must
ask. for example, what is literature? Usually, when discussing literature one determines
what it says {its subject matter) and how it says what it says (the approach to its themes). [ T
we are talking about literature in the sense of significant works possessing certain mini­
mal aesthetic qualities rather than in the sociological sense of everything that gets pub­
lished at a certain time or in the ideological sense of all the writings on certain themes,
this principle can more precisely be formulated as a double question. First, epistemologi­
cally, what possibility for aesthetic qualities is offered by different thematic fields (“sub­
jects“)? The answer given by the aesthetics prevalent at the moment is: an absolutely
equal possibility. With this answer the question is booted out of the field of aesthetics and
into the lap of ideologists, who pick it up by our default and proceed to bungle it.
Second, historically, how has such a possibility in fact been used? Once one begins with
such considerations, one comes quickly up against the rather unclear concept of realism
(not the prose literary movement in the nineteenth century but a metah istorical stylistic
principle), since this genre is often pigeonholed as nonrealistic. I would not object but
would heartily welcome such labels if one had first persuasively defined what is “real“ and
what is “reality.” True, this genre raises basic philosophical issues, but it is perhaps not
necessary to face them in an initial approach. Therefore I shall here substitute for
“reality” (whose existence independent of any observer or group of observers I do not at
ail doubt, in fact) the concept of "the author’s empirical environment,” which seems as
immediately dear as any.
ESTRANGEMENT AND COGNITION 5

also good rock salt which could be stolen or at the worst bartered
for. Their stories are a syncretic travelogue and voyage imaginaire,
daydream and intelligence report. This implies a curiosity about
the unknown beyond the next mountain range (sea, ocean, so­
lar system), where the thrill of knowledge joined the thrill of ad­
venture.
From Iambulus and Euhemerus through the classical utopia to
Verne’s island of Captain Nemo and Wells's island of Dr.
Moreau, an island in the far-off ocean is the paradigm of the
aesthetically most satisfying goal of the SF voyage. This is par­
ticularly true if we subsume under this the planetary island in
the aether ocean—usually the Moon—which we encounter from
Lucian through Cyrano to Swift’s niini-Moon of Laputa, and on
into the nineteenth century. Yet the parallel paradigm of the
valley, “over the range" (the subtitle of Butler’s SF novel
Erewhon) which shuts it in as a wall, is perhaps as revealing. It
recurs almost as frequently, from the earliest folktales about the
sparkling valley of Terrespial Paradise and the dark valley of the
Dead, both already in Gilgamesh. Eden is the mythological locali­
zation of utiopian longing, just as Wells’s valley in “The Country
of the Blind” is still within the liberating tradition which con­
tends that the world is not necessarily the way our present empir­
ical valley happens to be, and that whoever thinks his valley is the
world is blind. Whether island or valley, whether in space or
(from the industrial and bourgeois revolutions on) in time, the
new framework is correlative to the new inhabitants. The
aliens—Utopians, monsters, or simply differing strangers—are a
mirror to man just as the differing country is a mirror for his
world. But the mirror is not only a reflecting one, it is also a
transforming one, virgin womb and alchemical dynamo: the mir­
ror is a crucible.
Thus it is not only the basic human and humanizing curiosity
that gives birth to SF. Beyond an undirected inquisitiveness,
which makes for a semantic game without clear referent, this
genre has always been wedded to a hope of finding in the un­
known the ideal environment, tribe, state, intelligence, or other
aspect of the Supreme Good (or to a fear of and revulsion from
its contrary). At all events, the po^tbihty of other strange, co­
variant coordinate systems and semantic fields is assumed.
1.3. The approach to the imaginary locality, or localized day­
6 ESTRANGEMENT AND COGNITION

dream, practiced by the genre of SF is a supposedly factual one.


Columbus’s (technically or genologically nonfictional) letter on the
Eden he glimpsed beyond the Orinoco mouth, and Swift’s (tech­
nically nonfactual) voyage to Laputa, Balnibarbi, Glubbdubbdrib,
Luggnagg, "and Japan" represent two extremes in the constant
intermingling of imaginary and empirical possibilities. Thus SF
lakes off from a fictional ("literary”) hypothesis and develops it
with totalizing (“scientific”) rigor—the specific difference be­
tween Columbus and Swift is smaller than their generic prox­
imity. The effect of such factual reporting of fictions is one of
confronting a set normative system—a Ptolemaic-type dosed
world picture—with a point of view or look implying a new set of
norms; in literary theory this is known as the attitude of es­
trangement. This concept was first developed on non-naturalistic
texts by the Russian Formalists (“ostranenie,” Viktor Shklovsky)
and most successfully underpinned by an anthropological and
historical approach in the work of Bertolt Brecht, who wanted to
write “plays for a scientific age.” While working on a play about
the prototypical scientist, Galileo, he defined this attitude (“Ver-
fremdungseffekt”) in his Short Organon for the Theatre-. “A repre­
sentation which estranges is one which allows us to recognize its
subject, but at the same time makes it seem unfamiliar.” And
further: for somebody to see all normal happenings in a dubious
light, “he would need to develop that detached eye with which
the great Galileo observed a swinging chandelier. He was amazed
by that pendulum motion as if he had not expected it and could
not understand its occurring, and this enabled him to come at
the rules by which it was governed." Thus, the look of estrange­
ment is both cognitive and creative; and as Brecht goes on to say,
“one cannot simply exclaim that such an attitude pertains to
science, but not to“ art. Why should not art, in its own way, try to
serve the great social task of mastering Life?”2 (Later, Brecht

2. Viktor Shklovsky, "Iskusstvo kak priem," in Sborn&i po Irorii poeiicheshogo iszyka, 2


(Petrograd. ¡917). In the translation "Art as Technique," in Lee T, Lemon and Marion J.
Reis, eds., Atunon Formalist Criticism (Lincoln, NE, 1965), ostranenie is rendered so me w hat
clumsily as "defamtliariration,” See also Victor Erlich's classical survey, Russian Formalism
(The Hague, 1955).
Bertoli Brecht, "Kleines Organon fur das Theater," in his Gesammeilr Werke, 16
(Frankfurt. 1973), translated in John Willett, cd., Brecht On Theatre (New York, 1964), My
quotations arc from pp. 192 and 196 of this translation, but I have changed Mr. Willett's
ESTRANGEMENT AND COGNITION 7

would note that it might be time to stop speaking in terms of


masters and servants altogether.)
In SF the attitude of estrangement—used by Brecht in a dif­
ferent way, within a still predominandy “realistic” context—has
grown into the formal framework of the genre.
2. Science Fiction As Cognition
(Critique and Science)
2.1. The use of estrangement both as underlying attitude
and dominant formal device is found also in the imth, a “time-
less” and religious approach looking in its own way beneath (or
above) the empiric surface. However, SF sees the norms of any
age, including emphatically its own, as unique, changeable, and
therefore subject to a cognitive view. The myth is diametrically
opposed to the cognitive approach since it conceives human rela­
tions as fixed and supernaturally determined, emphatically deny­
ing Montaigne’s “la constance même n’est qu’un branle plus lan­
guissant.” The myth absolutizes and even personifies apparently
constant motifs from sluggish societies. Conversely, SF, which
focuses on the variable and future-bearing elements from the
empirical environment, is found predominantly in the great
whirlpool periods of history, such as the sixteenth-seventeenth
and nineteenth-twentieth centuries. Where the myth claims to
explain once and for all the essence of phenomena, SF first pos­
its them as problems and then explores where they lead; it sees
the mvthical static identity as an illusion, usually as fraud, at best
only as a temporary realization of potentially limitless contingen­
cies. It does not ask about The Xian or The World, but which
man?: in which kind of world?: and why such a man in such a
kind of world? As a literary genre, SF is fully as opposed to
supernatural or metaphysical estrangement as it is to naturalism
or empiricism.
2.2. SF is, then, a literary genre whose necessary and sufficient
conditions are the presence and interaction of estrangement and cogni-

translation of I'erfremdung as "alienation" into my "estrangement," since "alienation”


evokes incorrect, indeed opposite, connotations: estrangement was for Brecht an ap­
proach militating directly against social and cognitive alienation. See Ernst Bloch, "Enl-
frrmdung, t'rrfremdung: Alienation, Estrangement," in Erika Munk, ed., Brtcht (New York,
1972).
8 ESTRANGEMENT AND COGNITION

lion, and whose main formal device is an imaginative framework alter­


native to the author’s empirical environment.
Estrangement differentiates SF from the “realistic” literary
main stream extending from the eighteenth century into the
twentieth. Cognition differentiates it not only from myth, but
also from the folk (fairy) tale and the fantasy. "The folktale also
doubts the laws of the author’s empirical world, but it escapes
out of its horizons and into a closed collateral world indifferent
to cognitive possibilities, it does not use imagination as a means
of understanding the tendencies latent in reality, but as an end
sufficient unto itself and cut off from the real contingencies. The
stock folktale accessory, such as the Flying carpet, evades the
empirical law of physical gravity—as the hero evades social
gravity—by imagining its opposite. This wish-fulfilling element is
its strength and its weakness, for it never pretends that a carpet
could be expected to flv—that a humble third son could be ex­
pected to become king—while there is gravity. It simply posits
another world beside yours where some carpets do, magically,
fly, and some paupers do, magically, become princes, and into
which you cross purely by an act of faith and fancy. Anything is
possible in a folktale, because a folktale is manifestly impossi­
ble. Furthermore, the lower-class genre of folktale was from
the seventeenth-eighteenth centuries on transformed into the
more compensatory, and often simplistic, individualist fairy tale.
Therefore, SF retrogressing into fairy tale (for example, “space
opera” with a hero-princess-monster triangle in astronautic cos­
tume) is committing creative suicide.
Even less congenial to SF is the fantasy (ghost, horror, Gothic,
weird) tale, a genre committed to the interposition of anti-
cognitive laws into the empirical environment. Where the
folktale is indifferent, the fantasy is inimical to the empirical
world and its laws. The thesis could be defended that the fantasy
is significant insofar as it is impure and fails to establish a
superordinated maleficent world of its own, causing a grotesque
tension between arbitrary supernatural phenomena and the em­
pirical norms they infiltrate. Gogol’s Nose is significant because it
is walking down the Nevski Prospect, with a certain rank in the
civil service, and so on; if the Nose were in a completely fantastic
world—say H. P. Lovecraft’s—it would be just another ghoulish
thrill. When fantasy does not make for such a tension between
ESTRANGEMENT AND COGNITION 9

the supernatural and the author’s empirical environment, its


monotonous reduction of all possible horizons to Death makes of
it just a subliterature of mystification. Commercial lumping of it
into the same category as SF is thus a grave disservice and ram­
pantly socio-pathological phenomenon.
2.3, The pastoral, on the other hand, is essentially closer to
SF. Its imaginary framework of a world without money­
economy, state apparatus, and depersonalizing urbanization al­
lows it to isolate, as in a laboratory, two human motivations:
erotics and power-hunger. This approach relates to SF as al­
chemy does to chemistry and nuclear physics: an early try in the
right direction with insufficient foundations. SF has much to
learn from the pastoral tradition, primarily from its directly sen­
sual relationships which do not manifest class alienation. This
lesson has in fact often been absorbed, whenever SF has sounded
the rheme of the triumph of the humble (Restif, Morris, and
others, up to Simak, Christopher, Yefremov, etc.). Unfor­
tunately, the baroque pastoral abandoned this theme and
jelled into a conventional sentimentality, discrediting the gen­
re; but when pastoral escapes preciosity, its hope can fertilize
the SF field as an antidote to pragmatism, commercialism,
other-directedness, and technocracy.
2.4. Claiming a Galilean estrangement for SF does not at all
mean committing it to scientific vulgarization or even technologi­
cal prognostication, which it was engaged in at various times
(Verne, the United States in the 1920s and 30s, USSR under
Stalinism). The needful and meritorious task of popularization
can be a useful element of SF works at a juvenile level. But even
the roman scientifique, such as Verne’s From the Earth to the
Moon—or the surface level of Wells’s Invisible Man—though a
legitimate SF form, is a lower stage in its development. It is very
popular with audiences just approaching SF, such as the juvenile,
because it introduces into the old empirical context only one eas­
ily digestible new technological variable (Moon missile, or rays
which lower the refractive index of organic matter).3 The
euphoria provoked by this approach is real but limited, better

3. Note the functional difference from the anti-gravity metal in Wells's First Mm in the
Moan, which is an introductory or " plansi bility-t al ¡dating" device and not tiie be-all of a
much richer novel. Devices of plausibility are further discussed in chapter 4.
10 ESTRANGEMENT AND COGNITION

suited to the short story and a new audience. It evaporates much


quicker as positivistic natural science loses prestige in the
humanistic sphere after the world wars (compare Nemo’s
Nautilus as against the United States Navy's atomic submarine of
the same name), and surges back with prestigious peacetime ap­
plications in new methodologies (astronautics, cybernetics). As I
will argue in chapter 7, even in Verne the “science novel” has a
structure of transient estrangement, which is specific to murder
mysteries, not to a mature SF.
23. After such delimitations, it is perhaps possible at least to
indicate some differentiations within the concept of "cognitive­
ness” or “cognition.” As used here, this term implies not only a
reflecting of but also on reality. It implies a creative approach
tending toward a dynamic transformation rather than toward a
static mirroring of the author's environment. Such typical SF
methodology—from Lucian, More, Rabelais, Cyrano, and Swift
to Wells, London, Zamyatin, and writers of the last decades—is a
critical one, often satirical, combining a belief in the potentialities
of reason with methodical doubt in the most significant cases.
The kinship of this cognitive critique with the philosophical fun­
daments of modern science is evident.
3. The World of the Science Fiction Genre
(Concept and Some Functions)
3.0. As a full-fledged literary genre, SF has its own repertory
of functions, conventions, and devices. Many of them are highly
interesting and might prove very revealing for literary history
and theory in general. 1 shall discuss some of these—such as the
historically crucial shift of the locus of estrangement from space
to time—in the chapters that follow. I shall not, however, at­
tempt a systematic survey of such functions and devices, which
would properly be the subject of another book, one that encom­
passed modern SF as well. I should only like to mention that all
the estranging devices in SF are related to the cognition es­
poused, and that, together with the historical venerability of the
genre's tradition, this seems to me a second, methodological rea­
son for according SF much more importance than is usual in
academe. However, it might here be possible to sketch some
determining parameters of the genre.
3.1. In a typology of literary genres for our cognitive age,
ESTRANGEMENT AND COGNITION 11

one basic parameter would take into account the relationship of


the world (s) each genre presents and the “zero world” of empiri­
cally verifiable properties around the author (this being "zero" in
the sense of a central reference point in a coordinate system, or
of the control group in an experiment). Let us call this empirical
world naturalistic. In it. and in the corresponding “naturalistic” or
“realistic" literature, ethics is in no significant relation to physics.
Modern mainstream fiction is forbidden the pathetic fallacy of
earthquakes announcing the assassination of rulers or drizzles
accompanying the sadness of the heroine. It is the activity of the
protagonists, interacting with other, physically equally un­
privileged figures, that determines the outcome. However
superior technologically or sociologically one side in the conflict
may be, any predetermination as to its outcome is felt as an
ideological imposition and genoiogical impurity: the basic rule of
naturalistic literature is that man’s destiny is man.4 On the con­
trary, in the non-naturalistic, metaphysical literary genres dis­
cussed in 2.1. and 2.2., circumstances around the hero are
neither passive nor neutral. In the folktale and the fantasy, ethics
coincides with (positive or negative) physics, in the tragic myth ¡t
compensates the physics, in the “optimistic” myth it supplies the
coincidence with a systematic framework.
The world of a work of SF is not a priori intentionally oriented
toward its protagonists, either positively or negatively; the pro­
tagonists may succeed or fail in their objectives, but nothing in
the basic contract with the reader, in the physical laws of their
worlds, guarantees either. SF thus shares with the dominant lit­
erature of our civilization a mature approach analogous to that
of modern science and philosophy, as well as the omnitemporal
horizons of such an approach—aspects which will be discussed in
the following chapters.
3.2. As a matter of historical record, SF has started from a

4. In such cases as certain novels by Hardy and plays by Ibsen, or some of the more
doctrinaire works of the historical school of Naturalism, where determinism strongly
stresses circumstance at the expense of the main figures' activity, we have, underneath a
surface appearance of "naturalism," an approach to tragic myth using a shamefaced
validation for an unbelieving age. As contrary to Shakespeare or the Romantics, in this
case ethics follows physics in a supposedly causal chain (most often through biology}. An
analogous approach to fairy rale is to be found in, say, the mimicry of "naturalism" in
which Hollywood happy-end movies engage.
]2 ESTRANGEMENT AND COGNITION

prescientihc or protoscientific approach of debunking satire and


naive social critique and moved closer to the increasingly sophis­
ticated natural and human sciences. The natural sciences caught
up and surpassed the literary imagination in the nineteenth cen­
tury; the sciences dealing with human relationships might be
argued to have caught up with it in their highest theoretical
achievements but have certainly not done so in their alienated
social practice. In the twentieth century SF has moved into the
sphere of anthropological and cosmological thought, becoming a
diagnosis, a warning, a call to understanding and action, and—
most important—a mapping of possible alternatives. This histor­
ical movement of SF can be envisaged as an enrichment of and
shift from a basic direct model to an indirect model (both to be
analyzed at greater length in chapter 2). What matters here is
that the concept of a science fiction tradition or genre is a logical
corollary of the recognition of SF as the literature of cognitive
estrangement. It can be gleaned from my approach and exam­
ples that I think the literary genre which I am trying to define
embraces the subgenres mentioned in 1.1, from Greek and ear­
lier times until today (the Islands of the Blessed, utopias, fabu­
lous voyages, planetary novels, Slaatsromane, anticipations, and
dystopias—as well as the Verne-type romans scientifiques, the Well­
sian scientific romance variant, and the twentieth-century mag­
azine- and anthology-based SF semu stricto). If the argument
of this chapter holds, the inner kinship of these subgenres is
stronger than their obvious autonomous, differentiating fea­
tures. Some historical discussion of these kinships and differ­
ences will be attempted later on in this book; here I want only to
observe that the significant writers in this line were quite aware
of their coherent tradition and explicitly testified to it (the axis
Lucian-More-Rabelais-Cyrano-Swift-M. Shelley-Verne-Wells is a
main example). Also, certain among the most perspicacious sur­
veyors of aspects of the field, like Ernst Bloch, Lewis Mumford,
or Northrop Frye, can be construed as assuming this unity.
3.3. The novelty of such a concept shows most distinctly
when one attempts to find a name for the genre as it is here
conceived. Ideally this name should clearly set it apart from (I)
nonliterature, (2) the empiricist literary mainstream, and (3) non-
cognitive estrangings such as fantasy; furthermore (4) it should
try to add as little as possible to the already prevailing confusion
ESTRANGEMENT AND COGNITION J3

of tongues in this region. The academically most acceptable de­


signation has been that of a literature of utopian thought. The
concept is no doubt partly relevant, but fails to meet the first
criterion above; logically, such an approach was usually taught
and considered within the scope of either the history of ideas or
political and sociological theory. Although I would agree that
literature (and especially this genre) is most intimately involved
with life—indeed, that the destiny of humanity is its telos—I think
one should quickly add that literature is also more than an idea­
tional or sociological document. Since this is the rationale for any
systematic literary study and scholarship, I may not need to labor
the point.
The only proper way of searching for a solution seems to
require starting from the qualities defining the genre, since this
would take care of the criteria 1 to 3 at least. Taking the kindred
thesaurus concepts of science for cognition, and fiction for es­
trangement, I believe there is a sound reason for calling this
whole new genre Science Fiction (sensu. la to).
There are two main objections to such a solution. First, cogni­
tion is wider than science; I argued as much myself in 2.5. It is
much less weighty, however, if one takes "science” in a sense
closer to the German Wissenschaft, French science, or Russian
nauka, which include not only natural but also all the cultural or
historical sciences and even scholarship (cf. Literaturwissenschaft,
sciences humaines). As a matter of fact, that is what science has
been taken to stand for in the practice of SF: not only More or
Zamyatin, but the writings of Americans such as Asimov, Hein­
lein, Pohl, Dick, etc. would be completely impossible without
sociological, psychological, historical, anthropological, and other
parallels. Further, an element of convention enters into all
names (compare “comparative literature"), but it has proved
harmless as long as the name is handy, approximate enough, and
above all applied to a clearly defined body of works. The second
objection is that the use of “science fiction” confuses the whole
genre wath the twentieth-century SF from which the name was
taken. Given the advantages of the only term at hand fulfilling
the above criteria, I would argue that this is at worst a minor
drawback: nobody has serious trouble in distinguishing between
More’s book, the country' described in it, and the subgenre of
“utopia.” The trouble begins with the variety of unrelated inter­
14 ESTRANGEMENT AND COGNITION

disciplinary and ideological interpretations foisted upon such a


term; “science fiction” might perhaps escape the interdisciplinary
part of that obstacle race. Furthermore, there are always advan­
tages to acknowledging clearly one’s methodological premises. As
both Lukács and Eliot would agree, any tradition is modified and
reestablished by a sufficiently significant new development, from
whose vantage point it can be reinterpreted. This is, I would
maintain, the case with the mentioned ci-devant traditions, for
example, of “utopian literature,” in the age of science fiction. If
that is accepted, the new name is no drawback at all, but simply
an onomastic consummation.
4. For a Poetics of Science Fiction (Anticipation)
4.1. The above sketch should, no doubt, be supplemented by
a sociological analysis of the “inner environment" of SF, exiled
since the beginning of the twentieth century into a reservation or
ghetto which was protective and is now constrictive, cutting off
new developments from healthy competition and the highest
critical standards. Such a sociological discussion would enable us
to point out the important differences between the highest
reaches of the genre, glanced at here in order to define functions
and standards of SF, and its debilitating average.5
4.2. If the whole above argumentation is found acceptable, it
will be possible to supplement it also by a survey of forms and
subgenres. Along with some which recur in an updated form—
such as the utopia and fabulous voyage—the anticipation, the
superman story, the artificial intelligence story (robots, androids,
and so on), time-travel, catastrophe, the meeting with aliens, and
others, would have to be analyzed. The various forms and sub­
genres of SF could then be checked for their relationships to other
literary genres, to each other, and to various sciences. For exam­
ple, the utopias are—whatever else they may be—clearly
sociological fictions or social-science-fiction, whereas modern SF
is analogous to modern polycentric cosmology, uniting time and
space in Einsteinian worlds with different but covariant dimen­
sions and time scales. Significant modern SF, with deeper and
more lasting sources of enjoyment, also presupposes more com­

5. A first approach to the sociology of SF may be found in the special issue of


Science-Fictum Studies., November 1977. edited and with an introduction by me.
ESTRANGEMENT AND COGNITION 15

plex and wider cognitions: it discusses primarily the political,


psychological, and anthropological use and effect of knowledge, of
philosophy of science, and the becoming of failure of new realities
as a result of it. The consistency of extrapolation, precision of
analogy', and width of reference in such a cognitive discussion
turn into aesthetic factors. (I'hat is why the “scientific novel”
discussed in 2.3. is not deemed completely satisfactory—it is aes­
thetically poor because it is scientifically meager.) Once the elastic
criteria of literary structuring have been met, a cognitive—in most
cases strictly scientific—element becomes a measure of aesthetic quality,
of the specific pleasure to be'sought in SF. In other words, the cogni­
tive nucleus of the plot codetermines the fictional estrangement
itself.
2

SF and the Genological Jungle

Thanks to the Greeks, we can distinguish tragedy from comedy in


drama. . . . When we come to deal with such forms as the masque,
opera, movie, ballet, puppet-play, mystery-play, morality, com-
media dell'arte, and Zauberspiel, we find ourselves in the position
of Renaissance doctors who refused to treat syphilis because Galen
said nothing about it.
Northrop Frye

1. A View from the Mountain: Taxonomy and a System


1.0. As Northrop Frye has rightly remarked, “just as there is
nothing which the philosopher cannot consider philosophically,
and nothing which the historian cannot consider historically, so
the critic should be able to construct and dwell in a conceptual
universe of his own.”1 For the purposes of constructing the uni­
verse of this discussion, I take it (1) that no field of studies and
rational inquiry can be investigated unless and until it is at least
roughly delimited; (2) that there exist literary genres, as
socioaesthetic and not metaphysical entities; (3) that these entities
have an inner life and logic of their own, which do not exclude
but on the contrary presuppose a dialectical permeability to
themes, attitudes, and paradigms from other literary genres, sci­
ence, philosophy, and everyday socioeconomic life; (4) that the
genres pertinent to this discussion are naturalistic fiction, fan­
tasy, myth, folk tale, pastoral, and science fiction. I am assuming
that these four axioms will be justified by their cognitive yield, by
the light that they might throw upon the field of inquiry. Should
this assumption prove justified, it would go a long way toward
indicating that the basic and possibly central task of SF theory
and criticism at this historical moment is the construction of a

). Northrop Frye, .inatomy of Criticism (New York. 1966), p. 12.

16
SF AND THE GENOLOGICAL JUNGLE 17

heuristic model or models for “Science Fiction”—which is also


the hypothesis of this chapter.
A heuristic model is a theoretical structure based on analogy,
which does not claim to be transcendentally or illusionistically
“real” in the sense of mystically representing a palpable material
entity, but whose use is scientifically and scholarly permissible,
desirable, and necessary because of its practical results. An
example might be the construct according to which the
molecules of a gas behave like minuscule elastic billiard balls in
random motion. Though very little may be known or indeed
knowable about what gas molecules are “really” like, both at the
time this construct was promulgated and now it was certain be­
yond reasonable doubt that they were not elastic billiard balls of a
microscopic size. Yet this heuristic model was among the decisive
factors in the development of the whole discipline of ther­
modynamics. It had immense theoretical and practical conse­
quences, among others a giant step forward in human under­
standing of natural and perhaps even social processes. It seems
therefore unnecessary to reopen the debates of the medieval
nominalists and realists about the “real” existence of entities such
as SF or any other genre; such debates hinge on a pseudo­
question. An acceptable heuristic model or set of models for a
literary genre is as necessary for its understanding, for the set­
ting up of standards pertaining to it, as the theory of ideal gases
was for its time and discipline. In other words, however frag­
mented, laborious, or foolhardy this particular endeavor of mine
might be, the critical community concerned with SF will have to
evolve a theory of the genre which can serve as a framework for
its history and criticism. Anyway, poets—including the poets
among SF writers—have often reminded us that what the
positivistic or philistine mentality considers foolhardy is, in Gor­
ky’s words, “the wisdom of life.”
1.1 Conscious of the monsters and incubi lurkingjust beyond
my path, and averting piously my eyes from the bleached bones
of the pioneers fallen by its side, I proceed to recall my starting
point, the identifications which I worked out for the aforemen­
tioned genres in the preceding chapter. I brought forward some
arguments for their delimitation, which I shall here supply with
further argumentation and subsume under the following tax­
onomic system:
18 SF AND THE GEMOLOGICAL JUNGLE

Fiction is differentiated from other verbal structures by the


presence of a fable, plot, or narrative, through which the writer
endeavors to illuminate human relations to other people and the
universe. (At this point the normal poetological distinctions of
epic, dramatic, and lyric fiction could ensue, based on the differ­
ent stresses in the relationship of the narrator and the characters
or world of the fable, but such distinguishing does not fall within
my scope in this book. I will assume it—as well as certain other
distinctions, such as that between verse and prose—as given or at
least as for practical purposes discernible in literary theory from
Aristotle to Brecht, Frye, and Barthes, and in the literary prac­
tice which preceded the setting up of theories. My presenta­
tion has in mind at the moment epic prose—novels and
stories—only, though for all I know the resulting heuristic model
or models might have a wider scope.) Fiction, then, can be di­
vided according to the manner in which men's relationships to
other men and their surroundings are illuminated. If this is ac­
complished by endeavoring faithfully to reproduce empirical
textures and surfaces vouched for by human senses and common
sense, I propose to call it naturalisticfiction. If, on the contrary, an
endeavor is made to illuminate such relations by creating a radi­
cally or significantly different formal framework—a different
space/time location or central figures for the fable, unverifiable
by common sense—I propose to call it estrangedfiction. The nor­
mative trend of fiction after Boccaccio and Shakespeare has been
naturalistic in the above sense, though this does not at all hold
true for earlier stages of literature in our civilization nor in other
civilizations.
The world of naturalistic fiction has thus a straightforward
relationship to the "zero world” of empirically verifiable proper­
ties around the author. The ideal of Tom Jones, The Red and the
Black, Madame Bounty, TVar and Peace, The Idiot, Huckleberry Finn,
or Intruder in the Dust is to create a significant statement about the
human condition by holding a mirror to nature. In naturalistic
fiction, as in the zero world, physics stands in no significant rela­
tion to ethics. It is the activity of the protagonists, interacting
with other equally unprivileged figures, that determines the
course of narration and outcome of fable. In naturalistic fiction,
the basic rule is that man’s destiny is other humans and man­
made institutions. In such a model, relating ethics to physics
SF AND THE GEMOLOGICAL JUNGLE 19

(Hollvwoodian happy-end, say) signifies a descent into sentimen­


talism, into what is properly called sub-literature.
However, estranged fiction can quite legitimately postulate
that circumstances around the hero—according to the basic
“literary contract” making up a particular estranged genre—■
either are or are not passive and neutral. One, larger group of
estranged literary genres, which embraces various kinds of myths
and their later descendants—fantasy and folktale—is indeed de­
fined by a contract inverse to that of naturalistic fiction: their
world is actively oriented toward the hero. The folktale (Mdrchen,
later fairy tale) world is oriented positively toward its pro­
tagonist; a folktale is defined bv the hero’s triumph: magic
weapons and helpers are, with the necessary narrative retarda­
tions, at his beck and call. Inversely, the fantasy world is oriented
negatively toward its protagonist; a fantasy is defined by the
hero’s horrible helplessness. Both fantasy and folktale derive
from mythology : tne folktale from the victorious-hero myth and
the fantasy from the tragic myth. Thus, in the folktale and the
fantasy, ethics coincides with physics—positive (hero-furthering)
in the first case, and (hero-denving) in the second. In the tragic
myth ethics compensates the physics; Oedipus, Osiris or Christ
have to fail because of the empirical world they live in, but the
failure is then ethically exalted and put to religious use, usually
by postulating a metaphysical world bevond the empirical one in
which the narrative finds its true, compensatory ending. Parallel
to that, in the “optimistic” myth of Perseus, Saint George and
other light-bearing heroes, ethics not only coincides with hero­
furthering physics but also supplies a systematic cosmosociologi-
cal framework to normalize the coincidence.
The literary genres in which physics is in some magical or
religious way' determined by ethics, instead of being neutral to­
ward die hero or the total human population of the presented
world, deny the autonomy of physics and can properly be called
metaphysical. But not all estranged genres enter into such a con­
tract with their reader. Notably, the pastoral and SF worlds offer
no assurances as to the outcome of their protagonists’ endeavors.
(Phenomena such as the sentimentalized Baroque pastoral or the
“new maps of hell” of American SF represent particular, limited
historical and ideological uses which do not necessarily flow out
of the basic contract of the genre but are super added to it.)
20 SF AND THE OENOLOGICAL JUNGLE

Together with some préfigurations in the pastoral, ST is thus a


metaempirical and non-naturalistic, that is, an estranged, literary genre
which is not at the same time metaphysical. On the contrary, SF
shares with naturalistic literature, naturalistic science, and
naturalistic or materialist philosophy a common sophisticated,
dialectical, and cognitive epistemé.
The genological system discussed above can be presented
schematically by using the two parameters or binary oppositions
of naturalistic/estranged, and cognitive/noncognitive:

NATURALISTIC ESTRANGED

COGNITIVE "realistic” literature SF (& pastoral)

sub-literature metaphysical: myth,


NONCOGNÎTIVE
of “realism" folktale, fantasy

1.2. In order to test the above taxonomy, let us introduce a


new basic parameter of time and see whether the system can
make sense of it. Naturalistic literature ranges through all empir­
ical times. Though concentrating on the present, it has, parallel
with the rise of historical sciences and dialectical philosophy,
evolved the historical novel and drama, and it can even to some
degree (admittedly not to the same degree as non-naturalistic
literature) deal with the future in the form of hopes, fears,
premonitions, and dreams, as in the psychological novel be­
ginning with, say, Stendhal and Dostoevsky. Carelessness about
precise time location or restriction to a one-dimensional point­
consciousness in the present—both of which do not critically
question prevailing anthropological modes of behavior—is the
mark of the subliterature of mainstream “realism,’' from Renais­
sance street-ballads to contemporary kitsch. The metaphysical
genres shun historical time: myth is located above time, folktale
in a conventional grammatical past which is really outside time,
and fantasy in the hero’s abnormally disturbed, historiosophi-
cally dislocated present into which irrupts a “black" timelessness
or another extrahistorical time. Inversely, SF shares the omni­
SF AND THE GENOLOGICAL JUNGLE 21

temporal horizons of naturalistic literature, ranging through


all possible times. Though concentrating on the cognitively
plausible futures and their spatial equivalents, it can deal with
the present and the past as special cases of a possible historical
sequence seen from an estranged point of view—since any em­
pirical historical point or flow can be thought of as one realiza­
tion among practically innumerable possibilities. The scheme
from 1.1. sub specie temporis would thus look like this:

HISTORICAL ESTRANGED

PLURIDIMENSIONAL “realistic" literature SF

ONE DIMENSIONAL
sub-literature myth,
of “realism” folktale, fantasy

It is not surprising to anybody who has read Marx, Hegel, or


Augustine of Hippo that naturalistic in the temporal sphere
means historical. It is more interesting to note that temporal cog­
nition is allied to a free movement back and forth in time. Myth
in its timeless suffering or bliss, folktale in its world apart allied
to the empirical world by a grammatical past, and fantasy as the
present lifted out of time into black transcendency—all share the
impossibility of such a humanizing movement. Out of their sev­
eral shortcomings they have, as is known, made tremendous
virtues; yet the limitations remain.
2. An Ecological Jungle Trip:
Symbiosis, Parasitism, Mimicry, and Sundry
2.0. So far my analysis has been conducted on a level which,
no doubt, was abstracted from actual historical literary genres
but one which endeavored to treat them as ideal types or pure
heuristic models. In actuality, a particular work, literary opus,
trend, or school is almost never entirely pure. Literary genres
exist in historically precise and curious ecological units, interact­
ing and intermixing, imitating and cannibalizing each other. To
understand what one really has in mind when talking about SF,
22 SF AND THE GENOLOGIGAL JUNGLE

it is necessary to continue the analysis on the level of actual


happenings in the noncanonic literature or paraliterature of this
century. Only such a path, descending from the clear mountain
sights and its wide horizons into the luxuriant and steamy jungle
of literary genres, and supplementing an aerial survey with ac­
tual botanizing in the field, has a chance of leading to useful
results.
2.1. The relationship of SF to naturalistic literature, usually to
the species of adventure-journey, is by now relatively clear and can
be dealt with briefly. It is a relationship of filiation, best evi­
denced in the work of Jules Verne: SF has historically had one of
its roots in the compost heap of such juvenile or popular sublit­
erature, and in order to develop properly it has had to subsume
and outgrow it—the quicker the better for its generic affirma­
tion. It found congenial or congeneric elements in the cognitive
and marvelous bias of the voyage extraordinaire and its catalogues
of wonders seen along Ulysses’ or Captain Nemo’s way. The sea
haunts this filiation, the island story is its microparadigm or root
situation, and locomotion the connecting thread of its narration.
All the marvelous interstellar SF voyages and quests in Heinlein,
Blish, Van Vogt, and a thousand others, the Nietzschean, Co­
lumbian, or Sindbadian poetry of navigation—navigare necesse,
vivere non necesse—belong here. Such voyaging is an honorable,
though in retrospect one can scarcely fail to note that it is an
initial (and for the reader initiatory), function of SF. It acts much
in the way that a true long voyage does in the zero world,
dialytically—estranging the reader from familiar and usually
contemptible shores, dissolving his umbilical connections with
old and firm earth (or Earth), preparing him to accept the mar­
velous beyond seven seas or galaxies. When unduly prolonged,
this adolescence of SF means arrested development. It should be
kept in its proper humbly useful place in the ontogenetic de­
velopment of the reader as well as in the phylogenetic develop­
ment of the genre.
In dose proximity to the didactic aspect of the journey is the
papular science compost heap which can be found next to the
adventure-journey heap in the early phylogenetic stages of SF
from technologically developed countries. Verne used both, add­
ing a dash of puzzle in the manner of Poe and a barrelful of
Saint-Simonian romanticism. Unalloyed, or alloyed with the
SF AND THE OENOLOGICAL JUNGLE 23

baser metal of subliterary conflict and sentiment, this leads no


further than to a primitive technological or at best technocratic
extrapolation, as evidenced in Bacon’s New Atlantis, then in
Gernsback and the "SF reservation” between the two world wars.
A hybrid results that is neither good fiction nor interesting sci­
ence; it is dislodged the first time the shapers of public and
publishing opinion happen to read Wells—or, indeed, a good
straightforward essay of scientific popularization, which has from
the time of Friedrich Engels and Thomas Huxley been im­
measurably more exciting and less reactionary than Ralph 124C
41 +. Of course, it usually takes those shapers a generation or
two to acquire the necessary taste in reading. In the meantime,
the Gernsbacks keep SF alive at the cost of starving, stunting,
and deforming it; comparing The Iron Heel with the output in the
United States between the World Wars, one strongly suspects the
cost is too high.
2.2. In 2.1. it was discussed how older paradigms of marvel­
ous voyage, popular science essay, and individualist subliterature
(the Western and the sentimental story) interfere with the forma­
tion of an autonomous SF paradigm or model if their grip is not
loosened quickly. Unfortunately, a majority of what is published
as SF is still in that prenatal or, better, regression-to-womb stage:
it is simply the Western or some kindred sub-literary species
masquerading its structures—generally for venal and ideological
reasons—under the externals of SF: rockets, ray-guns, monsters,
or in the last dozen years their slightly more sophisticated equiva­
lents. Usually the symbiosis of popular science and juvenile ad­
venture finds it impossible to mimic SF without regressing into
their homologue of the fairy tale, with its victorious hero, foiled
villain, damsel in distress, and quaint helpers or marvelous help­
ing objects. Such sub-Vernean or Gernsbackian SF does not
change the fairy tale structure but only the motivation of its
devices: it pretends to explain away the supernatural by reassign­
ing it to natural science and noble scientists (who are energetic
and sentimental if young and in love with, absent-minded if old
and fathers of, the eternal feminine). However, the science is
treated as a metaphysical and not physical, supernatural and not
natural activity, as gobbledygook instead of rational procedure.
From Ralph, Buck Rogers, and the post-Stapledonian supermen
to Asimov’s psychohistoryr (which has at least the advantage of
24 SF AND THE GENOLOG1CAL JUNGLE

identifying the proper field of modern destiny, social relations),


such metaphysical gobbledygook vitiates some of the best-known
SF works. Neither cognitive nor magical but shamefacedly pass­
ing off a juvenile idea of magic for cognition, equating the
photon rocket with the flying carpet and global social destinies
with the victory of the third son, such a mimicry is like the newly
fashionable pop wines: a hyping-up of the old grape juice into
the new wine. In the perfectly just world of taste and poetic
creativity, this procedure reaps the reward of hypocrisy: fairy
talc readers rightly prefer the classics, sophisticated SF readers
disbelieve the fairy tale. Inversely, in the very imperfectly re­
tributive world of social taste and commercial SF, such a proce­
dure breeds generations of readers with juvenile taste, unable to
develop the standards by which to judge SF (not to mention
empirical human relations).
2.3. The more ambitious reader and writer cannot for long
be satisfied with such pap. Yet trying to find a fresh tack in the
cruel world of instant obsolescence, SF often veers from Scylla to
Charybdis. A further step down into pseudo-sophistication—
correlative, no doubt, to a marked decadence of cultural taste in
bourgeois society and its literary markets—is the parasitism of
Gothic, horror, and weird fantasy upon SF. Such fantasy is
characterized, as I have said, by the irruption of an anti-cognitive
world into the world of empirical cognition. One can understand
some readers’ panic flight from a science which produces nuclear
bombs, napalm, and nerve gases, from a reason which justifies
class societies in mutual balances of terror, condemning two-
thirds of the world to hunger and disease, and the remaining
third—“hypocrite lecteur, mon semblable, mon frère”—to the
boredom of a nine-to-five drudgery relieved by flashes of TV
commercials. Maybe such readers ought to have an escapist en­
clave of sword-and-sorcery or Cthulhu cosmologies—I cannot
say. But surely SF, built upon the premise that nature is neither
a childishly wicked stepmother (“As flies to wanton boys are we
to gods / They kill us for their sport”) nor inscrutably alien to
man—surely SF cannot allow its contract with the reader to be
contaminated by the Great Pumpkin antics of fantasy. Even
more perniciously than is the case with the bland fairy tale struc­
ture, the black ectoplasms of fantasy stifle SF completely. Its time
shrinks to the point-consciousness of horror, gloom, and doom,
SF AND THE GENOLOGICAL JUNGLE Ï5

its daydreams turn into an inchoate nightmare, and under the


guise of cognition the ancient obscurantist enemy infiltrates its
citadel. Fossilized fragments of reasoning are used to inculcate
irrationality, and the social energy of readers is expended on
Witches’ Sabbaths instead of focusing it on the causes for our
alienating, murderous, and stultifying existences: the power
structures holding back the hominization of the sapiens, the true
demonology of war and market breeding pride and prejudice.
At its best, in Swift and Cyrano, in Jack London and the dysto­
pian “new maps of hell,” in Lucian and Wells, in the great Uto­
pians and Zamyatin, SF has with different degrees of precision,
but with unerring precision of orientation, focused on these
power structures, on such demonology. It is at its worst, at its
most alienated and alienating, when it honors the parasitism and
vampirism of fantasy.
2.4. There has also been a great deal of talk about affinities
between SF and the mythological tale. Though also a story about
supernatural events involving superhuman figures, as different
from other metaphysical fiction (folktale and fantasy), the events
and figures of this genre form a systematic whole, a mythological
edifice of tales whose norms are supposed to have supertemporally
(timelessly or continuously) determined man’s basic relations to man and
nature. Obviously, all religious systems are in this sense mytholog­
ical. On the contrary, plays and stories are neither myth nor
ritual but fictional literature, although myths and rituals may
underlie their forms, plots, and sometimes their characters. For
example, Murray has convincingly shown that the forms of Attic
tragedy derive from Dionysian sacrificial rituals, and Cornford
has done an analogous job for Attic comedy? The Hellenic
tragic characters derive primarily from Homer, but through him
from other sacrificial rituals, which is why Homerian themes
fitted so well into the mythic pattern of tragedy. Thus,Jict:on can
be formally or morphologically analogous to myth, but it is not itself myth.
It uses mythical morphemes for nonmythic and—except in

2. Gilbert Murray, "Hamlet and Orestes," in his The Classical Tradition of Poetry (New
York, 1968). and "Excursus on the Ritual Forms preserved in Greek Tragedy," in jane
Ellen Harrison. Epiltgomena to tht Study of Greek Religion—Themis (New York. 1966); F. M.
Cornford, The Ongin of Attic Comedy (Gloucester, MA. 1966). See also other anlhropologi-
cal works by the Cambridge School that, as far as literary studies are concerned, culmi­
nate in George Thomson’s elegant Aeschylus and Athens (New York, 1968).
86 SF AND THE GEMOLOGICAL JUNGLE

folktale, fantasy, and subliterature—for anti-mythic ends. “Myth


and literature are separate and autonomous entities, though any
specific myth text can and should be considered as folk­
literature."34However—and this is in itself highly important and
largely justifies the attention that modern scholars have devoted
to myth—bearing in mind the caveats and distinctions discussed
earlier, it should be acknowledged that important aspects of lit­
erature (primarily, many basic and possibly most significant
plots) are mythomorphic. What a writer like Faulkner or Kafka
creates is not a myth but a personal fictional statement formally
analogous to myth in a radically different and indeed incompati­
ble cosmological or ideological context. In other words, a realistic
parable such as The Bear or an SF parable such as The Metamor­
phosis, although it uses a mythological bestiary as well as the
mythological pattern of trial and death with or without resurrec­
tion, is in its message and final impact very different from, often
diametrically opposed to the religious myth expressing a collec­
tive static vision. Kafka and Faulkner are—they cannot but be—
historical writers.
Obviously, SF will be as mythomorphic in some basic pat­
terns as other fictional genres are. Beyond that, SF shares with
myth the fictional estrangement, the “outer limits of desire” as
Professor Frye aptly formulated it/ and its formal closeness to
myth will extend beyond plots to many characters and situations,
But all attempts to transplant the metaphysical orientation of
mythology and religion into SF, in a crudely overt way as in C.S.
Lewis, Van Vogt, or Zelazny, or in more covert ways in very many
others, will result only in private pseudomyths, in fragmentary
fantasies or fairy tales.5 As 1 mentioned in my first chapter, myth
absolutizes and even personifies apparently constant motifs from
periods with sluggish social dynamics, and claims to explain the
eternal essence of phenomena. On the contary, SF claims to or­
ganize variable spatiotemporal, biological, social, and other
characteristics and constellations into specific fictional worlds and

3. Stanley Edgar Hyman, "The Ritual View of Myth and the Mythic," in Thomas A.
Sebeok. ed., Myth (Bloomington, 1970), p, 151.
4. Frye, p. 136.
5. See Harry Levin, "Some Meanings of Myth,” in Henry A, Murray, ed„ Myth and
MylAwioAing (Boston, 1969), pp. 111-12.
SF AND THE OENOLOGICAL JUNGLE 27

figures. Mathematically speaking, myth is oriented toward con­


stants and SF toward variables.
On a different level of fictional structuring, however, is the
treatment of religious beliefs or mythic situations as historical
material. When such mvthic elements are—by transposition, as it
were, into the demystifying key of SF—extracted from a
mythological paradigm and fitted into an SF one, what results is
perfectly legitimate, often first-class SF. As always, the critic will
in any particular instance have to rely on his literary tact and
sense of measure to pierce this intricate double mimicry and
parasitism, to decide with which type of interaction between SF
and myth he is faced. To mention only two favorites of mine,
Stapleclon and Walter Miller, Jr., I believe that at a certain point
(say in The Flames) Stapledon crosses the divide into pseudomyth,
that is, into fantasy, and that Miller does the same at the resolu­
tion of A Canticle for Leibowitz with the character of Mrs. Grales.
At such points the ideological attraction to myth as world view
and not as formal pattern got the best of the SF writer.
3. To Greener Fields and Pastures New:
The Extrapolative and the Analogical Models of SF
3.0. I would like now to try emerging from the jungle into
the cultivated territory of selected SF, and analyze what look to
be its two main species or models, the extrapolative and the
analogical one.
3.1. SF written from, say, the period of the French Revolu­
tion on (though not necessarily in preceding epochs) has come to
be considered as starting from certain cognitive hypotheses and
ideas incarnated in the fictional framework and nucleus of the
tale. This extrapolative model—of Mercier’s L’An 2440, London’s
Iron Heel, Wells’s When the Sleeper JVoAes and Men Like Gods,
Zamyatin’s We, Stapledon’s Last and First Men, Yefremov's An­
dromeda, Pohl and Kornbluth’s Space Merchants, or Brunner’s The
Jagged Orbit—seems based on direct, temporal extrapolation and
centered on sociological (that is, utopian and anti-utopian)
modeling. This is where the great majority of the “new maps of
hell” is taken to belong for which postwar SF is justly famous, in
all its manifold combinations of sociotechnological scientific cog­
nition and social oppression (global catastrophes, cybernetics,
dictatorships).
28 SF AND THE OENOLOGICAL JUNGLE

Yet already in Wells’s Time Machine and in Stapledon, this ex­


trapolating transcended the sociological spectrum (from every­
day practice through economics to erotics) and spilled into
“billion-year” biology and cosmology. The ensuing radical es­
trangements can, no doubt, be anticipated in a chronological fu­
ture, but they cannot, scientifically speaking, be extrapolated. By
this token, futuristic anticipation reveals that extrapolating is a
fictional device and ideological horizon rather than the basis for
a cognitive model. It is thus dubious—as will be discussed further
in chapter 4—that significant SF could be simply extrapolation.
Nonetheless, whatever its ostensible location (future, “fourth di­
mension,’’ other planets, alternate universes), the self-under­
standing of much SF—as shown in the historical section of
this book—was uneasily futurological. Being written in a histori­
cal epoch dominated by anticipatory expectations, this SF de­
manded to be judged by the “scientific” import of the tale’s
premises and the consistency with which such premises (usually
one or very few in number) were narratively developed to their
logical end, to a “scientifically valid” conclusion.
SF could thus be used as a handmaiden of futurological
foresight in technology, ecology, sociology, and so on. Whereas
this may at times have been a legitimate secondary function the
genre could be made to bear, any forgetfulness of its strict se­
condariness leads to confusion and indeed danger. Ontologically,
art is not pragmatic truth nor is fiction fact. To expect from SF
more than a stimulus for independent thinking, more than a
system of stylized narrative devices understandable only in their
mutual relationships within a fictional whole and not as isolated
realities, leads insensibly to the demand for scientific accuracy in
the extrapolated realia. Editors and publishers of such “hard”
persuasion, from U.S. pulp magazines to the Soviet Agitprop,
have been inclined to depress the handmaiden of SF into the
slavey of the reigning theology of the day (technocratic, psionic,
utopian, catastrophic, or whatever). Yet this fundamentally sub­
versive genre languishes in straitjackets more quickly than most
others, responding with atrophy, escapism, or both. Laying no
claim to prophecies except for its statistically probable share, SF
should not be treated as a prophet: it should neither be en­
throned when apparently successful nor beheaded when appar­
ently unsuccessful. As Plato found out in the court of Dionysius
SF AND THE GENOLOGICAL JONGLE 29

and Hythloday at Cardinal Morton’s, SF figures better devote


themselves to their own literary republics, which, to be sure, lead
back—but in their own way—to the Republic of Man. SF is finally
concerned with the tensions between Civitas Dei and Civilas Ter-
rena, and it cannot be uncritically committed to any momentary
city.
3.2. The analogic model of SF is based on analogy rather
than extrapolation. Its figures may but do not have to be an­
thropomorphic or its localities geomorphic. The objects, figures,
and up to a point the relationships from which this indirectly
modeled world starts can be quite fantastic (in the sense of em­
pirically unverifiable) as long as they are logically, philosophi­
cally, and mutually consistent. The analytic model can thus
comprehend the extrapolative one, but it is not bound to the
extrapolative horizon.
The lowest form of analogic modeling is that in which an
extrapolation backwards is in fact a crude analogy to the past of
the Earth, from geological through biological to ethnological and
historical. The worlds more or less openly modeled on the Car­
boniferous Age, on tribal prehistory, on barbaric and feudal
empires—in fact modeled on handbooks of geology and an­
thropology, on Spengler’s Decline of the West and Dumas pere’s
Three Musketeers—are unfortunately abundant in the foothills of
SF. Some of this may lie useful adolescent leisure reading, which
one should not begrudge; however, the uneasy coexistence of
such worlds with a superscience, which is supposed to provide an
SF alibi, largely or wholly destroys the story’s cognitive credibil­
ity. The E.R. Burroughs-to-Asimov space opera, cropping up in
almost all U.S. writers right down to Samuel Delany, belongs to
the uneasy territory between inferior SF and non-SF—to forms
that, as I argued earlier, mimic SF scenery but are modeled on
the structures of the Western and other avatars of fairy tale
and fantasy.
The purest form of analogic modeling w ould be the analogy to
a mathematical model, such as the fairly primary one explicated
in Abbott’s Flatland, as well as the ontological analogies found in
a compressed overview form in some stories by Borges and Lem.
A somewhat more humane narration with a suffering pro­
tagonist is to be found in, say, ¿apek’s Krakatit or Le Guin’s Left
Hand of Darkness, and even more clearly in Kafka’s Metamorphosis
30 SF AND THE GENOLOGICAL JUNGLE

or In the Penal Colony and Lem’s Solaris. Such highly sophisticated


philosophico-anthropological analogies are today perhaps the
most significant region of SF, indistinguishable in quality from
other superior contemporary writing. Situated between Borges
and the upper reaches into which shade the best utopias, anti-
utopias, and satires, this semantic field is a modern variant of the
“conte philosophique” of the eighteenth century. Similar to
Swift, Voltaire, or Diderot, these modem parables fuse new visions
of the world with an applicability—usually satirical and
grotesque—to the shortcomings of our workaday world. Depart­
ing from the older rationalism, a modern parable must be
open-ended by analogy to modern cosmology, epistemology, and
philosophy of science.”
The analogic model of SF Jails, however, clearly within cogni­
tive horizons insofar as its conclusions or import is concerned.
The cognition gained may not be immediately applicable, it may
be simply the enabling of the mind to receive new wavelengths,
but it eventually contributes to the understanding of the most
mundane matters. This is testified by die works of Kafka and
Twain, Rosny and Anatole France, as well as of the best of Wells
and the "SF reservation” writers.
4. The Jungle Explorer: .Medicine Man or Darwinist
4.0. Thus far I have not explicitly referred to the theory and
practice of SF criticism, since it is impossible to discuss an intel­
lectual activity before its field has been determined. The field of
SF criticism is SF, and this truism becomes significant when we
pause to consider how little agreement there is about the basic
parameters of SF. Having discussed them, in the remainder of
this chapter I would like to essay some remarks on SF criticism.
They will have to be as disjointed, tentative, and unsystematic as
that criticism, since the basic lesson one can draw from the his­
tory of literary criticism is that it is difficult for criticism to be
more significant than the works it criticizes.

6, I have al templed lo analyse some representative examples of such modern SF


parables in chapters 10 and 12 of this book, ä fnopos of Wells's Time Machine and Óapck's
War IVitA the Newts, in my afterword to Stanislaw Lem, Solaris (New York, 1971 and 1976),
enlarged into a parallel to US and Russian examples in "Stanislaw Lem und das mit­
teleuropäische soziale Bewusstsein der Science-fiction," in Werner Berthe!, ed., Insel AT
SF AND THE GENOLOGICAL JUNGLE 31

4.1. Beyond the necessary but subsidiary critical activity of


reviewing and chronicling, it seems that the most fashionable
critical approach to SF is that of mythical analysis. In order to
comment upon it, I shall have to try to disentangle the main
meanings of this protean and tantalizing term.
Few writers considering myth in the last third of a century
have failed to lament the divergent and indeed incompatible
meanings given to this term in different professional and
ideological fields of discourse. Though everyone—including
myself—has to try to group these meanings for purposes of an
overview, it is sometimes difficult to escape the conclusion of a
philologist that there are as many interpretations of myth as
there are critics. In ethnology “myth” is indistinguishable from
“legend" or "folklore.” Cultural historians “employ ‘myth’ with
the quite separate meaning of a popularly accepted cluster of
images.”* 7 I'he term can also be loosely used to mean “talc, fan­
tasy, mass delusion, popular belief and illusion, and plain lie”;
an essay as early as 1947 reduced this confusion of tongues to
the absurd by adopting the title of “The Modern Myth of the
Modern Myth.”8 But, cutting a long story short, it seems to me
that the literary theoretician has presently to deal with three
principal views of the field: that of Cassirer and his followers,
that of literary scholars who consider all literature to be some
kind of myth—a view most ably and influentially formulated by
Northrop Frye—and that of a third group which would insist, as
I argued earlier (see note 2), that literary artifacts are not myths
and yet that many of them are significantly marked by genetic
and morphological connections with myths.
4.1.1. Cassirer treats myth as a kind of symbolic vision cor­
relative to the mvthopoeic mode of consciousness, “inythopoeia”
meaning the world view and forms of expression characteristic
of a hypothetical early stage of culture “when language is still
largely ritualistic and prelogical in character.” In this view, myth

manach auf das Jahr 1976—Stanislaw Lem (Frankfurt, 1976); and in essays on Philip K.
Dick and Ursula K. Le Guin, reprinted in Mullen and Suvin, eds. (see Bibliography I).
7. Richard M. Dorson, “Theories of Myth and the Folklorist,” in Murray, ed., p. 84.
8. First quotation from Hyman, in Sebeok, ed.. p. 153; see also, for a psychologist’s
attack on loose definitions of myth, Henry A. Murray, “The Possible Nature of a
’Mythology' to Come,” in Murray, cd., p. 303. The second quotation is the tide of Donald
A. Stauffer's essay in English Institute Essays 1947 (New York, 1948).
32 SF AND THE GENOLOGICAL JUNGLE

“is simply a basic way of envisaging experience and carries no


necessary connotation of storytelling.’’9 Rather, all creative, poet­
ic, metaphoric thinking is “mythical.” To this it must be briefly
objected that metaphor is feasible only when some cognitively
defined terms with fixed meanings are available as points of
comparison, and that as far as literature is concerned poetic
metaphor and language begin exactly where mythology ends. In
the best mythical fashion, if poetry springs from the mother-soil
of mythology, it does so only by spurning or destroying its par­
ent. Finally, if everything (including science, philosophy, the arts,
and ail other aspects and motives of social practice) is myth or
mythopoeia, if in myth, as Cassirer says, “everything may be
turned into everything,”1011 then this term loses all usefulness for
distinguishing literature from anything else, let alone for any
distinctions within literature itself. Historically hypothetical,
philosophically idealistic, and aesthetically useless, Cassirer’s
hypothesis for all its influences in the American cultural climate
after World War II (for example, Susanne Langer) cannot con­
tribute to our present needs.
4.1.2. At the opposite extreme—but les extrêmes se rejoignent—
is the position which preserves the autonomy of literary studies
but affirms that myth is story and any story is myth. It possesses
a heroic paradigm in Frye’s Anatomy of Criticism. Though men­
tioning the secondary sense of myth as “untruth,”11 and of
“myth in the narrower and more technical sense” as stories about
“divine or quasi-divine beings and powers,”12 and then discuss­
ing a mythical phase or context of literary art which is primarily
concerned with “poetry as the focus of a community,”13 Frye
concentrates on a Cassirerian “mythical view of literature” which
leads “to the conception of an order of nature as a whole being
imitated by a corresponding order of words.”14 This is based on

9. P(hilip) Wfheelw right], "Myth,” in Alex Preminger, ed., Encyclopedia of Poetry and
Poetics (Princeton, 1965), pp. 538-89; see Ernst Cassirer, An Essay on Man (New Haven,
1962) and The Philosophy of Symbolic Forms, vol. 2 (New Haven, 1955).
10. Cassirer. Essay, p. 81.
11. Frye. p. 75.
12. Frye, p. 116; see also, on "the mythical or théogonie mode,’' pp. 120, 83-36, et
passim.
13. Frye, p. 99; see the whole section, pp. 95-99.
14. Frye. P- 118.
SF AND THE GENOLOGICAL JUNGLE S3

his belief, explicated in the section subtitled “Theory of Myths,”


that “in myth we see the structural principles of literature iso­
lated.”15 If structural principles are to mean isolatable formal
narrative patterns, this is acceptable as a basis of discussion sub­
ject to historical verification. However, if they are also meant to
subsume the motivation of a literary work, what the Theory of
Literature calls “the inner structure of psychological, social, or
philosophical theory of why men behave as they do—some
theory of causation, ultimately,”16 then 1 do not see how myth
can contain the structural principles of all literature or be the
"total creative act” which could account for all basic components
of the final impact or message of all literary modes and genres.
In other words, among many brilliant insights in Anatomy of
Criticism there is one about mythical patterns not only being for­
mally analogous to basic patterns in other literary modes—which
one would a priori expect in the imaginative products of the
same human species—but also being more clearly identifiable in
supernatural stories “at the limits of desire"17 than in stories
cluttered with surface naturalism. However, there is an essential
difference between this and treating the fourfold seasonal
mythos of Spring, Summer, Autumn, and Winter as the basic
organization of all literature and indeed all verbal structures im­
aginable, including science and history.18 Here the formal simi­
larity has been left behind, and literature has (by way of a seman­
tically redefined mythos) been identified to myth tout court, since
its original meaning of superhuman story has not been aban­
doned.1“ Unfortunately, this is the most easily vulgarized and
therefore possibly the best-known part of Frye’s book. Logically,
literature and verbal structures in general are finally reduced to
a central unifying myth, adumbrated in Milton and Dante but
fully manifest in the Bible, which is a “definitive” myth.20 All
writing, one might therefore expect, has in the past aspired to

15* Fryt, p. 136*


16. René We I lek and Austin Warren. Theory of Literature (Harmondsworth, 1973), p.
207 el passim.
17. Frye, p. 134.
18. Frye. p. 34 i el passim.
19. Frye: redefining mythos, pp. 134-40 and 158 ff. ; retaining the meaning of super­
human tale, e.g., p. 317.
20. Frye, pp. 120-21; also p. 315, 325 el passim.
54 SF AND THE OENOLOGICAL JL'NGLE

and will in the future be confined to variations on smaller or


larger bits of the Christian myth of salvation. Obviously such a
conclusion will finally be shared only by those who acknowledge
the hegemony of a cyclical theory’ of history and a closed
cosmology—that is, by anti-utopians. Therefore, this brilliant
work can persuade us that much literature is morphologically
informed by patterns which we might perhaps call mythical.
However, “mythical” then proves to be simply shorthand for
“basic narrative patterns which are seen at their clearest in some
myths.”
4.1.3, For, when we have rendered unto myth what is of the
myth, we must recognize that finally, for a cognitive pursuit such
as literary theory and criticism, myth as an instrument is fairly
limited. Philosophically, myth is an evasion of precise distinctions
and of full intellectual commitment: a myth is not true or false
but believable or unbelievable, vital or dead. On its own grounds
it is irrefutable, for as soon as it is queried as to its truth it is not
treated as myth but as historical cognition or formal hypothesis.
In other words, it seems to me that Frye has rendered a signal
service to poetics by his formal hypothesis, but I find myself
unpersuaded by his historical premises and his semantical glid­
ing between myth as a historical genre, mythos as a formal
paradigm, and both of them as a “structural principle or at­
titude.”21 I am unable to accept the conclusion that “in literary
criticism, myth ultimately means mythos, a structural organizing
principle of literary form,"22 which does not differentiate be­
tween the formal and structural functions of myth.
As distinct from Cassirer and the Cassirerian aspect of Frye, it
seems to me that myth cannot constitute a useful theory of his­
tory in general, and artistic or literary history in particular. Myth
is parascientifie and sometimes prescientific in its interpretations
of nature and society. Although some among its numerous con­
figurations are statistically bound to become precursors of scien­
tific ones, it is essentially an insufficiently critical human experience
which, for all its ideological and artistic uses, cannot be dignified
as anything more than a first significant step on the human way
to a cognition of reality. Speaking of the myth’s "unity of feel­

21. Frye, p. 310.


22. Frye, p. 341.
SF ANE THE GE NO LO GIC Al. JUNGLE 35

in g,” Cassirer rightly concludes that its pragmatic function is to


promote social solidarity through feelings of cosmic sympathy at
the time of social crisis.23 Myth embodies and sanctions au­
thoritarian social norms and the basic institutions which deter­
mine the life of each member of a certain collective authority­
structure. It is intrinsically—-whatever its surface innovations in
this age where every new car fashion is “revolutionary"—a con­
servative force, a guarantee of the status quo (say of the mass
existence of private cars). In the forceful words of David Bidney:
To my mind, contemporary philosophers and theologians,
as well as students of literature in general, who speak of the
indispensable myth in the name of philosophy and religion,
and anthropologists and sociologists who cynically approve
of myth because of its pragmatic social function, are under­
mining faith in their owm disciplines and are contributing
unwittingly' to the very degradation of man and his culture
which they otherwise seriously deplore. Myth must be taken
seriously as a cultural force but it must be taken seriously
precisely in order that it may be gradually superseded in the
interests of the advancement of truth and the growth of
human intelligence. Normative, critical, and scientific
thought provides the only self-correcting means of combat­
ing the diffusion of myth, but it may do so only on condition
that we retain a firm and uncompromising faith in the integ­
rity of reason and in the trans-cultural validity of the scien­
tific enterprise.24
Thus, the literary scholar and critic, building his autonomous
and yet rational conceptual world, must honor myth, in the Fry-
gian "narrow sense” of stories about superhuman beings, as both
occasionally fetching folk poetry and a reservoir of literary
forms. At the same time, the critic—and in particular the critic of
SF—must, I believe, abandon the belief that he has done much
more than his formal homework when he has identified Yef­
remov’s Andromeda as containing the myth of Perseus or Delany’s
Einstein Intersection and Verne’s Château des Carpates as contain­
ing the myth of Orpheus. He is still left face to face with the

23. Cassirer, £ssay, pp. 79-84.


24. David Bidney, "Myth. Symbolism, and Truth." in Sebeok. ed., p. 23.
S6 SF AND THE GENOLOGICAL JUNGLE

basic questions of his trade, namely, is the myth or mytheme


transmuted (1) into valid fiction; (2) into valid science fiction?
"Mythical analysis” as a self-sufficient critical method collapses at
this point; as an ideology7 it remains a contributing factor to the
Babylonian confusion of tongues, a particularly lethal quicksand
region on the path to SF.
4.2. Finally, it might be possible to sketch the basic premises
of a significant criticism, history, and theory of this literary
genre. From Edgar Allan Poe to Danton Knight and Stanislaw
Lem, including some notable work on the other subgenres from
the utopias to Wells and some general approaches to literature
by people awake to methodological interest, much spadework
has been done. If one may speculate on some fundamental fea­
tures or indeed axioms of such criticism, the first might be that
the genre has to be and can be evaluated proceeding from its
heights down, applying the standards gained by the analysis of
its masterpieces. We find in SF, as we do in most other genres of
fiction, that 80 to 90 percent of the works in it are sheer confec­
tionery. However, contrary to subliterature, the criteria for the
insufficiency of most SF are to be found in the genre itself. This
makes SF in principle, if not yet in practice, equivalent to any
other "major" literary genre. The second axiom of SF criticism
might be to demand of SF a level of cognition higher than that
of its average reader: the strange novelty is its raison d’etre. As a
minimum, we must demand from SF that it be wiser than the
world it speaks to.
In other words, this is an educational literature, hopefully less
deadening than most compulsory education in our split national
and class societies, but irreversibly shaped by the pathos of,
preaching the good word of human curiosity, fear, and hope.
Significant SF denies thus the “two-cultures gap” more efficiently
than any other literary genre I know of. Even more importantly,
it demands from the author and reader, teacher and critic, not
merely specialized, quantified positivistic knowledge (scientia) but
a social imagination whose quality of wisdom (sapientia) testifies
to the maturity of his critical and.creative thought. It de­
mands—to conclude the botanical marvelous voyage of this chap­
ter—that the critic be a Darwinist and not a medicine-man.
3
Defining the Literary Genre of Utopia:
Some Historical Semantics, Some
Genology, a Proposal, and a Plea

For if the matter be attentively considered, a sound argument may


be drawn from Poesy, to show that there is agreeable to the spirit of
man a more ample greatness, a more perfect order, and a more
beautiful variety than it can anywhere (since the Fall) find in na­
ture. ... it [Poesy] raises tlie mind and carries it aloft, accommodat­
ing the shows of things to be desires of the mind, not (like reason
and history) buckling and bowing down the mind to the nature of
things.
Francis Bacon
■'Utopia,” the neologism of Thomas More’s, has had a singu­
larly rich semantic career in our time. Having at its root the
simultaneous indication of a space and a state (itself ambiguously
hovering between, for example, French état and condition) that
are nonexisting (ou) as well as good (eu), it has become a territory
athwart the roads of all travelers pursuing the implications of the
question formulated by Plato as “What is the best form of or­
ganization for a community and how can a person best arrange
his life?”1 And have not the urgencies of the situation in which
the human community finds itself made of us all such travelers?
Utopia operates by example and demonstration, deictically. At
the basis of all utopian debates, in its open or hidden dialogues,
is a gesture of pointing, a wide-eyed glance from here to there, a
“traveling shot” moving from the author’s everyday lookout to
the wondrous panorama of a far-off land:
But you should have been with me in Utopia and personally
seen their manner and customs as I did. . . . [More, Utopia,
book 1]

1. Lams 3, 702b. See Plato, The Laws, trans, with introduction by A. E. Taylor (Lon­
don, 1960), p. 85.

37
38 DEFINING THE LITERARY GENRE OF UTOPIA

... it was winter when I went to bed last night, and now, by
witness of the river-side trees, it was summer, a beautiful
bright morning seemingly of early June. [Morris, News from
Nowhere, chapter 2]
We should both discover that the little towns below had
changed—but how, we should not have marked them well
enough to know. It would be indefinable, a change in the
quality of their grouping, a change in the quality of their
remote, small shapes. ... a mighty difference had come to
the world of men. [Wells, J Modem Utopia, chapter 1]
Morris’s abruptly beautiful trees can be taken (as they were
meant to be) for an emblem of this space and state: utopia is a
vivid witness to desperately needed alternative possibilities of
“the world of men," of human life. No wonder the debate has
waxed hot whether any particular alternative is viable and
whether it has already been found, especially in the various so­
cialist attempts at a radically different social system. In the heat
of the debate, detractors of this particular set of alternative
conclusions—often shell-shocked refugees from it—have tried to
deny the possibility and/or humanity of the utopian concept as
such. Other imprudent apologists—often intellectuals with a
solid position within the defended system—have taken the sym­
metrically inverse but equally counter utopian tack of proclaim­
ing that Civitas Dei has already been realized on Earth by their
particular sect or nation, in “God's own country” of North
America or the laicized Marxist (or pseudo-Marxist) experiments
from Lenin to Castro and Mao. Historians have transferred
these debates into the past: were Periclean Athens, Aqbar’s
India, Emperor Friedrich's Sicily, Münzer’s Mühlhausen, the
Inca state, or Jeffersonian U.S.A, utopian?
Such fascinating and tempting questions cannot fail to in­
fluence us in an underground fashion—defining our semantics
—-in any approach to a definition of utopia. But I propose to
confine myself here to a consideration of utopia as a literary
genre. No doubt this is not the first point about utopias-—that
would pertain to collective psychology: why and how do they
arise?—nor is it the last one—that would pertain to the politics
of the human species and perhaps even to its cosmology: how is
Homo sapiens to survive and humanize its segment of the universe?
DEFINING THE LITERARY GENRE OF UTOPIA 39

Such a politico-eschatological question has understandably arisen


out of twentieth-century heretic reinterpretations of the two
most systematic bodies of thought about man in our civilization:
the Judaeo-Christian one (in spite of its usual pat transfers of the
answer into the blue yonder of otherworldly post-mortems) and
the Marxist one (in spite of Marx’s and Engels’s scorn of subjec­
tive theorizing about ideal futures in their predecessors, the
“utopian socialists”), Ernst Bloch’s monumental philosophical
opus, culminating in Hope the Principle, has reinterpreted utopia
(as have some theologians such as Martin Buber and Paul Tillich)
as being any overstepping of the boundaries given to man, hence
a quality inherent in all creative thought and action. In a nar­
rower and more academic version, a similar reinterpretation of
“utopia” as any orientation that transcends reality and breaks the
bounds of existing order, as opposed to “ideology,” w'hich ex­
presses the existing order, was introduced by Karl Mannheim.2 3
But all these horizons, interesting and even inspiring as they are,
are beyond my scope here. I propose that an acknowledgment
that utopias are verbal artifacts before they are anything else, and
that the source of this concept is a literary genre and its
parameters, might be, if not the first and the last, nonetheless a
central point in today’s debate on utopias. If this is so, one cannot
properly explore die signification of utopia by considering its
body (texts) simply as a transparency transmitting a Platonic
idea: the signifiant must be understood as well as the signifié.
Thus, especially at this time of failing eschatologies, it might
even be in the interests of utopia (however widely redefined) if
we acted as physiologists asking about a species’ functions and
structure before we went on to behave as moralists prescribing
codes of existence to it: perhaps such codes ought to take into
account the makeup of the organism? And since discussions of
utopias are an excellent demonstration of the saying that people
who do not master history are condemned to relive it, the
physiological stance will have to be combined with an anamnesic
one, recalling the historical semantics (in sections 1 and 2) of

2. See Tillich (a representative essay from which is reprinted in Manuel, eci.). Buber.
Bloch, and Mannheim—-all in Bibliography II; also the rich anthology on the concept of
utopia; Neusüss, ed. (Bibliography II).
40 DEFINING THE LITERARY GENRE OF UTOPIA

utopia while trying to tease out its elements (in section 3) and
genological context (in sections 4 and 5).
1. Historical Semantics: Antediluvian
The first point and fundamental element of a literary defini­
tion of utopia is that any utopia is a verbal construction. This might
seem self-evident, but it is in fact just beginning to be more
widely recognized in the vanguard of “utopology." The Oxford
English Dictionary, for example, defines utopia in the following
ways:
1. An imaginary island, depicted by Sir Thomas More as
enjoying a perfect social, legal and political system,

b, transf. Any imaginary, indefinitely remote region, coun­


try, or locality.

2. A place, state, or condition ideally perfect in respect of


politics, laws, customs, and conditions.
..................................................................................... * » ■
b. An impossibly ideal scheme, esp. for social improve­
ment.

Obviously, the OED—whose latest examples come in this case


from the turn of the century—has not yet caught up to the
necessity and practice of defining utopia as a literary genre,3 If
we nonetheless look for clues in the above four definitions, we
shall see that the first one pertains to More’s “depiction” of a
locus which is, for the OED, defined by two aspects: (1) “imagi­
nary” removal from the author’s (and presumably the reader’s)
empirical environment; (2) sociopolitical perfection. The first as­
pect is then isolated in the semantic practice leading to definition
lb, and the second in the practice leading to 2, which is further
treated derisively by hardheaded pragmatists or ideologists of
the status quo in 2b. From all this a definition of utopia as a

3. See the stimulating discussion, with more lexicographic material, in Herb ruggen
(Bibliography II); also further French, German, and Spanish material in Rita Falke.
“Utopie—logische Konstrukcion und chimcrc,’' in Viilgradter and Krey, eds. (Bibliog­
raphy 11),
DEFINING THE LITERARY GENRE OF UTOPIA 41

literary form should retain the crucial element of an alternative


location radically different in respect of sociopolitical conditions from
the author’s historical environment. However, this element must
be valorized in the context of a literary-theoretical approach.
Only in OED 1 is there even a discreet mumble about the
utopia being an artistic artifact, hidden in the ambiguous “de­
picted” (about which more later). All the other definitions refer
to its qualities of perfection, remoteness, or impossibility. This
ontological equating of utopia to England, Germany, or any
other empirical country was an accepted nineteenth- and early
twentieth-century way of defining it. I shall adduce only a few
definitions from some better-known and more helpful works
pertaining to such a way of thinking, which might well—
regardless of their actual year—be called antediluvian:
(1) Utopias . . . are ideal pictures of other worlds, the exis­
tence or possibility of which cannot be scientifically demon­
strated, and in which we only believe. [Voigt, 1906]
(2) More depicted a perfect, and perhaps unrealizable, soci­
ety, located in some nowhere, purged of the shortcomings,
the wastes, and the confusion of our own time and living in
perfect adjustment, full of happiness and contentment.
[Hertzler, 1923]
(3) an ideal commonwealth whose inhabitants exist under
perfect conditions. [Encyclopedia Britannica, accepted by Ber-
neri, 1950]*
All of the above definitions or delimitations consider utopia sim­
ply as a Platonic idea and proceed to examine its believability and
realizability. Hertzler (2) is the most effusive and prolix among
them: the definition of utopias in general on which her whole
book is predicated, is effected by a definition of More’s work
prefaced with the statement that this definition isolates the dis­
tinctive characteristic applicable to all “imaginary ideal societies.”
The vagueness (“perhaps," “some nowhere”) and non-sequiturs

4. These definitions can be found in the following books (whenever in my quotes the
subject and predicate are missing, "utopia is" is implied): Voigt, p. 1; Hertzler, pp. 1-2;
Berneri, p. 320 (all in Bibliography II). A number of very useful approaches to utopia
are not referred to here, as they were not found cognate to a primarily literary-
theoretical viewing; a stilt greater number were found of little use except for a history of
"u topologic thought."
42 DEFINING THE LITERARY GENRE OF UTOPIA

(More depicted a society purged of “the confusion of our own


time”) make Hertzler a very good example—though greater of­
fenders could be found in the antediluvian age—of the useless­
ness to our endeavors of most surveys of “Utopian Thought” as
being idealistic and ideological. All the above definitions,
moreover, do not (except by vague suggestions inherent in
“commonwealth” or “society") distinguish between various reli­
gious "ideal pictures of other worlds” and utopias. This echoes
the (once?) widely-held unexamined premise that utopias are
really lay variants of paradise. Now if this is true, it is so only in
the sense which would make a counter project out of a variant.
Whereas it remains very important to pursue the historical un­
derground continuation of absolutistic religious and mythologi­
cal structures (especially those drawn from the Islands of the
Blessed and Terrestrial Paradises) in Plato, More, or a number
of other utopian writers, it should seem clear that there is little
point in discussing utopias as a separate entity, if their basic
humanistic, th is-worldly, historically alternative aspect is not
stressed and adopted as one of their differentiae genericae. "A
wishful construct has been explicated, a rational one, that does
not possess chiliastic certainties of hope any more, but postulates
the possibility of being constructed by its own forces, without
transcendental support or intervention,” observes Bloch even
about More's Utopia? What is literally even more important, such
a construct is located in this world. Utopia is an Other World imma­
nent in the world of human endeavor, dominion, and hypothetic
possibility—and not transcendental in a religious sense. It is a
nonexistent country on the map of this globe, a “this-worldly
other world.” No doubt, there is the pragmatic, Macaulayan
sense of utopia being anything intangible and impossibly far-off,
as opposed to immovable properly in one’s own property-owning
environment (“An acre in Middlesex is better than a principality
in Utopia”)56; this sense would also englobe all Heavenly and
Earthly Paradises. But from any point of view except that of a
property-owner and pragmatist, religion is, as Ruyer notes,
counter utopian. It is directed either towards Heaven (tran­

5. Bloch, p. 607.
6. Quoted in the 0ED\ see Thomas Babington Macaulay, "Lord Bacon," in his Critical,
Historical and Miscellaneous Essays and Poems (Albany, 1887), 2:229.
DEFINING THE LITERARY GENRE OF UTOPIA 43

scendence) or towards Middlesex (bounded empirical environ­


ment): in either case it is incompatible with a non-transcendenlal
overstepping of empirical boundaries.7 The teles of religion is,
finally, eternity or timelessness, not history. On the contrary, just
as the satire is an impossible possible—what is empirically possi­
ble is felt as axiologically impossible; it should not be possible—
utopia is a possible impossible. Subversion and rhetoric embrace
in a paradoxical socio-political revaluation of the Petrarchan ‘‘icy
fire" impossibilia—a “positive adynaton" in Barthes’s term.8
Thus, chemin faisant, we have found that the (still not too pre­
cise) element of historical alternative enters any definition which
would leave utopia intact as a literary genre and object of explo­
ration. We have still to pursue the metaphors adopted as a first
try at untying the embarrassing knot of utopia’s being a concept
and belief and yet, at the same time, obviously a (literary)
artifact—a “picture" (2 and 4) or a “description” (4 and 5):
(4) A. Nom donné par Thomas Morus au pays imaginaire
qu’il décrit dans son ouvrage: De optima reipublicae statu, deque
nova insula Utopia (1516), et dans lequel il place un peuple
parfaitement sage, puissant et heureux, grâce aux institutions
idéales dont il jouit.
B. Se dit par extension de tous les tableaux représentant,
sous la forme d'une description concrète et détaillée (et sou­
vent même comme un roman), l’organisation idéale d’une
société humaine. [Lalande, ed. of 1968, but text goes back at
least to 1928]
[A. Name given by Thomas More to the imaginary coun­
try which he describes in his work De optima reipublicae statu,
deque nova insula Utopia (1516), and into which he collocates
a people that is perfectly tvise, powerful, and happy, thanks
to the ideal institutions with which it is provided.
B. Said by extension of all pictures representing, by
means of a detailed and concrete description (often even as
a novel), the ideal organization of a human society.]
(5) la description d’un monde imaginaire, en dehors ... de
l’espace et du temps historiques et géographiques. C’est la

7. Ruyer (Bibliography 11), p. 31; see also Schwonke (Bibliography II), pp. 1-3, in
whose book this is a basic theme, and Gerber (Bibliography I), pp. 6-7.
8, Barthes (Bibliography II). p. 122.
44 DEFINING THE LITERARY GENRE OF UTOPIA

description d’un inonde constitué sur des principes différ­


ents de ceux qui sont à l’oeuvre dans le monde réel.9 [Ruyer,
1950]
[the description of an imaginary world, outside ... of
historical and geographic space and time. This is a descrip­
tion of a world based on principles that differ from those
underlying the real world ]
“Description” is derived etymologically from “writing,” but in an
archaic and ambiguous sense which, as it were, echoes the deri­
vation of writing from drawing. Above it is clearly employed
within the semantics pertaining to painting: “il décrit ... il place”
(in 4a. placing pertains to the way a landscape painter would
arrange his figures); and "tableaux représentant, sous la forme
d’une description” is a classic witness for my thesis (4b.). Even (5),
which is more abstract than the previous definitions, continues its
discussion in the immediately following line by contrasting such
descriptions to those of a non utopian novelist, who “lui, place des
personnages et des aventures imaginaires dans notre monde.”10
Utopia, as well as “our world,” is a scene for dramatis personae and
actions; the metaphor of author as puppeteer (stage manager),
never far beneath the metaphor of author as painter (scenog-
rapher), has here come nearer to the surface.
Such a dramatic metaphor, linked as it is to the “all the world's
a stage” topos, is potentially much more fruitful—since drama
fuses painting and literature, temporal and spatial arts—and
very appropriate for this dialogic form. Unfortunately, it has
not, to my knowledge, been taken seriously in defining utopias.
Thus such attempts at acknowledging the artificial character of
utopia have remained half-hearted. They have failed because
they did not acknowledge that it is a literary artifact. This is
crucial because the problems of “depicting” a radically different
(5) because perfect (4) imaginary world are in a literary artifact
quite distinct from the problems of a “tableau,” which exists in
an arrested moment of time and in a synoptic space. A picture

9. These definitions can be found in Lalande (Bibliography 11), p. 1179—and see the
whole discussion on pp. 1178-81—and Ruyer, p. 3. See also the definition of Dupont
(Bibliography HI C), p. 14. which is transitional between the first group of definitions and
this one. All the translations in this book, unless otherwise indicated, are mine.
10. Ruyer, p. 3; italics added.
DEFINING THE LITERARY GENRE OF UTOPIA 45

may perhaps approximate the status of a mirror of external real­


ity (though even rhe mirror reverses). In literature, a concrete
and detailed “description” or, better, verbal construction is not, in
any precise sense, a “re-presentation” of a preexisting idea
which would be the content of that representation or description
(where would such an idea preexist? with the Zeitgeist?). Literary
texts cannot be divided into body and soul, only into interlocking
levels of a multifunctional body, which is a human construct out
of verbal denotations and connotations. Only within such a con­
text can the definition of its thematic field—practically identical
from (2) to (5)—become a valid part of a literary definition. The
imaginary community (the term seems preferable to the ambiguous
"world”) in which human relations are organized more perfectly than in
the authors community can be accepted as a first approximation to
identifying lhe thematic nucleus of the utopian genre.
One further point should account for my substitution of
“more perfectly” in place of the "perfect” in (2) to (4). Though
historically most of the older utopias tried to imagine a certain
perfection, after Bacon's New Atlantis and Fénelon’s Télémaque
(not to forget Plato’s Laws) a suspicion ought to have arisen that
this is not inherent in lhe genre. That suspicion should have
grown into a certainty after Saint-Simon and Morris. By the time
Wells wrote his celebrated first page of A Modem Utopia distin­
guishing between static and kinetic utopias, the laggard academic
and literary critics of the genre found their work done for them.
Since then we have had no further excuse for insisting on abso­
lute perfection, but only on a state radically better or based on a
more perfect principle than that prevailing in the author's com­
munity, as a hallmark of the utopian genre.11 As for the “au-

11. See the analogous argument in Walsh (for the titles in this note see Bibliography
II), p, 25. The position of utopia midway between the corruptible world of class history
and ideal perfection is quite analogous—as will be discussed in section 4 of this chapter—
to the position of Earthly Paradise in religious thought; see for example the definition of
Athanasius of Alexandria:
The Terrestrial Paradise we expound as not subject to corruption in the way in
which our plants and our fruits get corrupted by putrefaction and worms. Nor is it,
on the other hand, wholly incorruptible, so that it would not in future centuries
decay by growing old. But if it is compared with our fruits and our gardens, it is
superior to all corruption; while if it is compared to the glory of the coming Good,
which eye hath not seen nor ear heard nor the heart of man comprehended, it is
and is reputed to be vasdy inferior.
46 DEFINING THE LITERARY GENRE OF UTOPIA

thor’s community," this phrase can be left conveniently plastic to


embrace whatever the author and his ideal readers would have
felt to be their community—from city to nation to planet.
2. Historical Semantics: Postdiluvian
In the last twenty years, at least in literary criticism and theory,
the premise has become acceptable that utopia is first of all a
literary genre or fiction. The Cold War “end of ideology” climate
might have contributed to this (it can be felt, for example, in the -
disclaimers in the Negley*Patrick book discussed below), but
more importantly, it has been part of a deeper epistemological
shift in literary shcolarship—a belated recognition that, as Frye
wrote, the literary critic “should be able to construct and dwell in
a conceptual universe of his own.”*12 1 shall again adduce only a
few definitions as characteristic examples for works of this
period, after the deluge of two world wars and two cycles of
worldwide revolutions:
(6) There are three characteristics which distinguish the
utopia from other forms of literature or speculation:
1. It is fictional.
2. It describes a particular state or community.
3. Its theme is the political structure of that fictional state
or community. . . .
Utopias are expressions of political philosophy and theory,
to be sure, but they are descriptions of fictional states in
which the philosophy and theory are already implemented
in the institutions and procedures of the social structure.
[Negley and Patrick, 1952]
(7) ... the literary ideal image of an imaginary social system
(Siaatsordnung). [Herbruggen, 1960]
(8) the utopian novel is the literary manifestation of a play­
ful synopsis of man, society, and history in a variable, image­
like (bildhaft) thought model possessing spatio-temporal au­
tonomy, which model permits the exploration of possibilities

Athanasii arc hie p. Alexandrini, Opera omnia quae extant . . . (Paris, 1698) 2:279, quoted
in Coli, p. 39- The insistence on utopia as wholly "ideal" can still be found in Herbrug­
gen—see note 13.
12. Northrop Frye, Anatomy of Criiieiem, p. 12.
DEFINING THE LITERARY GENRE OF UTOPIA 47

detached from social reality yet relating to it. [Krys man ski,
1963]
(9) la description littéraire individualisée d'une société im­
aginaire, organisée sur des bases qui impliquent une critique
sous-jacente de la société réelle.13 [Cioranescu, 1972]
[the individualized literary description of an imaginary so­
ciety, organized on bases which imply an underlying critique
of the real society.]
Negley and Patrick (6) seem to have been the first expressly to
enunciate a differentiation between the utopia of political scien­
tists and Geisteswissenschaftler (“expressions of political philosophy
and theory”) and that of the literary critics and theorists (“fic­
tional states,” theme and ideas “implemented”). Their pioneer­
ing status is evident in certain uneasy compromise with the older
conception which they are abandoning.14 But as well as their use
of the by-now dead metaphor of describing (which in a proper
context it would perhaps be pedantic to fault), their failure to
elaborate what exactly fictional implementation entails and their
de facto concentration in the book on socio political ideas and
structure unrelated to the literary structure leave their definition
somewhat isolated and without consequences. But their useful
and influential book at least indicated the horizons of studying
what they called in their preface, in a mixture of conceptual styles,
both “utopian thought in Western civilization” (old style) and
also, somewhat shamefacedly, “the literary genre of the utopists”
(newf style).
On the other hand, Herbrüggen (7) starts boldly and happily
by identifying utopia as literary, but then leaves it dangling in
intense vagueness by calling it not only “imaginary” but also the
“ideal image.” Later in this work, he has many just and stimulat­
ing things to say about its delimitation from other genres. In
particular, he has been a pioneer in drawing some structural
consequences from defining utopia as possessing a literary mode
of existence. However, a number of his parameters, including his

13. These definitions can be found in the books by Negley and Patrick, pp. 3-4;
Herbrüggen, p. 7; Krysmanski. p. 19; and Cioranescu. p. 22—all in Bibliography It.
14. No doubt, there were earlier implicit or incidental suggestions that fictional utopia
was primarily a literary genre, e.g. in Dupont—in spite of his definition and title—and in
Frye. Anatomy. But the voices of these, and possibly of other, precursors fell on deaf ears.
48 DEFINING THE LITERARY GENRE OF UTOPIA

definition, seem to fit More (his particular paradigm), or indeed


a utopian program, better than they would an ideal-typical
utopia.
Krysmanski's (8) sociological exploration of German “utopian
novels” of the twentieth century (which ought rather to be called
science fiction, as 1 shall argue in section 5) set itself the laud­
able aim of discovering and fully defining “the specific nature of
the utopian novel": his definition is the conclusion of a chapter
with that title. Unfortunately, for an analysis of a “literary man­
ifestation” (Erscheinungsform) it is far too little conversant with
fundaments of literary theory and criticism. One’s sympathy and
tolerance lie with his Aristotelian basic approach, striving for a
definition which must be precise and comprehensive, in which
case technical jargon is almost impossible to avoid. Nonetheless,
it is not onlv the Teutonic and Mannheimian “sociology of
knowledge" nature of the jargon which makes one pause, it is
primarily the arbitrariness and vagueness of the elements of the
definition, which seem to prove that modern definitions can be
every bit as prolix-cum-insufficient as the antediluvian ones. It
may be useful to draw our attention to the elements of playful­
ness, of simultaneous viewing or synopsis (Zusammenschau) of
man, society, and history, or of an exploration of possibilities.
But why “manifestation of a synopsis” (the German is still worse:
“Erscheinungsform der . . . Zusammenschau”)? Why “variable,"
“image-like,” and “spatio-temporal autonomy”—is not every Denk-
modell such? And the final clause evidently pertains to science
fiction in general, being too wide for utopia, which is bound up
with the (here missing) “more perfect community” concept.
As for Cioranescu’s book devoted to “utopia and literature," a
work full of stimulating and provocative statements, I shall re­
turn to later. At this point, it might suffice to point out with
relief how neat and with unease how over generalized his defini­
tion is (9). Are not Paradise, an Island of the Blessed, or satirical
SF covered by it as well? And, not to boggle at minor maters, just
what is “the real society”?
3. A Proposed Definition:
Utopia as Verbal Construction
The historico-semantical discussion of the preceding two sec­
tions has come up with the following elements for defining
DEFINING THE LITERARY GENRE OF UTOPIA 49

utopia: a radically different and historically alternative sociopolit­


ical condition; an alternative locus; an imaginary community in
which relations are organized more perfectly than in the author’s
community; the fictional or, more clearly, “verbal construction”
character of any such condition, location, or community; the
particular or individualized character of any such construct as
opposed to general and abstract utopian projects and programs.
I shall now commit the utopian imprudence of proposing after
the above critique a construct or definition of my own:
Utopia is the verbal construction of a particular quasi­
human community where sociopolitical institutions, norms,
and individual relationships are organized according to a
more perfect principle than in the author's community, this
construction being based on estrangement arising out of an
alternative historical hypothesis.
I have indicated earlier in general outline the importance to be
alloted to the element of verbal construction. This can be fully
demonstrated only in particular analyses of utopian works. But
its relevance can be seen even in a general answer to the ques­
tion: what type of verbal construction? As Frye has pointed out,
utopia belongs to a narrative form and tradition which he calls
anatomy (or Menippean satire) rather than to the novel. The
anatomy deals less with illusionistic “people as such than with
mental altitudes” and at its most concentrated “presents us with a
vision of the world in terms of a single intellectual pattern.”1516
Our critical judgments should take this into account; in particu­
lar, there is no point in expecting from a characterization and
plotting which are more allegorical than naturalistic the qualities
and criteria induced from the psychological novel, from Prevost
to Proust or Richardson to Henry James.10 To take one example,
the conclusions of Gerber’s interesting book on twentieth-cen­
tury utopias (or rather SF) are vitiated by his assumption and

15. Frye, pp. 309 and 310.


16. The famous quarrel between James and Wells—available in Leon Edel and Gor­
don N. Ray, eds., Henry Janus and H. G. Wells (Urbana, IL, 1958which resulted in a
draw rather than in the vindication of the psychological novel the Jamesians saw in it, is a
clear example of the collision between the "anatomic" or allegorical and the "novelisttc"
or individualistic orientations.
50 DEFINING THE LITERARY GENRE OF UTOPIA

definition of utopia as a novel.17 To take another, Elliott has


aptly complained about one of the dominant interpretations of
More’s Utopia:
We are given no sense . . . that these questions exist, not as
abstract political, religious, or philosophical propositions,
but as constitutive elements in a work of art. What is wanted
instead of the Catholic interpretation of communism is an
interpretation of Utopia that will show us how the question
of communism is incorporated into the total structure of the
work.18
Further, some basic structural characteristics of utopia seem to
flow logically from its status as a discourse about a particular,
historically alternative, and better community. Since such a dis­
course will necessarily present an opposition which is a formal
analogy to the author’s delimited environment and its way of life,
any utopia must be (1) a rounded, isolated locus (valley, island,
planet—later, temporal epoch). Since it has to show more per­
fectly organized relationships, the categories under which the
author and his age subsume these relationships (government,
economics, religion, warfare, etc.) must be in some way or other
(2) articulated in a panoramic sweep whose sum is the inner or­
ganization of the isolated locus; as Barthes remarks about
Fourier (and some other writers), the syntax or composition of
elements is identified with creation in such works.19 Since not
only’ the elements but also their articulation and coordination
have to be based on more perfect principles than the categoriza­
tion in the author’s civilization (for example, the federalist
pyramid from bottom up of More’s Utopia as opposed to the
centralist pyramid from top down of More’s England and
Europe), (3) a formal hierarchic system becomes the supreme order
and thus the supreme value in utopia; there are authoritarian
and libertarian, class and classless utopias, but no unorganized
ones. (Morris’s reticence about organization and hierarchy in
News From Nowhere places that work halfway between utopia

17. Gerber, final two chapters, and in particular pp. 121-22. See the critique by Elliott
(Bibliography II), p. 104 and the whole chapter "Aesthetics of Utopia."
18. Elliott, pp. 28-29.
19. Barthes, p. 9; this whole discussion is indebted to Barthes's book, though I do not
wholly share his horizons.
DEFINING THE LITERARY GENRE OF UTOPIA 51

and Earthly Paradise; see chapter 8). Usually the installation of


the new order must be explained—a contract theory, as Frye
observes, is implied in each utopia (King Utopus, the socialist
revolution, gas from a comet, etc., being the arbiters or contract­
makers). The utopian contract is necessarily opposed to the domi­
nant contract-myth in the author’s society as the more reverent
“contract behind the contract,”20 a human potential which exist­
ing society has alienated and failed to realize. Lastly, utopia is
bound to have (4) an implicit or explicit dramatic strategy in its
panoramic review conflicting with the “normal” expectations of
the reader. Though formally closed, significant utopia is themat­
ically open; its pointings reflect back upon the reader’s "topia.” I
have already hinted at that in section 1, and one critic has even
conveniently found a three-act dramatic structure in More's
Utopia.21 Whether this is exact or not, there is no doubt that an
analysis of ideational protagonists and settings in Burkean
“dramatistic” terms is here appropriate.22 For example, utopia is
invariably a frame-within-a-frame, because it is a specific wond­
rous stage, set within the world stage; techniques of analyzing the
play-within-the-play could be profitably employed when dealing
with it. The varieties of the outer frame—usually some variant of
the imaginary voyage23—have been readily noticeable and as
such the object of critical attention; less so their correlation of
say, the humanistic symposium of More or the socialist dream-
which-might-be-a-vision of Morris with the experience in the
inner frame. Even on the stylistic and not only compositional
level, such a strategy should be fruitful: “l’écriture," remarks
Barthes of Fourier, “doit mobiliser en même temps une image et
son contraire [the writing must mobilize at the same time an
image and its opposite].”24
Finally, “verbal construction” as a definitional element by-

20. Northrop Frye, "Varieties of Literary Utopias." in Manuel. ed.k p. 38.


21. Edward Surtz, S.J., "Utopia as a Work of Literary Art,” in Edward Surlz, S.J., and
J.H. Hexter. eds.. The Complété Works of St. Thomas More (New Haven. 1965), 4: exxv-
cliii, especially in the chapter "Dramatic Technique, Characterization, and Setting."
22. E.g. Kenneth Burke, The Philosophy of Literary Form (New York, 1957).
23. Historically this is especially significant in antiquity and Renaissance, when most
utopias and imaginary voyages were combined, but it does not have to persist as an
explicit combination. See the excellent survey of Gove (Bibliography 111 A), much in
need of newer follow-ups.
24. Barthes, p. 115,
52 DEFIXINC THE LITERARY GENRE OF UTOPIA

passes, I hope, the old theologizing quarrel whether a utopia can


be realized, whether in fact (according to one school) only that
which is realizable or on the contrary (according to another but
equally dogmatic school) only that which is unrealizable can be
called utopia. Neither prophecy nor escapism, utopia is, as many
critics have remarked, an “as if,”2526 an imaginative experiment or
“a methodical organ for the New.”28 Literary utopia—and every
description of utopia is literary—is a heuristic device for perfec­
tibility, an epistemological and not an ontological entity.
“L’utopie est un jeu, mais un jeu sérieux. L’utopiste a le sens des
possibilités autres de la nature, mais il ne s’éloigne pas de la
notion de la nature [Utopia is a game, but a serious game. The
utopian author envisages the other possibilities of nature, but he
does not let go of the notion of nature]” argued Ruyer in two
chapters which remain among the best written on the “utopian
mode.”27 He referred to utopian subject matter as “les possibles
latéraux [the lateral possibilities]” and compared the utopian ap­
proach or view to the hypothetico-deductive method in experi­
mental sciences and mathematics (for example, non-Euclidean
geometries). If utopia is. then, philosophically, a method rather
than a state, it cannot lie realized or not realized—it can only be
applied. That application is, however, as important as it has been
claimed that the realization of utopia is: without it man is truly
alienated or one-dimensional. But to apply a literary text means
first of all (wherever it may later lead) to read as a dramatic dia­
logue with the reader.28 Besides requiring the willingness of the

25. See Hans Vaihinger, Die Philosophic des ,4ir Ob (Leipzig, I920) or The Philosophy of
"As If" trans. C. K. Ogden (New York. 1924). The verbal mode appropriate to this is the
subjunctive: see Elliott, p. 115: Samuel R. Delany. "About Five Thousand One Hundred
and Seventy Five Words," in Clareson, ed.. S77 (Bibliography I); Michael Holquist, "How
to Play Utopia." in Jacques Ehrmann, ed.. Game, Play, literature (Boston, 1971). particu­
larly illuminating in his discussion of utopias as a literature of the subjunctive in
"hypothetical or heuristic time." p. 112; and Chude-Gilbert Dubois, “line architecture
fixionelle," Revue des sciences humaines 39, No. 155 (1974): 449-71.
26. Bloch, p. 180.
27. Ruyer, chapters 1 and 2; the first quotation is from p. 4 and the later one p. 9;
Ruyer acknowledges the stimulus of an observation by l-aiande. p. 1180. Unfortunately,
the analysis of actual utopian characteristics and works in the rest of Ruyer’s book is
much less felicitous.
28. Some of my conclusions are very similar to those of Harry Berger, Jr., in his more
synoptic, seminal introductory discussion of the “other world" in “The Renaissance
World: Second World and Green World," The Centennial Review 9 (1965): 36-78. Regret­
DEFINING THE LITERARY GENRE OF UTOPIA 53

reader to enter into dialogue, the application of utopia depends


on the closeness and precision of his reading.
4. Comment: Utopia as Historical Estrangement
I have thus far worked upon certain premises, among them
that scholarly inquiry is possible only when oriented towards,
and by, an at least approximately delimited and defined field
and that valid definitions in literary studies—as in anything—are
historical and not transcendental, or “contextualist” and not “es-
sentialist.” Proceeding further, it is necessary to add that the
basic diachronic way to define the context of a work of art is to
insert it into the tradition and system of its genre (meaning by
that a socioaesthetic entity with a specific inner life, yet in a con­
stant osmosis with other literary genres, science, philosophy,
everyday socioeconomic life, and so on). Understanding par­
ticular utopias really presupposes a definition and delimita­
tion of their literary genre (or, as we shall see, subgenre), its
inner processes, logic, and telos. What is, then, the distinctive set
of traits of the literary genre “utopia,” its differentia generica?
1 have argued in my first two chapters for a division of prose
literature into naturalistic and estranged genres. The literary
mainstream of the individualistic age endeavors faithfully to re­
produce empirical textures, surfaces, and relationships vouched
for by human senses and common sense. Utopia, on the con­
trary, endeavors to illuminate men’s relationships to other men
and to their surroundings by the basic device of a radically dif­
ferent location for the postulated novel human relations of its
fable; and I have proposed to call literary genres which have
such a different formal framework “estranged.” One should in­
sist on the crucial concept of a radically different location, of an
alternative formal framework functioning by explicit or implicit ref­
erence to the author’s empirical environment. Without this ref­
erence, nonutopian readers, having no yardstick for comparison,
could not understand the alternative novelty. Conversely, with­
out such a return and feedback into the reader’s normality there
would be no function for utopias or other estranged genres: “the

fully I must add that I believe his particular argument about Utopia—that More differs
radically from Hythloday—to be wholly unconvincing.
54 DEFINING THE LITERARY GENRE OF UTOPIA

real function of estrangement is—and must be—the provision of


a shocking and distancing mirror above the all too familiar real­
ity.”29 No-place is defined by both not being and yet being like
Place, by being the opposite and more perfect version of Place. It
is a “positive negation,” a “merveilleux réel,”30 the standing on its
head of an already topsy-turvy or alienated world, which thus
becomes dealienated or truly normal when measured not by
ephemeral historical norms of a particular civilization but by
“species-specific" human norms. Utopia is thus always pre­
dicated on a certain theory of human nature. It takes up and
refunctions the ancient topos of mundus inversus: utopia is a for­
mal inversion of significant and salient aspects of the author’s
world which has as its purpose or lelos the recognition that the
author (and reader) truly live in an axiologically inverted world.
It follows, as has been increasingly recognized in modern inves­
tigations (and as has been mentioned in passing in section 1),
that the explicit utopian construction is the logical obverse of any
satire.31 Utopia explicates what satire implicates, and vice versa.
Furthermore, there are strong indications that the two are in fact
phylogenetically connected in the folk-inversions and “saturas”
of the Saturnalias, whose theme was sexual, political, and
ideological reversal, in fact total existential “reversal of values, of
social roles, of social norms.”32 The best argument in favor of
that can be found in the ontogenesis of individual works, in—to
stick to utopias and cognate estranged genres—the most promi­
nent titles of the tradition which runs from Lucian’s True His­
tories and More's Utopia through Fourier, Bellamy, Morris, Wells,
and Zamyatin to modern SF. A guess could even be hazarded

29. Ernst Bloch, “Entfremdung, Verfremdung," Verfremdungen, 1 (Frankfurt, 1963),


English as "Entfremdung, Verfremdung-. Alienation, Estrangement,'’ trans. Anne Halley and
Darko Suvin, in Erika Munk, ed.,BrrrAt (New York, 1972). p. 10. For "estrangement," see
the discussion and references in my first chapter (Shklovsky and Brecht), as well as Bloch.
Das Pnnzip Hoffnung.
30. "Positive negation” is the term used in Mikhail Bakhtin’s fundamental Tvorclustuo
Fransua Rable . . . (Moscow, 1965), English as Rabelais and His World (Bibliography 11), p.
403; but see also this whole book for a rich and persuasive account of folk humor as the
source for inverting and negating a dominant, upper-class feeling of reality. "Merveilleux
réel” is an expression of Barthes's, p. 101.
31. Sec Frye, Anatomy, pp. 309-12; Lalande, p. 1180; Negley and Patrick, pp. 5-6;
and especially Elliott, chapter 1. "Saturnalia, Satire, and Utopia."
32. Elliott, p. 11.
DEFINING THE LITERARY GENRE OF UTOPIA 55

that the significance and scope of writings in this tradition can be


gauged by the degree of integration between its constructive-
utopian and satiric aspects: the deadly earnest blueprint and the
totally closed horizons of “new maps of hell” both lack aesthetic
wisdom.
However, besides satire (which can be, like utopia, both a
mode and a genre) the estranged literary genres comprise sev­
eral which are differentiated from utopia by not situating what
Aristophanes calls their topos apragmon in the field of an alterna­
tive historical hypothesis. The most relevant ones are, in ascend­
ing order, myth, fantasy, folktale, Cockayne, and Terrestrial
Paradise.
I have tried to deal w ith myth in my earlier chapters, and I can
only repeat that, although it is also shaped as a specific form of
estrangement, myth is diametrically opposed to a historical ap­
proach.33 Conceiving human relationships to be fixed and
supernaturally determined, myth claims to explain phenomena
by identifying their eternal essence; conceiving human relation­
ships to be changeable and maternally determined, history at­
tempts to explain phenomena by identifying their problematic
context. From a historical point of view, myth itself is a historical
phenomenon and problem, an illusion when not a fraud. Litera­
ture is, in fact, never truly a myth (though mythological tales are
literature) but only, in certain cases, formally analogous to
mythical structure or niythomorphic. Thus, for example, the
myth of the Golden Age can have many formal analogies and
elements in common with utopia, but utopia is its opposite:
. . . man's effort co work out imaginatively what happens—or
what might happen—when the primal longings embodied in
the myth confront the principle of reality. In this effort man
no longer merely dreams of a divine state in some remote
time; he assumes the role of creator himself.
A characteristic of the Golden Age ... is that it exists
outside history, usually before history beings: in illo tem­
pore. 34

33. See also Ruyer, pp. 4-6. For all my admiration of Professor Frye's insights, here J
obviously disagree with the horizon and main terminology of his work—and in particular
with his dassitying Dante's Paradiso and Purgatario as utopian, in Manuel, ed.. p. 34.
34, Elliott, pp. 8-9.
56 DEFINING THE LITERARY GENRE OF L'TOPIA

Folktale and fantasy, being morphological and ideological descen­


dants of fragmented mythology' (in the case of fantasy privatized
to boot), can be regarded in a similar way. Neither of them pre­
tends to be historically oriented or in historical time. Both take
place in a context of supernatural laws oriented towards the
protagonist, whereas for humanistic historiosophy—including
utopia—nature is neutral and man’s destiny is man.
Somewhat closer to utopia is Cockayne (Cuccagna, Schlaraffen-
land), a widespread folk legend of a land of peace, plenty, and
repose, probably refurbished by the student-poets of goliardic
and “prandial” libertinism.35 This legend is interesting here be­
cause the land where roasted fowls fly into your mouth, rivers
flow with cream or wine, and sausages with a fork stuck into them
run around crying “eat me, eat me!” is obviously an inverted
image of the hunger, toil, and violence in the authors’ everyday
lives. Cockayne is already an inverted parallel world that relates,
if not yet to a historical hypothetical possibility organized into in­
stitutions, then at least to everyday human needs and not to
transcendental doctrines:
Lt fiction parallèle, la préoccupation pour le destin de
l’homme et la solution strictement matérialiste sont les trois
traits fondamentaux qu’ont en commun l’utopie et la pays
de Cocagne. . . .
Le matérialisme ainsi entendu ignore les restrictions men­
tales et transcende la matière pour la transformer en divinité
tutélaire et en providence.36
[The parallel fiction, the preoccupation with human de­
stiny and the strictly materialist solution are the three fun­
damental traits which utopia and Cockayne have in com­
mon. . . .
Taken thus, materialism ignores mental restrictions and
transcends matter in order to transform it into patron deity
and providence.]

35. See Bakhtin's chapter “Banquet Imagery," especially pp. 296-98, and Morton
(Bibliography It), pp. 15-27. For some further references to Cockayne see Ackermann,
Bonner (both in Bibliography lit B), Boas, pp. 167-68, Patch, pp. 51 and 170-71 (both
in Bibliography It), Gatz. pp. 116-21. Grauss, Manuel and Manuel (all in Bibliog­
raphy III B), and note 36.
36. Cioranescu, pp. 57 and 59. but see his whole passage on pp, 55-62, which pre­
DEFINING THE LITERARY GENRE OF UTOPIA 57

Clearly, as Cioranescu notes, this does not jibe with the funda­
mental utopian context of a neutral nature: but utopia wishes to
achieve by cognitive means and in a context of hypothetically
inflected history what the legend of Cockayne achieved in a pure
wishdream outside the terrible field of history. While still a
folktale, Cockayne can be readily transferred to the vicinity of
utopia by allying its dream to a cognitive context, as in Rabelais,
The Earthly Paradise may be even nearer to utopia. Outside
official Christianity, it is as a rule not transhistoric, but can be
reached by an ordinary voyage. It is divided from other lands by
a barrier, which makes it usually an island in the sea—an Island
of the Blessed, as the Greek tradition from time immemorial has
it and as many other writings, anonymous or famous, also know
it, to wit, the Celtic blessed island or Dante’s Paradiso Terrestre
in the western sea.37 Often, especially in versions unaffected by
religious rewriting, the inhabitants are not disembodied, but are
simply more perfect people. The implied critique of the author’s
environment is explicated in a whole group of “other world’’
tales.38 The magical or folktale element is clearly present in the
perfect climate, the freedom from cares and strife, and often in
the arrested time on such blessed islands (so that a return from
them entails instant aging or turning to dust). And yet, the prox­
imity of utopia of Terrestrial Paradise in its unbowdlerized ver­
sions is impressively indicated by a tale such as that of the
Guarani Land-Without-Evil. That land, also called the House of
Our Ancestress,
is difficult to reach, but it is located in this world. Although
. . . it entails paradisiacal dimensions (for instance,

sents l he best analysis of Cockayne 1 know of. For connections with satire see also Elliott,
pp. 16-17.
37. A general survey on ideas about the Golden Age. Eden, and Paradise is to be
found in Manuel and Manuel, who, however, fail to make the crucial distinction between
heavenly and earthly paradise. On Greek tales see Bonner, Lovejoy and Boas, the com­
ment in Bloch, chap. 36 (all in Bibliography it), and a number of works from Bibliog-
raphy III B, especially Gau. Finlev. Pöhlmann, Rohde, and Winston, For medieval tales
and beliefs about localized "other worlds” See Boas. Goli, Graf, “Il Mito del Paradiso
Terrestre." Patch (all in Bibliography II}, and a number of works from Bibliography
III B, especially Curtius, Graus, Kampers. Peters, and Westropp; Coli, p. 13Ü, and Patch,
p. 135, comment on the accessibility and material reality of Eden for medieval minds. See
also Giamatti (Bibliography II) for Renaissance echoes.
38. See Patch, p. 128, and Coli, p. 130,
58 DEFINING THE LITERARY GENRE OF UTOPIA

immortality)—the Land-Without-Evil does not belong to the


Beyond. . .. One arrives there . . . [not only] in soul or spirit,
but in flesh and bones. . . . [It] is thus a world at once real
and transfigured, where life continues according to the same
familiar model, but . . . without misery or sickness, without
sins or injustice, and without age.39
Is such a country outside history, as Eliade thinks? It is certainly
outside empirical or known history, but it is at the same time an
alternative, hypothetically possible, and supremely desirable his­
tory on Earth. All the above qualifications could be applied to
utopia, not only in my proposed definition but according to most
of the quoted definitions too. It lacks only More’s great discovery
of focusing on sociopolitical institutions and norms as a key to
eliminating misery, sickness, and injustice. The usual utopian
answer, communal ownership, is here preserved (the Guaranis
did not need to attain it) by means of what Bloch calls a “medical
utopia” (search for immortality, eternal health, and youth).
If not utopia, this is a fraternal genre: an early and primitive
branch of SF.
Comment: Utopia as a More Perfect
5.
Organized Community
Finally, the relationships of utopia to other genres of what I
have in the earlier chapters called “cognitive estrangement”—
SF, pastoral, and nonfictional works—should also be discussed.
This will account for the necessity of all my definitional ele­
ments between “verbal construction” and the final clause. Just
like Cockayne, the pastoral is akin at least to libertarian utopia in
its rejection of money economy, cleavage between town and
country, and state apparatus. But just like Cockayne, it is primar­
ily a unomia, a land without formalized institutions, without or­
ganized superstructures of community life.40 If Cockayne is the
land for sensualists, Earthly Paradise for heroes, and pastoral for

39. Mircea Eliade, “Paradise and Utopia,” in Manuel, ed., pp. 273-75. For paradises
located on Earth see also Boas, pp. 154-74, Graf. pp. 15 and 24, and Goli, p. 91; and for
the arrival in flesh at Earthly Paradise the Hellenic testimonies in Lovejoy and Boas, pp.
25-30 and 290-303, where further bibliography can also be found.
40. Cioranescu, pp. 60-61.
DEFINING THE LITERARY GENRE OF L'TOPIA 59

swains (sheperds as philosophers, poets, and lovers), utopia is the


land for naturalistic human figures just slightly larger (more
virtuous) than evervdav nature.
The definitional element of a particular community is neces­
sary, as observed in section 3, in order to differentiate utopia
from general beliefs, programs, and unlocalized projects. How­
ever, as soon as the blueprints and beliefs become localized and
approach a narrative (as in much of the writing of utopian
socialists), there is little delimitation provided by any definition of
utopia I can think of. The usual escape clause is that utopia is
belles lettres or fiction, while Saint-Simon or Fourier are lettres or
nonfiction. But that distinction, though sufficiently normative
in the eighteen th-cen tun- to allow Swift to base the formal
framework of Gulliver's Travels on playing with it, is historically a
fugitive one. What was the Guarani legend of Land-Without-Evil
or Columbus's letter on finding the Terrestrial Paradise beyond
the Orinoco for the authors, fiction or nonfiction? And for us?
What is, for that matter, the Bible—theology or '‘literature” in the
sense of fiction? The term “literature” has always wavered be­
tween a populist or sociological inclusive extreme (everything
published in printed form) and an elitist or aesthetical exclusive
extreme (only those “belles” works worthy of entering into a
normative history of “literature”). In brief, the eighteenth­
nineteenth century escape clause does not seem co me to work
any longer, since it deals in subjective values and intangible in­
tentions. Suppose it were found that the Supplement to Bougain­
ville's Toyage had been written by Bougainville instead of
Diderot—would it cease to be utopian? And if Fourier had pub­
lished his vision of anti-lions and a sea of lemonade with Jules
Verne’s editor, would it thereby become SF? We are beginning to
move in the Borgesian world, where the same text has opposite
meanings according to the intention of the author. This is good
satiric fun, but any literary theory which can be built upon such
premises would have to reject most that we now dignify with
such a name. The same dilemma applies to ethnological reports:
if literature is not defined by being right or wrong but by il­
luminating human relationships in certain ways and by certain
means, I see no way of delimiting Levi-Strauss’s sequence on myths
from fictional literature or belles lettres. Reports on the perfect
60 DEFINING THE LITERARY GENRE OF UTOPIA

Inca empire, it has been argued, had inspired More. This is


probably inexact, but such a report, especially if related at se­
cond hand, would have been generically indistinguishable from
the Utopia (although, among other things, surely less witty). If I
have argued all along in this chapter for utopia as literature, it is
precisely because of such a breakdown in the philosophy of liter­
ature. The resulting inchoate mass should at least be judged by
taking into account the whole text and not arbitrary essences
abstracted from it: as imaginative, though not imaginary.4'
The definitional element of quasi-human refers to such com­
munities as those of Swift’s Houyhnhnms, Stapledon’s Eight­
eenth Men {Homa sapiens being the First Men), or the numer­
ous aliens and cybernetic intelligences of modern SF.41 42 It con­
notes that utopias are in a strange and not yet clarified way an
allegorical genre akin to the parable and analogy. In the par­
able or analogy, the premises do not have to be realistic as long
as the upshot is clear. Thus, utopia is always aimed at human re­
lations, but its characters do not have to be human or even out­
wardly anthropomorphic. Their relationships and commun­
ities, though, will make sense only insofar as they can be judged
as similar or dissimilar to human ones.
The element of community differentiates utopias on the one
hand from "robinsonades,” stories of castaways outside of an
alternate community.43 On the other hand, this terminology tries
to steer a middle course in the debate which seems to have raged
in Mitteleuropa between State worshippers and Kantian or
anarchist individualists among critics, an echo of which is heard
in Krvsmanski’s Solomonic solution of a “synopsis” of man,
society, and history. The “anarchists” (for example, Berneri)
stressed the moral behavior of individuals, the “archists” the
normative power of institutions. Too narrow an interest in gov­
ernmental apparatus leads to the deadly boredom of eight­
eenth-century Staatsromane in the narrow sense—say, certain
works extolling constitutional monarchies in the South Seas. Too

41. Frys, “Varieties of Literary Utopias." p. 32.


42. Sec e.g. Robert Boguslaw's discussion of mtn as "operating units" in The New
Utopians (Englewood Cliffs, NJ, 1965), passim, which effects a witty juxtaposition of
utopias and “system design.”
43. See Bruggernau (Bibliography II), especially pp. 187-89.
DEFINING THE LITERARY GENRE OF UTOPIA 61

wide a sense of utopia, which with Bloch would embrace medical,


biological, technological, erotic, and even philosophical wish­
dreams, leads to incorporating Don Juan and Faust, the Theses
an Feuerbach and The Magic Flute, into utopia: a somewhat over­
weening imperalism. The middle course suggested in what is, I
hope, my prudent use of “community where sociopolitical insti­
tutions, norms, and individual relationships are organized ac­
cording to a more perfect principle” (see section 3), focuses on
the sociopolitical concern with justice and happiness, on the
“radical eudemonism” of utopia’s “detailed, serious discussion
of political and sociological matters.”44 And if utopia is not a
myth valid for all eternity but a historical genre, the acknowledge­
ment of its context in the adjunct “than in the author’s commu­
nity” seems mandatory—most utopias would not be such for
most of us today without that adjunct, since one man’s perfec­
tion is another man’s (or class’s) terror.

Yet, finally, it cannot be denied that sociopolitical perfection,


though I believe it historically crucial in our epoch, is logically
only a part of Bloch’s spectrum, which extends from alchemy
through immortality to omniscience and the Supreme Good. All
cognition can become the subject matter of an estranged verbal
construction dealing with a particular quasi-human community
treated as an alternative history. This “cognitive estrangement” is
the basis of the literary genre of SF. Strictly and precisely speak­
ing, utopia is not a genre but the sociopolitical subgenre of science
fiction. Paradoxically, it can be seen as such only now that SF has
expanded into its modern phase, “looking backward” from its
englobing of utopia. Further, that expansion was in some not
always direct ways a continuation of classical and nineteenth­
century utopian literature. Thus, conversely, SF is at the same
lime wider than and at least collaterally descended from utopia;
it is, if not a daughter, yet a niece of utopia—a niece usually
ashamed of the family inheritance but unable to escape her
genetic destiny. For all its adventure, romance, popularization,
and wondrousness, SF can finally be written only between the

44. First quotation from Barthes, p. 86. second from Elliott, p. 110.
62 DEFINING THE LITERARY GENRE OF UTOPIA

utopian and the anti-utopian horizons. All imaginable intelligent


life, including ours, can be organized only more or less perfectly.
In that sense, utopia (and anti-utopia) is first of all a literary
genre; but finally, as Bloch notes, it is a horizon within which
humanity is irrevocably collocated. My main point is that without
a full, that is, literal and literary, analysis we are bound to over­
simplify and miscontrue those horizons. For any sane under­
standing of utopia, the simple basic fact to start from remains
that it is not hypostasis of the Holy Ghost, the Zeitgeist, or what­
not, but a literary genre induced from a set of man-made books
within a man-made history.
4

SF and the Novum

0. It is often thought that the concept of a literary genre (here


SF) can be found directly in the works investigated, that the
scholar in such a genre has no need to turn to literary theory
since he/she will find the concepts in the texts themselves. True,
the concept of SF is in a way inherent in the literary objects—the
scholar does not invent it out of whole cloth—but its specific
nature and the limits of its use can be grasped only by employing
theoretical methods. The concept of SF cannot be extracted intu­
itively or empirically from the work called thus. Positivistic critics
often attempt to do so; unfortunately, the concept at which they
arrive is then primitive, subjective, and unstable. In older to
determine it more pertinently and delimit it more precisely, it is
necessary to educe and formulate the differentia specifica of the SF
narration. My axiomatic premise in this chapter is that SF is
distinguished by the narrative dominance or hegemony of a fictional
“novum" (novelty, innovation) validated by cognitive logic.
1. The Novum and Cognition
1.1 What is the common denominator the presence of which
is logically necessary and which has to be hegemonic in a narra­
tion in order that we may call it an SF narration? In other words,
how can the proper domain of SF be determined, what is the
theoretical axis of such a determining? The answering is clouded
by the present wave of irrationalism, engendered by the deep
structures of the irrational capitalist way of life which has re­
duced the dominant forms of rationality itself (quantification,
reification, exchange value, and so on) to something narrow,
dogmatic, and sterile inasmuch as they are the forms of reason­
ing of the dominant or of the dominated classes. Nonetheless, I
do not see any tenable intrinsic determination of SF which would
not hinge on the category of the novum, to borrow (and slightly
63
64 SF AND THE NOVUM

adapt) a term from the best possible source, Ernst Bloch.’ A


novum of cognitive innovation is a totalizing phenomenon or
relationship deviating from the author’s and implied reader’s
norm of reality. Now, no doubt, each and every poetic metaphor
is a novum, while modern prose fiction has made new insights
into man its rallying cry. However, though valid SF has deep
affinities with poetry and innovative realistic fiction, its novelty is
“totalizing” in the sense that it entails a change of the whole
universe of the tale, or at least of crucially important aspects
thereof (and that it is therefore a means by which the whole tale
can be analytically grasped). As a consequence, the essential ten­
sion of SF is one between the readers, representing a certain
number of types of Man of our times, and the encompassing and
at least equipollent Unknown or Other introduced by the
novum. This tension in turn estranges the empirical norm of the
implied reader (more about this later). Clearly the novum is a
mediating category whose explicative potency springs from its
rare bridging of literary and extraliterary, fictional and empiri­
cal, formal and ideological domains, in brief from its unalienable
historicity. Conversely, this makes it impossible to give a static
definition of it, since it is always codetermined by the unique, not
to be anticipated situationality and processuality that it is sup­
posed to designate and illuminate. But it is possible to distinguish
various dimensions of the novum. Quantitatively, the postulated
innovation can be of quite different degrees of magnitude, run­
ning from the minimum of one discrete new “invention” (gadget,
technique, phenomenon, relationship) to the maximum of a set­
ting (spatiotemporal locus), agent (main character or characters),
and/or relations basically new and unknown in the author’s en­
vironment. (Tangentially I might say that this environment is
always identifiable from the text’s historical semantics, always
bound to a particular time, place, and sociolinguistic norm, so
that what would have been utopian or technological SF in a given
epoch is not necessarily such in another—except when read as a
product of earlier history; in other words, the novum can help us
understand just how is SF a historical genre.)
1.2 The novum is postulated on and validated by the post-

1. In particular: Ernst Bloch, Das Prinzip Hojfnung (Bibliography II) and Experimmtum
Mundi (Frankfurt, 1976).
SF AND THE NOVUM 65

Cartesian and post-Baconian scientific method. This does not


mean that the novelty is primarily a matter of scientific facts
or even hypotheses; and insofar as the opponents of the old
popularizing Verne-to-Gernsback orthodoxy protest against such
a narrow conception of SF they are quite right. But they go too
far in denying that what differentiates SF from the “super­
natural" literary genres (mythical tales, fairy tales, and so on, as
well as horror and/or heroic fantasy in the narrow sense) is the
presence of scientific cognition as the sign or correlative of a
method (way, approach, atmosphere, sensibility) identical to that
of a modern philosophy of science.23 4Science in this wider sense
of methodically systematic cognition cannot be disjoined from
the SF innovation, in spite of fashionable currents in SF criticism
of the last 15 years—though it should conversely be clear that a
proper analysis of SF cannot focus on its ostensible scientific
content or scientific data. Indeed, a very useful distinction be­
tween “naturalistic” fiction, fantasy, and SF, drawn by Robert XT
Phiimus, is that naturalistic fiction does not require scientific ex­
planation, fantasy does not allow it, and SF both requires and
allows it?
Thus, if the novum is the necessary condition of SF (differ­
entiating it from naturalistic fiction)/ the validation of the
2. Beyond the discussion in chapter 1, see also my essays ‘“Utopian1 and 'Scientific’,”
The Minnesota Review N.S. No. 6 (J976), and “Science and Marxism. Scientism and Mar-
quit," ibidem No. 10 (1978).
3. The distinction is to be found in Robert M. Phiimus, “Science Fiction: From its
Beginning to 1870.“ in Barron, ed, (Bibliography I), pp. 5-6. My defining of SF is
indebted to some earlier discussions. In particular. I find myself in some respects near to
Kingsley Amis’s definition in chapter I of .Vein Maps of Hell (Bibliography I)—with the
significant difference of trying to go beyond his evasive basing of the SF innovation "in
science or technology, or pseudo-scicnce or pscudo-technology’’ (p. 18).
4. Works avowedly written within a nonrealislic mode, principally allegory (but also
whimsy, satire, and lying tall tale or Munchhausenjade), constitute a category for which
the question of whether they possess a novum cannot even be posed, because they do not
use the new worlds, agents, or relationships as coherent albeit provisional ends, but as
immediately transitive and narratively nonautonsmous means for direct and sustained reference
to the author’s empirical world and some system of belief in it. The question whether an
allegory is SF, and vice versa, is, strictly speaking, meaningless, but for classifying pur­
poses has to be answered in the negative. This means that—except for exceptions and
grey areas—most of the works of Kafka or Borges cannot be claimed for SF: though I
would argue that 1 n the Penal Colony and “The Library of Babel" would be among the ex­
ceptions. But admittedly, much more work remains to be done toward the theory of
modern allegory in order to render more precise the terms underlined in this note (see
also section 2.2. of this chapter).
66 SF AND THE NOVUM

novelty by scientifically methodical cognition into which the


reader is inexorably led is the sufficient condition for SF. Though
such cognition obviously cannot, in a work of verbal fiction, be
empirically tested either in the laboratory or by observation in
nature, it can be methodically developed against the background
of a body of already existing cognitions, or at the very least as a
“mental experiment” following accepted scientific, that is, cogni­
tive. logic. Of the two, the second alternative—the intrinsic, cul­
turally acquired cognitive logic—seems theoretically the crucial
one to me. Though I would be hard put to cite an SF tale the
noveltv in which is not in fact continuous with or at least analog­
ous to existing scientific cognitions, I would be disposed to accept
theoretically a faint possibility of a fictional novum that would at
least seem to be based on quite new, imaginary cognitions, be­
yond all real possibilities known or dreamt of in the author’s
empirical reality. (My doubts here are not so much theoretical as
psychological, for I do not sec how anybody could imagine some­
thing not even dreamt of by anyone else before; but then I do
not believe in individualistic originality.) But besides the “real”
possibilities there exist also the much stricter—though also much
wider—limits of “ideal” possibility, meaning any conceptual or
thinkable possibility the premises and/or consequences of which
are not internally contradictory.5 Only in “hard” or near-future
SF does the tale's thesis have to conform to a “real possibility”—
to that which is possible in the author’s reality and/or according
to the scientific paradigm of his culture. On the contrary, the
thesis of any SF tale has to conform to an “ideal possibility,” as
defined above. Any tale based on a metaphysical wish-dream—
for example omnipotence—is “ideally impossible” as a coherent
narration (can an omnipotent being create a stone it will not be
able to lift? and so forth), according to the cognitive logic that
human beings have acquired in their culture from the begin­
nings to the present day. It is intrinsically or by definition impos­
sible for SF to acknowledge any metaphysical agency, in the lit­
eral sense of an agency going beyond physis (nature). Whenever it
does so, it is not SF, but a metaphysical or (to translate the Greek
into Latin) a supernatural fantasy-tale.

5. I have been stimulated by the discussion of Ivan Foht. “Slika covjeka i kosmosa.”
Radin Beograd: Treci program (Spring 1974): 523-60.
SF AND THE NOVUM 67

1.3. Thus science is the encompassing horizon of SF, its “ini­


tiating and dynamizing motivation.’’6 I reemphasize that this
does not mean that SF is “scientific fiction” in the literal, crass, or
popularizing sense of gadgetry-cum-utopia/dystopia. Indeed, a
number of important clarifications ought immediately to be at­
tached: I shall mention three. A first clarification is that “horizon”
is not identical to “ideology.” Our view of reality or conceptual
horizon is, willy-nilly, determined by the fact that our existence
is based on the application of science(s), and I do not believe we
can imaginatively go beyond such a horizon; a machineless
Arcadia is today simply a microcosm with zero-degree industrial­
ization and a lore standing in for zero-degree science. On the
other hand, within a scientific paradigm and horizon, ideologies
can be and are either fully supportive of this one and only
imaginable state of affairs, or fully opposed to it, or anything
in between. Thus, anti-scientific SF is just as much within the
scientific horizon (namely a misguided reaction to repressive
—capitalist or bureaucratic—abuse of science) as, say, literary
utopia and anti-utopia both are within the perfectibilist hori­
zon. The so-called speculative fiction (for example, Ballard’s)
clearly began as and has mostly remained an ideological inver­
sion of “hard” SF. Though the credibility of SF does not depend
on the particular scientific rationtile in any tale, the significance
of the entire fictive situation of a tale ultimately depends on the
fact that “die reality that it displaces, and thereby interprets”7 is
interpretable only within the scientific of cognitive horizon,
A second clarification is that sciences humaines or historical-
cultural sciences like anthropology-ethnology', sociology, or lin­
guistics (that is, the mainly nonmathematical sciences) are equally
based on such scientific methods as: the necessity and possibility
of explicit, coherent, and immanent or non supernatural expla­
nation of realities; Occam’s razor; methodical doubt; hypothesis­
construction; falsifiable physical or imaginary (thought) ex peri -

6, Jan Trzyn.idlov.ski. “Prdba poetyki science fiction," in K, Bucizyk. ed., Z teorii i


historii litrratury (Warsaw. 19631. p. 272, see also Stanislaw Lern (Bibliography I); Rafail
Nudel’nian, "ConversaUcni in a Railway Compartment” (Bibliography VI); and Joanna
Russ. “ Towards an Aesthetic of Science Fiction,” in R, D. Mullen and Darko Suvin,
eds., Srinice-Firtiun Studies . . . 1973-1915 (Bibliography I), pp. 8-15.
7. Robert M. Philmus (Bibliography III A), p. 20.
68 5F AND THE NOVUM

merits; dialectical causality and statistical probability; progres­


sively more embracing cognitive paradigms; et sim. These “soft
sciences” can therefore most probably better serve as a basis
for SF than the “hard” natural sciences; and they have in fact
been the basis of all better works in SF—partly through the
characteristic subterfuge of cybernetics, the science in which
hard nature and soft humanities fuse. A third clarification,
finally, is that science has since Marx and Einstein been an
open-ended corpus of knowledge, so that all imaginable new
corpuses which do not contravene the philosophical basis of the
scientific method in the author’s times (for example, the
simulsequentialist physics in Le Guin’s The Dispossessed) can play
the role of scientific validation in SF.
1.4. It may be objected to this that a look into bookstores will
show thaL a good proportion of what is sold as SF is constituted
by tales of more or less supernatural or occult fantasy. However,
this is the result of an ideological and commercial habit of lump­
ing together SF (fiction whose novum is cognitively validated)
and fantastic narrative. A misshapen subgenre born of such
mingling is that of “science-fantasy,” extending from Poe
through Merritt to Bradbury’, about which 1 can only repeat the
even more pathological level—internalized in fictional creation
—this has led to tales that incongruously mingle science-fiction al
and fantastic narrative. A misshappen subgenre born of such
mingling is that of “science-fantasy,” extending from Poe
through Meritt to Bradbury, about which I can only repeat the
strictures of the late James Blish, who noted how in it “plausi­
bility is specifically invoked for most of the story, but may be cast
aside in patches at the author’s whim and according to no vis­
ible system or principle,” in “a blind and grateful abandonment
of the life of the mind.”8 In supernatural fantasy proper, the
supposed novelty rejects cognitive logic and claims for itself a
higher “occult” logic—whether Christian, a-Christian and indeed
atheistic (as is the case of H. P. Lovecraft), or, most usually, an

8. William Atheling, Jr., Mort Issues at Hand (Bibliography I), pp. 98 and 104. A
further warning in the same chapter that the hybrid of SF and detective tale leads^as I
would say. because of the incompatibility between the detective tale’s contract of infor­
mative closure with the reader and the manifold surprises inherent in the SF novum
system—to a trivial lower common denominator of the resulting tale has so far been
developed only by Rafail Nudel'man (see note 20).
SF AND THE NOVUM 69

opportunistic blend of both, openly shown in the more self­


confident nineteenth century by something like Marie Corelli’s
“Electric Christianity’' (the enormous popularity of which is
echoed right down to C. S. Lewis). The consistent supernatural
fantasy tale—one which does not employ only a single irruption
of the supernatural into everyday normality, as in Gogol’s Nose
or Balzac’s Peau de Chagrin, but develops the phenomenology of
the supernatural at the expense of the tension with everyday
norm—is usually (in England from Bulwer-Lytton on) a proto­
Fascist revulsion against modern civilization, materialist rational­
ism, and such. It is organized around an ideology unchecked by
any cognition, so that its narrative logic is simply overt ideology
plus Freudian erotic patterns. If SF exists at all, this is not it.
One of the troubles with distinctions in genre theory is, of
course, that literary history is full of “limit-cases." Let us briefly
examine one of considerable importance, Stevenson’s The Strange
Case of Dr. Jekyll and Mr. Hyde. Despite my respect for Stevenson’s
literary craftsmanship, I would contend that he is cheating in
terms of his basic narrative logic. On the one hand, his moral
allegory of good and evil takes bodily form with the help of a
chemical concoction. On the other, the transmogrification Jekyll-
Hyde becomes not only unrepeatable because the concoction
had unknown impurities, but Hyde also begins “returning" with­
out any chemical stimulus, by force of desire and habit. This
unclear oscillation between science and fantasy, where science is
used for a partial justification or added alibi for those readers
who would no longer be disposed to swallow a straightforward
fantasy or moral allegory, is to my mind the reason for the
elaborate, clever, but finally not satisfying exercise in detection
from various points of view—which in naturalistic fashion masks
but does not explain the fuzziness at the narrative nucleus. This
marginal SF is therefore, to my mind, an early example of "sci­
ence fantasy." Its force does not stem from any cognitive logic,
but rather from Jekyll’s anguish over his loss of control and from
the impact of the hidden but clearly underlying moral allegory.
The latter is particularly relevant to Victorian bourgeois repres­
sions of the nonutilitarian or nonofficial aspects of life and it also
holds forth an unsubstantiated promise that the oscillation be-
tw'een SF and fantasy does not matter since we are dealing with
full-blown allegory anyway (see note 4).
70 SF AND THE NOVUM

2. Narrative Consequences of the Novum


2.1. The presence of the novum as the determining factor of
an SF narration is crucially testable in its explanatory power for
the basic narrative strategies in this genre. First of all, the domi­
nance or hegemony of the cognitive novelty means that an SF
narration is not only a tale that includes this or that SF element
or aspect: utopian strivings or dystopian terrors of some kind, as
in the majority of world literature; moral allegories or tran­
scendental visions of other worlds, better or worse from our
own, as in much literature down to Milton, Swedenborg, and
coundess imitators; use of new technological gadgets, as in many
James Bond tales; and so on. An SF narration is a fiction in
which the SF element or aspect, the novum, is hegemonic, that is,
so central and significant that it determines the whole narrative
logic—or at least the overriding narrative logic—regardless of
any impurities that might be present.9
2.2. Furthermore, the novum intensifies and radicalizes that
movement across the boundary of a semantic field (defined by
the author’s cultural norm) which always constitutes the fictional
event.10 In “naturalistic” fiction this boundary is iconic and
isomorphic: the transgression of the cultural norm stands for a
transgression of the cultural norm; Mme. Bovary’s adultery
stands for adultery. In SF, or at least in its determining events, it
is not iconic but allomorphic: a transgression of the cultural
norm is signified by the transgression of a more than merely

9. A major objection against so-called thematic studies of SF elements and aspects,


from J. O. Bailey’s Pilgrims Through Space and Time (Bibliography I)—in 1947 no doubt a
pioneering work—to present-day atomistic and positivistic SF critics, is that these studies
ignore the determining feature of what they are studying: the narrative logic of a fic­
tional tale. Correlatively, they tend to become boring catalogs of raisins picked out of the
narrative cake, and completely desiccated in the process. This does not mean that critical
discussions of. say, artificial satellites, biological mutations, or new sexual mores in SF (or
other fiction) cannot be, for some strictly limited purposes, found useful: and for such
purposes we should probably know where the mutations, satellites, or sex patterns first
appeared and how they spread. But we should not be lured by this very peripheral
necessity into annexing any and every tale with a new gadget or psychic procedure into
SF, as, for example, Bailey did with Wilkie Collins's Moonstone and Thomas Hardy's Two
on a Tower. SF scholarship that does this is sawing off the branch on which it is sitting: for
if these and such works are SFjust like, say, Wells's Invisible Man, then in fact there is no
such thing as SF.
10. Jurij Lotman. The Structure of the Artistic Text, .Michigan Slavic Contributions No.
7 (Ann Arbor, 1977), pp. 229 ff.
SF AND THE NOVUM 71

cultural, of an ontological, norm, by an ontic change in the


character/agent’s reality either because of his displacement in
space and/or time or because the reality itself changes around
him. I do not know a better characterization than to say that the
novelty makes for the SF narration’s specific ontolytic effect and
properties. Or perhaps—since, as differentiated from fantasy
tale or mythological tale, SF does not posit another superordi-
nated and "more real” reality but an alternative on the same
ontological level as the author’s empirical reality—one should say
that the necessary correlate of the novum is an alternate reality,
one that possesses a different historical time corresponding to dif­
ferent human relationships and sociocultural norms actualized
by the narration. This new reality overtly or tacitly presupposes
the existence of the author's empirical reality, since it can be
gauged and understood only as the empirical reality modified in
such-and-such ways. Though I have argued that SF is not—by
definition cannot be—an orthodox allegory with any one-to-one
correspondence of its elements to elements in the author’s real­
ity, its specific modality of existence is a feedback oscillation that
moves now from the author's and implied reader’s norm of real­
ity to the narratively actualized novum in order to understand
the plot-events, and now back from those novelties to the au­
thor’s reality, in order to see it afresh from the new perspective
gained. This oscillation, called estrangement by Shklovsky and
Brecht, is no doubt a consequence of every poetic, dramatic,
scientific, in brief semantic novum. However, its second pole is in
SF a narrative reality sufficiently autonomous and intransitive to
be explored at length as to its own properties and the human
relationships it implies. (For though mutants or Martians, ants or
intelligent nautiloids can be used as signifiers, they can only sig­
nify human relationships, given that we cannot—at least so far—
imagine other ones.)
2.3. The oscillation between the author’s “zero world" and the
new reality induces the narrative necessity of a means of reality
displacement. As far as I can see, there are two such devices: a
voyage to a new locus, and a catalyzer transforming the author’s
environment to a new locus; examples for the two could be
Wells’s Time Machine and Invisible Man. The first case seems bet­
ter suited to a sudden and the second to a gradual introduction
of a new1 reality; no doubt, all kinds of contaminations and twists
on these two means are thinkable. When the in medias res tech­
72 SF AND THE NOVUM

nique is used in any particular SF tale, the means of displace­


ment can be told in a retrospective or they can, apparently, to­
tally disappear (more easily in a space/time displacement: our
hero is simply a native of elsewhere/elsewhen). However, this
semblance conceals the presence of displacement in a zero-form,
usually as a convention tacitly extrapolated from earlier stories;
the history of the genre is the missing link that made possible,
for example, tales in another space/time without any textual
reference to that of the author (as in most good SF novels of the
last 20 years).
2.4. The concept of novum illuminates also the historical
vicissitudes of justifying the reality displacement. In naturalistic
tales the voyage can only start in the author’s space, and the
account of the new reality has to arrive back into that space so
that its telling may be naturalistically plausible. However, it
would then be logically necessary that the account of such a
sensational voyage to a new reality should in its turn become a
catalyzer, inducing changes in the author’s and reader’s envi­
ronment. Since ibis in fact, as the reader knows, has not hap­
pened, naturalistic SF has had to invent a number of lightning-
rods to dissipate such expectations. Verne pretended not to
notice its necessity, while Wells in some of his tales pretended we
all knew it already—ploys which today make those narrations
sound as if they assumed an alternate time-stream in which
Nemo or the Invisible Man had in fact (as different from the
reader’s time-stream) been the scourge of the seas or of southern
England. Many earlier writers tvent through other extraordinary
contortions to satisfy naturalistic plausibility, usually a contami­
nation of the “manuscript in a bottle” device (the news of a
voyage to the Moon just having arrived by volcanic eruption
from it and just being served piping hot to you, dear reader) and
the “lost invention” device (a one-shot novelty confined to the
experience of a few people and unable to extend beyond them
because of the loss of the invention), as in The First Men in (he
Moon. But the most plausible variable for manipulation was time,
inasmuch as setting the tale in the future immediately dispensed
with any need for empirical plausibility. The shift of SF from
space into future time is not simply due to an exhaustion of
white spots on the mappa mundi. Rather, it is due to an interac­
tion of two factors: on the one hand, such a narrative conve­
SF AND THE NOVL’M 73

nience, stunted within strict positivist ideology ; on the other, the


strong tendency toward temporal extrapolation inherent in life
based on a capitalist economy, with its salaries, profits, and pro­
gressive ideals always expected in a future clock-time.
Thus space was a fully plausible locus for SF only before the
capitalist way of life, from very early tales about the happy or
unhappy valley or island—known to almost all tribal and ancient
societies—to More and Swift. An Earthly Paradise or Cockayne
tale, a humanist dialogue and satire, all happen in a literary or
imaginative space not subject to positivistic plausibility. But a
triumphant bourgeoisie introduces an epoch-making epistemo­
logical break into human imagination, by which linear or clock­
time becomes the space of human development because it is the
space of capitalist industrial production. The spatial domin­
ions of even the largest feudal landowner are finite; capital,
the new historical form of property—that shaper of human ex­
istences and relationships—has in principle no limits in extrapo­
lated time. Through a powerful system of mediations infusing
the whole human existence, time becomes finally the equiva­
lent of money and thus of all things. The positivist ideology
followed capitalist practice in eventually perfecting an image of
time rigid ¡fled “into an exactly delimited, quantifiable continuum
filled with quantifiable ‘things’ . . . : in short, [time] becomes
space.”" Imaginative times and spaces are now resolved into
“positive," quantified ones. All existential alternatives, for better
or worse, shift into such a spatial ¡zed future, which now becomes
the vast ocean on whose other shore the alternative island is to be
situated. Positivism shunts SF into anticipation, a form more ac­
tivistic than the spatial exemplum because achievable in the im­
plied reader's own space. When the industrial revolution be­
comes divorced from the democratic one—a divorce which is the

11. Georg Lukács, History and Class Consdousnrss (London, 1971), p. 90. The whole
seminal essay “Reification and the Consciousness of the Proletariat," developing insights
from Marx’s Capital, is to be consulted; also Georg Si mm el, Philosophie des Grides (Munich.
1930), VVerner Sombart, Der Mndeme Kapila&smus, I—II/I (Munich, 1917), and Lewis
Mumford (Bibliography IV A). I have tried to apply Lukacs's ideas on quantification and
reification in my essays "On Individualist World View in Drama," Zagadmenia rodzajow
literackih 9. No. I (1966). and “Beckett's Purgatory of the Individual." Tulane Drama
Review 2. No. 4 (1967), and in the historical part of this book, especially in the essay on
Verne as the bard of movement in such a quantified space.
71 SF AND THE NOVUM

fundamental political event of the bourgeois epoch—activism be­


comes exasperated and leads to the demands for another epis­
temological and practical break, signalled by Blake’s Jerusalem in
England’s green and pleasant land and the cosmic “passionate
attraction” of Fourier’s phalansteries. Such imaginative energies
converge in Marx, the great prefigurator of the imaginative shift
still being consummated in our times. Rather than identify it as
"postindustrial" (a fairly reified and vague term), I would tend to
call the new episteme—since it is in our century' marked by names
such as Einstein, Picasso, Eisenstein, and Brecht—one of spat­
iotemporal covariance, simulsequentialism, or humanist rela­
tivism and estrangement: in brief, one of alternate historical
realities. I would argue that in such a historical perspective, all
significant SF from Zamyatin, Capek, and Lem to Le Guin,
Disch, and Delany is neither simply spatial, as in Lucian or More,
nor simply temporal, as in all the followers of The Time Machine
and When the Sleeper Wakes, but spatiotemporal in a number of
very interesting ways, all of which approximate a reinvention
and putting to new uses of the precapitalist and preindividualis-
tic analogic times and spaces of the human imagination.
The main difference with such medieval and premedieval con­
ceptions could perhaps be expressed in terms of destiny. As
Ixjtman remarks, literary functions can be divided into two
groups, the active forces and the obstacles.12 Right down to Swift
(in SF and in literature in general), the obstacles are inhuman
and superhuman forces, at best to be ethically questioned by the
tragic poet and hero but not to be materially influenced.
Whether they are called gods, God, Destiny, Nature, or even
History is relatively less important than the fact that they are
transcendental, empirically unchangeable. The great enlighten­
ing deed of the bourgeoisie was to reduce the universe to indi­
viduals. which also meant identifying the obstacles with men,
who are reachable and perhaps removable by other individual
men. I would imagine that a truly modern literature (and SF),
corresponding to our epoch, its praxis and episteme, would corre­
spond to the third dialectical term to follow on such fatalistic
collectivism and humanistic individualism. We have learned that
the institutional and imaginative products of men—states, corpo­

12. Lotman, p, 239.


SF AND THE NOVUM 75

rations, religions, wars, and the like—can very well become a


destiny for each of us: tragedy is again possible in the twentieth
century (as the October Revolution and Second World War,
Dubcek and Allende can teach us), though it is the tragedy of
blindness—of failed historical possibilities—rather than of lucid­
ity. The obstacles are superindividual but not inhuman; they
have the grandeur of the ancient Destiny but they can be over­
come by other men banding together for the purpose. Men are
the historical destiny of man; the synthesis in this historical triad
is a humanistic collectivism.
2.5. The alternate reality logically necessitated by and pro­
ceeding from the narrative kernel of the novum can only func­
tion in the oscillating feedback with the author’s reality suggested
in 2.2 because it is as a whole—or because some of its focal rela­
tionships are—an analogy to that empirical reality. However fan­
tastic (in the sense of empirically unverifiable) the characters or
worlds described, always de nobis fabula narratur. Though SF is
not orthodox allegory, it transmits aesthetic information in direct
proportion to its relevance and aesthetic quality. The alternative
is for it to operate in semantic emptiness spiced with melodra­
matic sensationalism as a compensatory satisfaction, in a run­
away feedback system with corrupt audience taste instead of
with cognition of tendencies in the social practice of human re­
lationships.13 The clear dominance of that kitsch alternative in
the present historical period should not, however, prevent us
from discussing the significant models of SF, its horizons and
yardsticks.
In my second chapter I considered heuristic models of SF
under the headings of (1) extrapolation, which starts from a
cognitive hypothesis incarnated in the nucleus of the tale and
directly extrapolates it into the future, and (2) analogy, in which

13. “The information gained, concerning a hypothesis, may perhaps be thought of as


the ratio of the a posteriori to the a priori probabilities (strictly the logarithm of this
ratio)"—Colin Cherry, On Human Communication (Cambridge, MA, 1966), p. 63. Thus,
the information gained from a work of literature is a logarithmic (that is, alas, much
diminished) ratio of the existential possibilities imaginable and understandable by an
ideal reader after reading, to those imaginable and understandable before the reading.
The information is a function of the rearrangement of the reader’s understanding of
human relationships. “In general, where we speak of information, we should use the
word form.” argues René Thom in his impressive Stabilité structurelle et morphogénese
(Reading, MA. 1972), p. 133.
7B SF AND THE NOVUM

cognition derives only from the final import or message of the


tale, and may perhaps be only indirectly applicable to pressing
problems in the author’s environment. This analysis is, so far as
it goes, useful in challenging the defining of all SF as extrapola­
tion (to which the title of a critical journal devoted to SF still
witnesses); but it does not go far enough. In that chapter I also
noted that any futurological function SF might have was strictly
secondary, and that stressing it was dangerous since it tended to
press upon SF the role of a popularize!' of the reigning ideology
of the day (technocratic, psionic, utopian, dystopian, hip, or
whatever). Thus, although extrapolation was historically a con­
vention of much SF (as analyzed at length in the second section of
this book), pure extrapolation is flat, and the pretense at it masks
in all significant cases the employment of other methods.
Theoretical defining of any SF as extrapolation should therefore
be decently and deeply buried.14 It seems clear that SF is mate­
rial for futurology (if at all) only in the very restricted sense of
reflecting on the author's own historical period and the pos­
sibilities inherent in it: Bellamy’s and Morris’s different socialist
twenty-first centuries use the anticipation device so effectively
because they are about incipient collective human relationships
in the 1880s as they (differently) saw them, while 1984 or 2001
are about incipient collective human relationships in 1948 or
1967 as certain aspects of or elements within Orwell’s or Ku­
brick’s mind saw them.
Any significant SF text is thus always to be read as an analogy,
somewhere between a vague symbol and a precisely aimed para­
ble, while extrapolative SF in any futurological sense was (and is)
only a delusion of technocratic ideology—no doubt extremely
important for the historical understanding of a given period of
SF, but theoretically untenable. For extrapolation itself as a sci­
entific procedure (and not pure arithmetic formalization) is pre­
dicated upon a strict (or, if you wish, crude) analogy between
the points from and to which the extrapolating is carried out:
extrapolation is a one-dimensional, scientific limit-case of analogy. As
Peirce put it, a scientific “effect” (or “phenomenon") “consists in

14. Wells knew this already in 1906, see note 19 in my chapter 10 and the self-
criticism it refers to. On the discussion of extrapolative, analogical, and other models for
SF see also Philnius, note 3. and Fredric Jameson, "Generic Discontinuities in SF’ and
"World Reduction in Le Guin," both in Mullen and Suvin, eds., pp. 28-39 and 251-60.
SF and the novum 77

the fact that when an experimentalist shall come to act accord­


ing TO A CERTAIN SCHEME THAT HE HAS IN MIND [caps blit not
italics mine], then will something else happen, and shatter the
doubts of sceptics, like the celestial fire upon the altar of
Elijah.”15 Specifically, the SF “future-story” has been well iden­
tified by Raymond Williams as
the finding and materialization of a formula about society. A
particular pattern is abstracted, from the sum of social ex­
perience, and a society is created from this pattern ... the
“future” device (usually only a device, for nearly always it is
obviously contemporary society that is being written about
, . .) removes the ordinary tension between the selected
pattern and normal observation.16
Clearly, neither is the future a quantitatively measurable space
nor will the ensemble of human relationships stand still for
one or more generations in order for a single element (or a
very few elements) to be extrapolated against an unchanging
background—which is the common invalidating premise of
futurological as well as of openly fictional extrapolation. The
future is always constituted both bv a multiple crisscrossing of
developments and—in human affairs—by i mem io ns, desires,
and beliefs rather than only by quantifiable facts. It is Peirce's
scheme or Williams’s pattern rather than the end-point of a line.
Furthermore, anticipating the future of human societies and
relationships is a pursuit that shows up the impossibility of using
the orthodox—absolute or scientistic—philosophy of natural sci­
ence as the model for human sciences. It is a pursuit which
shows, first, that all science (including natural sciences) is and
always has been a historical category, and second, that natural or
“objective” and human (cultural) or “subjective” sciences are ul­
timately to be thought of as a unity: “Natural science will in lime
include the science of man as the science of man will include
natural science. There will be one science”—remarked an acute
observer already in the first part of the nineteenth century.17 As
15. Charles Sanders Peirce, "What Pragmatism Is,” Collected Papers, ecl. Charles
Hartshorne and Paul Weiss (Cambridge. MA, 1934). para. 425, p. 284.
16. Raymond Williams. The Long Revolution (Harmondsworth, 1971), p. 307.
17. Karl Marx, ‘‘Private Property and Communism," Writings of the Young Marx on
Philosophy and Society, ed. Ixiyd D. Easton and Kurt H. Guddat (Garden City. NY, 1967),
p. 312.
78 SF AND THE NOVUM

a corollary, the valid SF form or subgenre of anticipation—tales


located in the historical future of the author’s society—should be
strict!} differentiated from the technocratic ideology of extrapo­
lation on the one hand and the literary device of extrapolation
on the other. Extrapolating one feature or possibility of the au­
thor’s environment may be a legitimate literary device of hyper-
boiization equally in anticipation-tales, other SF (for example,
that located in space and not in the future), or indeed in a
number of other genres such as satire. However, the cognitive
value of all SF, including anticipation-tales, is to be found in its
analogical reference to the author’s present rather than in pre­
dictions, discrete or global. Science-fictional cognition is based on
an aesthetic hypothesis akin to the proceedings of satire or pas­
toral rather than those of futurology or political programs.
The problem in constructing useful models for SF is, then, one
of differentiating within analogy. If every SF talc is some kind of
analogy—and I think that The Time Machine or The Iron Heel,
Heinlein’s Future History or Pohl-Kornbluth’s Space Merchants,
even Stapled on's Last and First Men or Yefremov’s Andromeda, are
primarily fairly clear analogies to processes incubating in their
author’s epoch—then just what is in each case the degree and the
kind of its anamorphic distortion, its “version” of reality? How is
their implied reader supposed to respond to and deal with ¿1
narrative reality that is an inverted, reverted, converted, everted,
averted, subverted Other to his certainties of Self and Norm—
certainties which, as Hegel says, are clouded by (heir very illusion
of evidence and proximity, bekannt but not erkanntV9 A partially
illuminating answer to this group of questions would also clear
up why some of these versions pretend—sometimes with convic­
tion, most often by pure convention—to be situated in an ex­
trapolated future.
2.6. A final narrative consequence of the novum is that it
shapes the SF “chronotope” (or chronotopes?). A chronotope is
“the essential connection of temporal and spatial relationships, as
shaped in literary art.” In it, “the characteristics of time are un­
folded in space, while space is given meaning and measured by18

18. Georg Wilhelm Friedrich Hegel, Phdnomenoiogte des Gristes. Sdmllicht Wtrkf (Leip­
zig, 1949), 2:28.
SF AND THE NOVUM 79

lime”1920—and both are blended into a particular plot structure.


Now the novelty in SF can be either a new locus, or an agent
(character) with new powers transforming the old locus, or a
blend of both. The connection between the active forces (the
protagonistfs]) and the obstacles to be reduced (the locus) deter­
mines the homogeneity of a tale. If the protagonists and the loci
necessarily imply and richly reinforce each other—as do Wells’s
l ime Traveller and the sequence of his devolutionary visions of
the future, or Le Guin’s Shevek, his physics, and the binary
planetary socio politics and psychology of The Dispossessed—then
we have a tale of a higher quality than the wish-dreams of, say, a
Van Vogt, where all the obstacles are fake since the protagonist
is a superman enforcing his will both on enemies and supposed
allies.
As for plot structures, if SF is organized around an irreversible
and significant change in its world and agents, then a simple
addition of adventures, where plus ça change plus c’est la même
chose, is an abuse of SF for purposes of trivial sensationalism,
which degrades the genre to a simpler and less organized plot
structure. Nudclman has to my mind brilliantly demonstrated
the incompatibility of the plot structure of the cyclical detective
tale, the conclusion of which returns the universe “to its equilib­
rium and order," the linear structures of the additive adventure
tale, and the spiral structures of valid SF, the plot of which alters
the universe of the tale.30 On the contrary, the easiest narrative
way of driving a significant change home is co have the hero or
heroine grow into it (or better, to have the hero or heroine
define it for the reader by growing with it), and much valid SF
uses the plot structure of the “education novel,” with its initially
naive protagonist who by degrees arrives at some understanding
of the novum for her/himself and for the readers.
As these two examples and other discussions in this chapter
may indicate, it should be possible to engage in analytic evalua­
tions of SF that would be neither purely ideological nor purely
formalistic, by starting with the necessities of literary structure
brought about by some variant of a novum.
19. M. Bakhtin. Vopraiy litrratuvy i èstrtikt (Moscow, 1975), pp. 234—35.
20. Rafail Nudel'tnan, “An Approach to the Structure of Le Guin’s SF." in Mullen and
Suvin, eds.. pp. 240-50.
80 SF AND THE NOVUM

3. The Novum and History


3.1 The novum as a creative, and especially as an aesthetic,
category is not be fully or even centrally explained by such
formal aspects as innovation, surprise, reshaping, or estrange­
ment, important and indispensable though these aspects or fac­
tors are.21 The new is always a historical category since it is always
determined by historical forces which both bring it about in so­
cial practice (including art) and make for new semantic meanings
that crystallize the novum in human consciousnesses (see 1.1 and
2.2) An analysis of S F is necessarily faced with the question of
why and how was the newness recognizable as newness at the
moment it appeared, what ways of understanding, horizons, and
interests were implicit in the novum and required for it. The
novelty is sometimes directly but sometimes in very complex
ways (for example, not merely as reflection but also as préfigura­
tion or negation) related to such new historical forces and
patterns—in the final instance, to possibilities of qualitative dis­
continuity in the development of human relationships. An
aesthetic novum is either a translation of historical cognition and
ethics into form, or (in our age perhaps more often) a creation of
historical cognition and ethics as form.
3.2. Probably the most important consequence of an under­
standing of SF as a symbolic system centered on a novum which
is to be cognitively validated within the narrative reality of rhe
tale and its interaction with reader expectations is that the
novelty has to be convincingly explained in concrete, even if
imaginary, terms, that is, in terms of the specific time, place,
agents, and cosmic and social totality of each tale. This means
that, in principle, SF has to be judged, like most naturalistic or
“realistic" fiction and quite unlike horror fantasy, by the density
and richness of objects and agents described in the microcosm of
the text. Another way of interpreting the Phil mus distinction
from 1.2. would be to set up a further Hegelian triad, where the
thesis would be naturalistic fiction, which has an empirically vali­
dated effect of reality, the antithesis would be supernatural

21, See, for development of estrangement and similar notions after the Formalists
and Brecht, Hans Robert Jauss, Literalurgescluchtii als Provokatiim (Frankfurt, 1970), as well
as critiques of and improvements on jauss handily asembled in Peter Uwe Hohendahl.
ed., Sozialgeschir.kle unrl Wirkung.wthetik (Frankfurt, 1974).
SF AND THE NOVUM 81

genres, which lack such an effect, and the synthesis would be SF,
in which the effect or reality is validated by a cognitive innova­
tion. Obversely, the particular essential novum of any SF tale
must in its turn be judged by how much new insight into imagi­
nary but coherent and this-worldly, that is, historical, relation­
ships it affords and could afford.
3.3 In view of this doubly historical character of the SF
novum—born in history and judged in history—this novum has
to be differentiated not only according to its degree of mag­
nitude and of cognitive validation (see 1.1. and 1.2.), but also
according to its degree of relevance. What is possible should be dif­
ferentiated not only from what is already real but also from what
is equally empirically unreal but necessary. Not all possible novel­
ties will be equally relevant, or of equally lasting relevance, from
the point of view of, first, human development, and second, a
positive human development. Obviously, this categorization im­
plies, first, that there are some lawlike tendencies in men’s social
and cosmic history, and second, that we can today (if we are
intelligent and lucky enough) judge these tendencies as parts of a
spectrum that runs from positive to negative. I subscribe to both
these propositions and will not argue them here—partly for
rhetorical convenience, but mainly because 1 cannot think of any
halfway significant SF narration that does not in some way sub­
scribe to them in its narrative practice (whatever the author’s
private theories may be).
Thus a novum can be both superficially sweeping and cogni­
tively validated as not impossible, and yet of very limited or brief
relevance. Its relationship to a relevant novelty will be the same
as the relationship of the yearly pseudo-novum of “new and
improved" (when not “revolutionary”) car models or clothing
fashions to a really radical novelty such as a social revolution and
change of scientific paradigm making, say, for life-enhancing
transport or dressing. The pseudo-novum will not have the
vitality of a tree, an animal species, or a belief but, to quote
Bergson, the explosive, spurting élan of a howitzer shell explod­
ing into successively smaller fragments, or “of an immense fire­
works, which continually emits further firesparks from its
midst.”22 In brief, a novum is fake unless it in some way partici-

22. Henri Bergson. L'fvolution creatnce (Paris. 1907). pp. 99 and 270; see also Bloch's
82 SF AND THE NOVUM

pates in and partakes of what Bloch called the “front-line of


historical process"—which for him (and for me) as a Marxist
means a process intimately concerned with strivings for a
dealienation of men and their social life. Capricious contingen­
cies, consequent upon market competition and tied to copyright
or patent law, have a built-in limit and taboo defined precisely
by the untouchable sanctity of competition (a palpable ideology
in much SF), Of brief and narrow relevance, particular rather
than general (kath’hekaston rather than kath’holon, as Aristotle
puts it in Poeties), they make for a superficial change rather than
for a true novelty that deals with or makes for human relation­
ships so qualitatively different from those dominant in the au­
thor’s reality that they cannot be translated back to them merely
by a change of costume. All space operas can be translated back
into the Social Darwinism of the Westerns and similar adven­
ture-tales by substituting colls for ray-guns and Indians for the
slimy monsters of Betelgeuse. Most novels by Asimov can be
returned to their detective-story model by a slightly more com­
plex system of substitutions, by which, for example, Second
Foundation came from Poe’s Purloined Letter.
3.4. Since freedom is the possibility of something new and
truly different coming about, “the possibility of making it differ­
ent,"*23 the distinction between a true and fake novum is, interest­
ingly enough, not only a key to aesthetic quality in SF but also to
its ethico-political liberating qualities. As always in art, ethical
pathos and effect or communal (jxjlitical) relevance are the ob-
ve-se of aesthetic consistency. They fuse in the realization that,
finally, the only consistent noveltv is one that constitutes an
open-ended system “which possesses its novum continually both
in itself and before itself; as befits the unfinished state of the
world, nowhere determined by any transcendental supra worldly
formula.”24 This connects with my argument in 1.3. about valida-
commenl on him in Dai Prinzip Hof]nung. p. 23J—my whole argument in 3.1,-3.4. is
fundamentally indebted to Bloch, See on originality within a capitalist market also Bertolt
Brecht. Gesammelte Werke (Frankfurt. 1973). 1-20, passim—for example. 15:199-200—
and Theodor W. Adorno, Aesthetische Throne (Frankfurt, 1970). pp. 257 ff.
23. Bloch, Experimentum, p. 139; see also Antonio Gramsci, ll Materialismo storico e la
filosofía di Benedetto Croce (Torino, 1948), quoted from Selections from the Prison Notebooks,
ed. Quintin Hoare and Geoffrey Nowell Smith (New York, 1971 >, p. 360, to whom 1 am
also much indebted.
24. Bloch. Expenmenium, p. 143.
SF AND THE NOVUM 83

tion for SF being based on science as an open-ended corpus of


knowledge, which argument can now be seen to be ultimately
and solidly anchored to the bedrock fact that there is no end to
history, and in particular that we and our ideologies are not the
end-product history has been laboring for from the time of the
first saber-toothed tigers and Mesopotamian city-states. It follows
that SF will be the more significant and truly relevant the more
clearly it eschews final solutions, lx? they the static utopia of the
Plato-More model, the more fashionable static dystopia of the
Huxley-Orwell model, or any similar metamorphosis of the
Apocalypse (let us remember that the end of time in the Apocalypse
encompasses not only the ultimate chaos but also the ultimate
divine order).25
3.5. An imaginary history each time to be reimagined afresh
in its human significance and values may perhaps borrow some
narrative patterns from mythological tales, but the “novelty” of
gods validated by unexplained superscicnces at the beck of the
Cambridge School's or von Däniken’s supermortals is a pseudo­
novelty, old meat rehashed with a new sauce. SF's analogical
historicity may or may not be mythomorphic, but—as I have
argued in chapter 2—it cannot be mythopoetic in any sense ex­
cept the most trivial one of possessing “a vast sweep” or “a sense
of wonder”: another superannuated slogan of much SF criticism
due for a deserved retirement into the same limbo as extrapola­
tion. For myth is reenactment, eternal return, and the opposite
of a creative human freedom.
True, even after one subtracts the more or less supernatural
tales (science-fantasy, sword-and-sorcery, and the like,) 90 per­
cent of SF will have plot structures escaping from history into
Westerns, additive sensationalist adventures, or rehashes of
mythography. However, as Kant said, a thousand years of any
given state of affairs do not make that state necessarily right.
Rather, reasons for rhe wrongness should be sought.
3.6 Thus this analysis has finally arrived at the point where
history, in the guise of analogical historicity, is found to be the
25. 1 have attempted to expand on this in my “The Open-Ended Parables of Stanis­
law Lem and Solaris," afterword to Stanislaw Lem. Solaris (New York, 1976); it is incorpo­
rated into a parallel to the orthodox Soviet and American SF models in my “Stanislaw
Lem und das mitteleuropäische soziale Bewusstsein der Science Fiction.“ in Werner
Berthel, ed., Insel Almanach auf das Jahr 1976: Stanislaw Lern (Frankfurt, 1976).
84 SF AND THE NOVUM

next and crucial step in the understanding of SF: story is always


also history, and SF is always also a certain type of imaginative
historical tale (which could be usefully compared and contrasted
to the historical novel). -All the epistemological, ideological, and
narrative implications and correlatives of the novum lead to the
conclusion that significant SF is in fact a specifically roundabout
way of commenting on the author’s collective context—often re­
sulting in a surprisingly concrete and sharp-sighted comment at
that. Even where SF suggests—sometimes strongly—a flight from
that context, this is an optical illusion and epistemological trick.
The escape is, in all such significant SF, one to a better vantage
point from which to comprehend the human relations around
the author. It is an escape from constrictive old norms into a
different and alternative timestream, a device for historical es­
trangement, and an at least initial readiness for new norms of
reality, for the novum of dealienating human history. I believe
that the critic, in order to understand it properly, will have to
integrate sociohistorical into formal knowledge, diachrony into
synchrony. History has not ended with the “post-industrial” soci­
ety: as Bloch said, Judgment Day is also Genesis, and Genesis is
every day.
II

history
Introduction to Older SF History

Let’s be realistic—let’s demand the impossible.


Anonym us Sorbonensis (May 1968)
The history of science fiction, as the genre is defined in thefirst part of
this book, gives rise to a number of significant and fascinating problems,
which can in the present state of our knowledge be rather identified than
resolved. One problem is the appearance of what seem to be temporal
groupings or clusters—periods with a noticeably higher frequency of SF
texts, separated from each other by gaps with a statistically significant
lower frequency of SF texts. I am, alas, incompetent to even enter into
tribal and extra-European narrative traditions (such as the Chinese one)
and their independent but rich histories.1 But the Euro-Mediterranean
tradition alone, of which this second part wishes to give a partial and
abbreviated overview, consists—so far as we can now tell—most probably
of six clusters: the Hellenic one (from folk myths and legends reactualized
in Aeschylus and Aristophanes to Plato, Theopompus, Euhemerus,
Hecataeus, and lambulus), the Hellenistic-cum-Roman one (from Virgil
to Antonius Diogenes and Lucian), the Renaissance-Baroque or
Columbus-to-Louis-XIV one (ca. 1500-1660), the cluster of the demo­
cratic revolution (mainly 1770-1820), the fin-de-siecle cluster (ca.
1870-1910), and the modern SF cluster in the last 50 years or so. In the
meantime, in periods of absolutist practice and world view—be they
Ptolemaic or Newtonian—the subversive tradition of SF was driven un­
derground (for example, the oral literature, apocrypha, and heretical
writings of the Middle Ages) or into exile (French SF between the Fronde
and the Encyclopedists; or the “lowbrow’’ and juvenile magazines and
novels of the century after Frankenstein, from which Verne, the utopias
grouped around Bellamy, and Wells emerge as volcanic archipelagos
from an ocean). Not all of this is quite clear, because SF (if one agrees to
this namefor the genre grouping alternative historical worlds) has been a
suppressed and neglected, often materially and most always ideologically
persecuted tradition: it is hardly an accident that except for conservatives
such as Aeschylus, Aristophanes, and Plato its first two clusters survive

I. Sec Simon in Bibliography III A. and Bauer, Chesneaux, and Nuila in Bibliography
in b.
87
88 INTRODUCTION TO OLDER SF HISTORY

only in fragments and references, or that from Kepler, Francis Godwin,


and Cyrano to Mark Twain its texts often had to be published posthu­
mously.
Thus both modern and older SF are only now beginning to be iden­
tified in scholarly bibliographies, and it is possible we willfind out that its
historical frequency is perceptibly higher. Nonetheless, I do not believe
that new data will substantially affect the basic cultural hypothesis of a
coherent literary tradition of SF as part of a popular literature that (like
many forms of humor and “obscenity") spread through centuries by word
of mouth and other unofficial channels, and penetrated into officially
accepted, normative, or “high" Literature and Culture only at favorable
historical moments. (For example, evidence is emerging about the possibly
high incidence of utopias and marvelous voyages in the period 1660-
1770, but when these works were not rendered more or less harmless by
the author’s timidity, as in the cases of Paltock or Fénelon, they were
forbidden by the authorities, as in the cases of Foigny or Vairasse.)
However, those works which did break through the surface of officially
recorded and recordable “higher” culture almost by definition had to be
significant; their resonance and echoes were certainly sufficient to estab­
lish an apparently tenuous yet potent and lucidly self-conscious intellec­
tual and formal tradition. Plato, Lucian, More, Rabelais, Swift, Di­
derot, Verne—writers who succeeded in breaking through because of
■superior personal talent, or cunning and luck in finding an interstitial
political time and social space in which to go public, or (usually) a
combination of both factors—are therefore not merely among the foun­
tainheads and transmitters of that tradition. Even more importantly, in
view of the largely suppressed SF tradition, the achievement of each such
major writer not only has to but also legitimately can indicate and stand
for the possibilities of a largely mute inglorious epoch. (In this book, the
first two European clusters will be dealt with only obliquely, through their
effect on the Renaissance.) No doubt this perverts somewhat what “really
happened" in cultural and literary history, but no more so than any
historical investigation, dealing, as it must always, with a choice from
whatever data have survived rather than wie es eigen dieh gewesen
(how it really was). An ideal history—especially a history of culture—
would have to be a geology, interested perhaps as much in the hollows
produced by absence of data as in the fullnesses produced by their pres­
ence, or a geography of the ocean depths as much as of the visible islands.
I confess that this book is not such an ideal, although I propose to suggest
how SF, sustained by subordinate social groups with which it achieves
INTRODUCTION TO OLDER SF HISTORY 89

and loses cultural legitimacy, is like an iceberg showing only a fraction


above the silent surface of officially recorded culture, and how the islands
limn not only themselves but also the oceans from which they grow.
It could be argued that SF always fuses the old rhetorical trope of “the
impossibilities” (impossibilia) with the equally venerable notion of the
wishedfor country into a new and fertile form in which autonomous
worlds are opposed to the author’s empirical environment and its norms;
and that, historically, at least the initial impulse for SF comes always
from the yearnings of a repressed social group and testifies to radically
other possibilities of life. Nonetheless, the different historical functions
and purposes in the various clusters have molded the SF tradition into
different subgenres. Its central watershed is around 1800, when space
loses its monopoly upon the location of estrangement and the alternative
horizons shift from space to time (however this shift might be curbed by
ideological hesitations about a truly different future). Some reasons for
the shift to anticipation have been brought forward in chapter 4, and its
meaning will be further discussed from the beginning of chapter 6 on.
5

The Alternative Island

The really philosophical writers invent the true, by analogy. . . .


Honoré de Balzac:
1. The Sociopolitics of Happiness:
More’s Utopia and its SF Context
1.1. In the first part of Thomas More’s Utopia (1516) a long
discussion of England’s social ills culminates in Hythloday’s fa­
mous passage on the destruction of the medieval peasantry:
“Your sheep,” I answered, “which are usually so tame and
so cheaply fed, begin now, according to report, to be so
greedy and wild that they devour human beings themselves
and devastate and depopulate fields, houses, and towns. . . .
there are noblemen, gentlemen, and even some abbots,
though otherwise holy men, who . . . leave no ground to be
tilled; they enclose every bit of land for pasture; they pull
down houses and destroy towns, leaving only the church to
pen the sheep in. . . . [trans. G.C. Richards, ed. Edward
Surtz]
This description, embedded in so acute an analysis of what na­
scent capitalism means to the people that Marx quoted it in Capi­
tal, is a masterpiece of indignant humanist sarcasm. The noble­
men who rage like earthquakes razing entire districts, the holy
men who are brutally indifferent to their spiritual flock and leave
churches standing only as profitable sheep-pens, the land which
is no longer communal tilling ground for a stable yeomanry but
a private enclosure for rich landlords who throw tenants out
onto the roads to beg and rob, and finally the erstwhile meek
sheep which have now7 turned into man-devouring beasts—all
this, couched in the careful verisimilitude of a traveler’s report
from exotic countries, amounts to a picture of a world upside down
being born in the shambles of the natural one. Rejecting all
partial and reformist solutions to such radical evils, the second
90
THE ALTERNATIVE ISLAND 91

part of Utopia will therefore present a radically different model


of sociopolitical life: a country that governs itself as a classless
extended family. That country—whose punning name means a
good place which is (as of now) nowhere—is an England recre­
ated in a more perfect shape. It is an island of the same size
and subdivision as England, but round instead of triangular; it
has the same natural resources, pegged to an economy based
on agriculture, but it is a just and happy country because it has
abolished private property in land and other means of pro­
duction. Instead of the monarchic pyramid in which power
flows from the top down, it is, at least in principle, a democratic
centralism that acknowledges no political elite, with a power
pyramid established from the bottom up. Where Europe slav­
ishly worships obscene war and gold, Utopia despises both;
while it sometimes has to fight wars, it uses gold for chamber
pots and slaves' fetters. It lives a distributive, egalitarian, pre in­
dustrial communism; much like tribal societies, or medieval vil­
lages, monasteries, and guilds, it is federalist and patriarchal. Its
organization is of a piece with its way of life, the best example
being the network of mutually equidistant halls where daily
meals are an occasion for pleasurable communion ¡n both physi­
cal and spiritual nourishment. Hythloday’s review of such “laws
and customs” in Utopia is a model of clarity and force fulness,
which answers the objections of his dialogue partners (including
a “More” manipulated for self-protective irony) simply by taking
to its logical end the gesture of pointing. It finds this “best state
of society” based upon the pursuit of an ultimately ethical plea­
sure attainable only in a social order with a truly collective
economy and culture. Happiness for each reached by economic
justice for all is the final goal of a possible social organization—a
startingly subversive idea.
Utopia is thus the reaffirmation of a world consonant with
human nature. This “new island” at the antipodes puts the
upside-down monstrosity of European class society back on its
feet: rhe estrangement is a dealienation. Yet a static human na­
ture working itself out in a family model—both concepts taken
from medieval Christianity—makes for a certain clogging rigidity
of relationships in Utopia, in contradiction to its fundamental
ideal of a higher Epicureanism. The Utopians possess slaves
(criminals and war prisoners), an official religion (albeit mostly
92 THE ALTERNATIVE ISLAND

deistic and tolerant of all creeds but not—unforgivably—of


atheists), and barbarous provisions against adultery. Also, the
representative democracy is tempered by a permanent rule of
the Elders, the family fathers, and of the learned. Together
with a proper subsistence-economy concern for husbanding re­
sources, this subordinates freedom to an egalitarian balance, en­
forced where necessary by stringent measures (for example,
travel restrictions). For all its dry wit, there is an air of schematic
blueprint, of groundplan «ithout adornments, about More’s pic­
ture of Utopia. But finally, it is an open-ended narrative (the
Utopians accept Greek learning and show interest in Christ’s
collectivism), the first picture of an egalitarian communism with
a relatively well-defined tolerance.
1.2. More’s Utopia subsumes all the SF forms of its epoch (and
consequently fulfills the same function as Wells does for recent
SF history). Il fuses the permanent though sometimes primitive
folk longings for a life of abundance and peace with a high-
minded intellectual constructs of perfect—that is, communist—
human relations known from antiquity on: it translates the Land
of Cockayne and the Earthly Paradise into the language of the
philosophical dialogue on the ideal state and of Renaissance
discovery-literature as reinterpreted by More’s unique blend of
medieval collectivism and Christian humanism. These forms
have been discussed in chapter 3, but a brief recapitulation will
poi111 out the specificity of More's synthesis.
Cockayne, the land of peace, sloth, and plentiful food—motifs
well known already in antiquity, and constituted into a special
country and topos in the Middle Ages—is already an inverted
world which relates to earthly human needs, and like utopia
proposes a strictly materialist solution. It can therefore be trans­
formed into utopia by relying on human intervention instead of
on a magical parallel world, and all utopias, beginning with
More, will retain their abhorrence of human degradation by war,
toil, and hunger. Next in the family of wondrous lands are the
Islands of the Blessed at the limits of the ocean. Found already
in tribal tales, Chinese and Mesopotamian legends, and Homer,
such an Elysium was originally a place of magical fertility and
contentment to which the blessed heroes were admitted in the
flesh. In the Middle Ages such locations in far-off mountains
(like the Himalayan Uttara Kuru, the echoes of which spread
the alternative island 93

from China to Europe) or seas (for example, the Celtic legendary


islands of sensual beauty) came to be conflated with the Earthly
Paradise. It is situated in this world, and before rewritings for
religious purposes its inhabitants were simply more perfect hu­
mans, endowed with happiness, health, youth, and immortality.
Echoes of such folk legends are heard in Dante’s account of
Ulysses’s final heroic voyage toward the Paradise Terrestrial, on
which he is drowned by a jealous God intent on preserving his
monopoly over the right of passage. In fact, Dante’s Comedy in­
corporates in its astrophysical and metaphysical universe almost
all the SF elements transmitted to More through the Middle
Ages, when—after Augustine of Hippo’s Civitas Dei—“the
utopia is transplanted to the sky, and called the Kingdom of
Heaven.”1 The Comedy subsumes discussions of several ideal po­
litical states, traditions of damned and blessed places, the search
for the perfect kingdom, and Dante's own superb vision of the
perfectly just City of God.
More was well aware of such subgenres as the Earthly
Paradise, but he rejected their location outside history and took
at least the first major step toward instituting in the alternative
island a historical rather than a magically arrested time. Bidding
also “a curl farewell” to the mythical conservatism of a Golden
Age of happy forefathers, he resolutely located Utopia in an
alternative but humanly attainable present, momentous exactly
because nonexistent among Europeans. As in Plato’s liepublic
(which looms large in the background of More’s work), human
destiny consists of men and their institutions; but, in direct op­
position to Plato, the just place can result from a heroic deed like
King Utopus’s cutting off the “new island” from the tainted con­
tinent. Men’s norms and institutions are not the province of reli­
gion and magic but of sociopolitics, and lime is measured in
terms of creative work. That is why Utopia differs radically from
Plato’s curious combination of caste society and ruling-caste
communism. Plato’s dialogue develops an argument for a
timelessly ideal (today quite anti-utopian) blueprint, set up in
order to escape popular, monarchic, or imperfectly oligarchic
government. More’s dialogue dramatically unfolds an actually
present state of classless self-government. More lacks all sym-

I. Lewis Mumford (Bibliography 111 A), p. 59.


94 THE ALTERNATIVE ISLAND

pa thy for both Plato's erotic communism and his caste system. /\s
for the notion that a just state depends on a community of
goods. More was much closer to the early Christian Fathers and
peasant insurgents—like John Ball—who extolled communism
than he was to Plato. Besides, this notion was so widespread in
Hellenic literature before and after Plato that Aristophanes
could mock in Ecclesiazusae (The Assemblywomen) a female attempt
at instituting egalitarian communism without money and toil,
and in The Birds a Cloudcuckooland where “everything is every­
body’s’’ and things illegal in Athens or on Olympus are deemed
beautiful and virtuous. All such references—characteristically
surviving only in fragments or rebuttals—speak of a setup
where:
... all shall be equal, and equally share
All wealth and enjoyments, nor longer endure
That one should be rich, and another be poor.
(Ecclesiazusae, 11.590-91, trans. Rogers]
Such an omnia sint communia is from that time forth the constant
principle differentiating consistent utopian literature from the
established society.
When Hythlodav is introduced to “More" he is compared to
Plato, but also secondarily to Ulysses, the hero of wondrous voy­
ages to the island of Circe, that of the Phaeacians, and so on.
The genre of imaginary voyage, as old as fiction, was the natural
vehicle of the Earthly Paradise and utopian tales, though it often
led simply to entertaining worlds whose topsy-turviness was only
playful and not also didactic. But it could also lead to just
peoples in happy lands at the limits of the world, from Hyper­
boreans to Ethiopians, from Plato’s Atlanteans to Euhemerus's
Panchaeans (and in the Middle Ages from Mandeville’s Suma­
trans to the subjects of Prester John). The most significant and
nearest in spirit to More is a fragment by lambulus (ca. 100
B.C.) about the equatorial Islands of the Sun where the usual
magically fertile nature enables men to live without private
property and state apparatus, in a loose association of com­
munities. In such joyous work as picking fruit each in turn serves
his neighbor. They practice erotic communism, eugenics, and
euthanasia (at the age of 150); the sciences, especially astronomy,
are well developed but the liberal arts are more valued as leading
THE ALTERNATIVE ISLAND 95

to spiritual perfection. Writing at the time of the great Mediter­


ranean slave and proletarian revolts, Iambulus presents a
plebeian Hellenic negation of the warring empires, the privatiza­
tion of man, and the division of labor. His happy islanders live in
the fields, under the open southern sun; and his account ot their
radical collectivism (found bv a voyager-narrator later expelled
for his harmful old habits) is the best that has even fragmentarily
survived from the host of similar tales.
Such tales were renewed by the great geographic discoveries:
Hythloday is also introduced as a participant in the voyages of
that Vespucci who had lent his name to America and set Europe
abuzz by describing the “perfect liberty’’ of the natives' tribal
communism and epicureanism. Thomas More transformed all
such strange new horizons, with their potent dissolving effects
on class society, into a systematic verbal construction of a par­
ticularized community where sociopolitical institutions, norms,
and personal relations are organized according to a more perfect
principle than that prevalent in the author's community (as 1
argue in chapter 3 that literary utopias should be defined). This
estranged place is presented as an alternative history: whoever its
author, however he twists utopian cognition, it always flows from
the hope of repressed and exploited social classes, expressing
their longing for a different but this-worldiy other world. Sud­
den whirlpools in history which both further and permit its ap­
pearance in literature—the times of lambulus, More, Fourier,
Morris, or indeed our own—have therefore the makings of
great ages of SF. For utopias, being social-science-fiction, the
sociopolitical variant of the radically different peoples and loca­
tions of SF, are the sociopolitical subgenre oj SF.
More's greatness resides thus not only in ethics or prose style.
Utopia supplied the name because it supplied the logically ines­
capable Ur-model for later literary utopias: a rounded and iso­
lated location articulated in a panoramic sxveep showing its inner
organization as a formal, ordered countersystem which is at the
same time utopia’s supreme value. The coming about of the new
order is explained by a new social contract; in More’s age, the
contract-maker is usually a founding hero, but later it will in­
creasingly be a democratic subversion—openly, as in Morris’s
socialist revolution, or transposed into cosmic analogs as tenu­
ous as Wells’s gas from a comet. Finally, though topographically
96 THE ALTERNATIVE ISLAND

closed, uiopias are presented by a dramatic strategy which counts


on the surprise effects of its presentations upon the reader: sig­
nificant utopian writings are in permanent dialogue with the
readers, they are open-ended—as in More.
2. The Dissociation of Play an e> Truth: Rabeiais to Bacon
2.0. More conveyed “full sooth in game.” François Rabelais’s
imaginative voyage through a sequence of wonderful places bois­
terously perfected such a fusion of urgent truth and witty play to
deal with the full compass of earthly preoccupations and pos­
sibilities. But by the last books of his pentalogy on rhe giants
Gargantua and Pantagruel (1532-64) the joke had become
grimmer and thinner. By the time of Campanella and Bacon, the
formal exercise of utopia had dissociated intellectual gravity
from plebeian play; in the process, “truth” itself grew increas­
ingly ideological.
2.1. Gargantua’s and Pantagruel’s sallying out of Utopia to
Paris and the ends of the world, and their insistence on the drink
and food of the body as well as of the spirit, are emblematic of
Rabelais's integration of sensual with philosophic materialism, of
folk chronicles about the deeds of enormous and valiant giants
with an uproarious intellectual critique of the sum total of con­
temporary life. Phis critique is inescapable because it reaches
from rational argument and farce to the colossal deployment of
synonyms and neologisms, idioms taken literally and fields en­
compassed encyclopedically. Language itself is no longer god-
given but a medium of human labor, enjoyment, and folly; it is
formally presented as such in the SF parable of the congealed
words in The Fourth Book. The sequence of events, too, bodies
forth a gay and dynamic process of imbibing knowledge from
the various provinces of reality passed in critical review—from
war and education in the first two books, through marriage and
sex in the third, to the wondrous and horrible islands of religion,
law, and finance in the fourth and fifth books. The basic attitude
of this work is “a gaiety of spirit” equated with the wine of the
grape as well as the wine of learning and freedom, of friendli­
ness and life itself. Such a draught is a blasphemous transubstan-
tiation in which matter becomes its own conscious and cognitive
enjoyment, substituting for service of the divine (divin) that of
the vine (du vin). The folk enjoyment in gigantism is not sepa­
THE ALTERNATIVE ISLAND 97

rated from goodness and wisdom. Rather, matter is treated as


not only the sole reality but also the supreme good, of which
there can never be too much. Rabelais’s whole work is one huge
navigation toward liberated matter and unalienated man. This
cognitive “imaginary voyage" is the exploration of a dangerous
freedom: “You must be the interpreters of your own enterprise”
is the final conclusion.
Thus “pantagruelism” is the liberation of a human quintes-
ence from the impure actuality, an unbridled creation of a new
human nature scorning contemporary unnatural Europe—as
when Pantagruel transforms the bad, aggressive king Anarch
into a good though henpecked hawker of green sauce. It oscil­
lates between sheer fantasy and simple inversion. The latter is
seen in the anti-abbey or “free university” of Theleme, set
against the old educational and monastic institutions. Formally,
this is the most clearly utopian passage in Rabelais, though it is
not his boldest creation but an elite assembly of young people
noble enough to follow the inner-directed commandment “do
what you will”. More importantly, “pantagruelizing” entails as­
similating the whole reality of that age and regurgitating it
transmuted by his laughing philosophy, just as Gargantua com­
prehended whole countries in his throat and regurgitated the
narrator who visited them. To that end, Rabelais employs with a
serene greediness all available SF traditions and all forms of
delighted estrangement—Greek satire and medieval legends,
marvelous voyages such as Navigatio Saudi Rrendani, Plato and
Villon, More and Lucian. Almost incidentally, he produced some
episodes of SF that will stand as its constant yardstick.
2.2. Rabelais adapted the episode in Gargantua’s throat and
the whole marvelous voyage in the second half of his opus from
the classical tradition subsumed in Lucian of Samosata. In True
Histories (ca. 160) Lucian laughingly settled the score with the
whole tradition of vegetative myths, from the mythological tales
themselves, through Homer’s voyages, to popular Hellenistic ad­
venture romances. His narrator’s journey to various wondrous
islands, flight to the Moon, Morning Star, and Sun, life inside a
huge whale, in Cloudcuckooland, on the Island of the Blessed,
and so forth is a string of model parodies, each translating a
whole literary form into a critical, that is, cognitive, context. The
island of vine-women is a parody of Circe’s and other islands of
98 THE ALTERNATIVE ISLAND

erotic bliss, the war of the Selenites against the Heliotes intro­
duces aliens and combats more grotesque than in any romance
or myth; but both are also models for later SF meetings and
warfare with aliens, Lucian uses the mythical scheme of journeys
based on the cycle of death and rebirth, darkness and day, clos­
ing and opening, for ironic subversion. Its spectrum ranges
from ironic events, situations, and characters, through parodic
allusions and wordplay, to direct sarcasm. For example, his
tongue-in-cheek extrapolation of colonial warfare into inter­
planetary space is rendered utterly ridiculous by a farcically
pedantic and scabrous description of the semi-human Selenites.
Lucian's whole arsenal of demystification amounts to a value sys­
tem in which vitality is equated with freedom. Being confined to
the country within the whale, with its oppressive fish-people, is
Lucian’s equivalent of an infernal descent, after his flight
through imperialist heavens. “Lucian the Blasphemer" presented
the nonexistent “quite lightly, quite easily, as if he were an inhab­
itant of the Fortunate Isles themselves.”2 His humanistic irony
embodied in aesthetic delight became the paradigm for the
whole “prehistory" of SF, from More and Rabelais to Cyrano and
Swift.
2.3. In More and Rabelais this tradition led to the “alchem­
ical” procedure of creating a new homeland by a transmutation
of the baser elements in the old country (England or the Tou-
raine), so that Rabelais’s Active narrator could call himself "ab­
stractor of the Quintessence." Actuality proved different: the
marvelous countries became colonies, More died beheaded,
Rabelais barely escaped the stake, knowledge and sense were
again viciously sundered by religious wars and monarchist abso­
lutism. In the profound crisis of the age, the first wave of the
revolutionary middle class had separated itself from the people,
and had been destroyed or absorbed by church and state. At
the beginning of the seventeenth century this was clearly spelled
out by the burning at the stake of Giordano Bruno, the heroic
philosopher who had proclaimed an infinite universe with an in­
finite number of autonomous and equivalent worlds.
The new power cast a spell even over utopographers. Shake­
speare allotted a conservative function to the wondrous place in

2. Ernst Bloch (Bibliography It), 1:507.


THE ALTERNATIVE ISLAND 99

the western seas by placing the educational island of The Tempest


under the rule of monarchist magic; by dividing its servant-aliens
into one representative airy, angelic if repining, goody and one
earthy, sexually and politically libertine, subversive, in a word
demonic, baddy; and by using it for a laboratory demonstration
of the supreme necessity for vertical political order. To round
things properly off, he also included in Gonzalo’s speech an Aris-
tophanically unfair sideswipe at the older “contrary” utopianism
of freedom: the new “commonwealth” is in a propagandistically
disingenuous way made to fuse utopian, primitivistic, and down­
right silly traits, so that it could be only too flippantly refuted.
When that speech does touch upon the central Morean con­
tradiction of a vertically ordered, patriarchal freedom, the mat­
ter is not explored but used as a sleight-of-hand substitution for
the virtuous simplicity of Montaigne's cannibals; therefore, their
happy primitive communism leaves a sour aftertaste in Shake­
speare’s reversion of Caliban. Beyond this, The Tempest is an an­
thology of elements of and views on a new locus—with the ex­
ception of any sympathetic to egalitarian community, which is
ruled a priori out of court in the travesty of the plebeians’ at­
tempt at a bestial takeover. Hierarchy, the aptly named Chain of
Being as sternly benevolent salvation from dystopian chaos,
clearly prevails. Yet the Shakespearean tension of Christian
humanism—and beyond that, of the poet in class society—
produced also, in Miranda, a naively pure glance at the (if only
potentially) “beauteous mankind" and “brave new world," and in
Prospero’s "revels” speech a melancholy adieu to even the gran­
dest verticals of human society and life as “insubstantial,” tran­
sient stuff of space and time. True, the official ideology of
Elizabethan morals and politics—indeed of politics as personal
morals—colors all the supple and masterly estrangements occur­
ring within the “sea-change” that affects in different ways all the
dramatis personae on this new island with the only too familiar
absolutist relationships. Nonetheless, and most importantly, the
other pole of the Shakespearean tension created Ariel, the
emblematic cosmic representative of the tempestuous and
metamorphic island; the yearnings for self-governing freedom,
to be repressed in a colonial island or civil society, were judged
allowable at least for a pure spirit. The rich Renaissance lyricism
hides a syllogism that already prefigures Swift’s dry Houyhn-
100 THE ALTERNATIVE ISLAND

hnms: if only intelligent beings (psychozoa) were not men but


possessed another nature, a radically different com mon-wealth
“by contraries”—or the perfect anarchism of a fusion with
nature—might be possible: and beautiful beyond our ken.
2.4. Disbelieving in a changeable human nature, utopog-
raphers had to cast about instead for better powers over it.
Among a host of less consistent attempts, the nearest to More
was Doni’s sarcastic “World of the Fools” (in I Mondi, 1552), but
even his plebeian egalitarianism succumbed to the pessimistic
and static view’ of the age. Most memorably, in southern, Cath­
olic Europe Tommaso Campanella reinstated astrology, that
fantastic pseudo-science of absolutism, as the guiding principle
of his City of the Sun (bp. 1623);3 and in northern, Protestant
Europe Francis Bacon, in New Atlantis (bp. 1627), perspicaciously
discerned a natural science acting as esoteric religion to be the
wave of the future. Campanella, though formally prolonging
Iambulus’s and More* line, describes a perfect theocracy some­
where behind India, in the seas of the old caste empires and on
an island so large it is almost a continent. The traditional utopian
abolition of private ownership, along with stimulating ideas on
dignifying labor and on education, is mystifyingly incorporated
into a monastic bureaucratism with an impersonal, militaristic
order that regulates all relations, from times for sexual inter­
course to the placement of buttons, in strict and grotesque detail
fixed by astrology. For the explosive horizontal of the Renais­
sance, Campanella substituted a dogmatic vertical that descends
from the Sun of Power to men. More’s urbane talk between
friendly humanists has in Campanella rigidified into a one-track
exposition from one top oligarch to another.
Bacon’s “great instauration,” based on the rising force of
capitalist manufacture and its technological horizons, was in the
following three centuries to prove more virulent than Campanel­
la’s monastic nostalgia. For Bacon the social system is an open
question no longer; rather, (he key for transforming the world is
a power over nature exercised by, and largely for, a politically

3. Since in this unseemly and subversive genre it is not too rare that books get
published much later than the normal few years after the known date of composition—as
in the cases of Campanella. Bacon, Kepler, Godwin, and Cyrano—such book publication
is in this work indicated by its date being preceded by ''bp."
THE ALTERNATIVE ISLAND 101

quite conservative. quasi-Christian priestly hierarchy. The orga­


nized application of technology in New Atlantis is not a break­
through to new domains of human creativity or even (except for
some agricultural and biological techniques) of natural sciences;
the only use mentioned for “stronger and more violent” engines
is in artillery, for the old destructive purposes. Conversely, sci­
ence becomes a patriarchal, genteel, and highly ceremonial reli­
gion, one that could be characterized—much as its later offshoot,
Saint-Simonism, would be—as Catholicism minus Christianity,
Scientists are a self-sufficient aristocracy of experts manipulating
or “vexing” nature and other men; as against Plato, More, and
Rabelais, their “science does not so much exude from wisdom
as wisdom exudes from science,”■* and gold is not a sign of base­
ness but of permanent abundance in possessions and power.
The very name of Bacon’s country aims to improve Plato both
by correcting his account of Old Atlantis and by presenting a
New Atlantis the old perfection of which has withstood not only
political but even geological contingencies (the narration ends
with an indication that its science can prevent earthquakes,
Hoods, comets, and similar phenomena).
The major positive claim of New Atlantis is that it delivers the
goods—abundance of things and years, and social stability—by
employing the lay miracles of science. In Bacon’s historical
epoch, even such a filling in of extant technical possibilities,
without a radical change in human relationships, constituted a
huge and euphoric program; the goal of the “research founda­
tion” of Salomon’s House is formulated as “the knowledge of
causes, and secret motions of things; and the enlarging of the
bounds of human empire, to the effecting of all things possible.”
But though this science is guarded by experts w'ho can, interest­
ingly enough, refuse to divulge dangerous discoveries, it is by its
own definition ethically indifferent: nuclear bombs and gas
ovens in concentration camps will be some of the “things" possi­
ble to effect. New Atlantis is starry-eyed over inquiring into the
“secret motions and causes” of fruits, winds, sounds, and clocks,
yet it does not think of inquiring into motions and results with
respect to the mother of the family, who is condemned to ritual

4. Howard B. White (Bibliography III C), p. 106: see also passim, especially pp.
223-25 and 171-72.
102 THE ALTERNATIVE ISLAND

seclusion, or to the population sundered from Salomon’s House.


The work thus gives a foretaste of that combination of tech­
nology and autocracy which in fact became the basis of Euro­
pean empires at home and abroad. At this point in history,
the utopian tradition fell under the sway of an upper-class ideol­
ogy which staves off human problems by technocratic extrapola­
tion, by quantitative expansion promising abundance within a
fundamentally unchanged system of social domination. Bacon's
“science” thus turns out to be as mythical as Campanella's astrol­
ogy, though more efficient. As a verbal vision, New Atlantis, with
its heavy-handed, propagandist insistence on a power hierarchy,
on opulence, and on resplendent signs of public status, and with
its stifling world which becomes interesting only when grandiose
projects are enumerated, is in fact much inferior to the fanatic
splendor of The City of the Sun. It is symptomatic of the quality of
imagination in the ensuing age that this work (one of Bacon’s
poorest) should have become the master of its thought. The
“outrageous piece of ‘miraculous evangelism’ " which founded
New Atlantis, its stuffy ceremonials and barbarous human rela­
tions, completes the picture of this “curious alliance of God,
Mammon and Science.”5
Thus the developing utopian tradition dragged into the open
the latent contradictions in More’s crypto-religious construct­
ing of Utopia. After the Rabelaisian flowering, Campanella and
Bacon mark a reaction against Renaissance libertarian human­
ism; the logical next step was the end of utopia as an indepen­
dent cognitive form. Official repression would have worked
toward this in any case; but it would not have succeeded so
swiftly and well had not the utopian camp been betrayed from
within. Having lost a fertile connection with popular long­
ings, utopia—with a few partial exceptions in the eighteenth
century—disappears from the vanguard of European culture
until Fourier and Chernyshevsky. Ironically, Bacon fought
medieval scholasticism but inaugurated a new dogmatism of
technocracy, and Campanella rotted for decades in papal prisons
but announced a return to the closed, mythic world-model of
Flato. History is cruel to “final solutions,”

5. Fir si quotation from R, W. Chambers (Bibliography III C)t p. 362; second quota­
tion from V. Dupont (Bibliography III C), p. 146,
THE ALTERNATIVE ISLAND 103

3. Monsters and Satellites: The Satirical Defence of Man


3.0. Expelled once again from official culture, significant SF
shifted from utopian seas to the planets, which were the fore­
front of attention after Copernicus and Galileo. This detour led
to Swift and reclaimed the wondrous islands for an oblique, satir­
ical defense of basically utopian values by way of a sharp critique
of the authors' anti-utopian actuality.
3.1. The "planet romance” (roman planétaire) presented a crit­
ical mirror to corrupted sublunary Earth by means of Lucianic
islands in the sea of ether, mainly by means of the Moon. “Am I
doing anything more monstrous [than Campanella, More and
Erasmus] if, in a vivid description of the monstrous habits of our
age, I transpose the scene from Earth to Moon for the sake of
caution?” asked Kepler plaintively after his Dream (bp. 1634). In
it, scientific speculation about other inhabited worlds turns into a
vivid description of Selenite biology and civilization determined
by such cosmographie factors as the search for water. Wells’s Ca-
vor was to remember this first attempt at a scientific exobiology
and cosmo-ethnology. Similarly, the hero of Francis Godwin’s
The Man in the Moon (bp. 1638) finds giant Selenites whose social
class depends on height and resistance to light, and who live in a
kind of Earthly Paradise (at the time often located on the Moon)
kept pure by deporting unsuitable babies to Earth. This sub­
genre produced in Savinien Cyrano’s two novels constituting
The Other World (bp. 1657-62) a masterpiece of embattled wit.
Cyrano’s first narrative, The States and Empires of the Moon, plays
on the opposition of “moon” and “world”:
. . , they [the Moon authorities] dressed me with great splen­
dour as a mark of shame; they made me ride on the platform
of a magnificent chariot, drawn bv four princes who were yoked
to it, and here is what they made me announce at every cross­
roads in the town:
“People, I declare unto you that this moon here is not a
moon, but a world; and that that world down there is not
a world but a moon. Such is what the Priests deem it good for
you to believe!” [trans. Geoffrey Strachan]
A supple alliance of simple inversion (the dress and chariot) and
sophisticated satire on the ideological use of language (the proc­
lamation) is here used for a burlesque debunking of such earthly
104 THE ALTERNATIVE ISLAND

authoritarianism as the Inquisition's against Galileo. The Moon


is literally an upside-down world: in it, youth—the time when
vitality is at its height—is revered instead of age, and noblemen
wear bronze reproductions of genitalia in honor of creation in­
stead of the sword, an instrument of destruction. This inclination
to make love, not war, is of a piece with universal sexual rights of
each to each and condemnation of warfare as dishonorable
whenever one side has any advantage over the other. But
Earth-bound criteria of truth are also demolished when con­
fronted with the revolutionary vision that other planets of our
Sun (and, by implication, of an infinite number of other stars)
are inhabited by conscious beings: the universe has no privileged
center. Sometimes Cyrano uses this cosmological, personal, and
political declaration of autonomy for seriocomic exaggeration, as
in the episode of the “thinking cabbage”; yet his high-minded
vision of interdependent independences lends itself to witty as­
sociation with a thoroughgoing materialist atomism in physics,
amounting to a total rejection of the religious and absolutist
world view. Even the extinction of individuality in, death is only
one phase in the omnipresent metamorphoses of creative matter,
and it is therefore met with joy instead of grief. This deep and
intimate concern for natural sciences is integrated into a se­
quence of satirical intellectual adventures, into a quest for
knowledge as freedom—poles apart from Bacon. To cite one
example, the argument of an interlocutor on the Moon that all
matter is pervaded by emptiness and has thus the freedom of
movement permitting “all things to meet in every thing” is not
only a brilliant “experimental though imaginary verification” of
philosophico-scientific value,6 it also grounds Cyrano’s liber­
tarianism in the very structure of the universe. To cite another,
the Bible is satirized when the prophet Elijah arrives into Lunar
Paradise by throwing a condensed magnet repeatedly upward
from his iron chariot; but his flight also sketches some technical
problems, such as braking during ground fall. Cyrano’s appar­
ently whimsical yet profoundly consistent and dialectical use of
innovating imagination, in both the boldness of its atheist philos­
ophy of science and cosmology- and its poetical wit, would remain
unsurpassed until the nineteenth century.

6. H, Weber (Bibliography 111 D). p. 28.


THE ALTERNATIVE ISLAND 105

As Cyrano rises from Earth he encounters beings with pro­


gressively more refined senses and therefore greater intelligence:
the “demons” from the Sun, for example, can sensually com­
prehend magnetism and tides (gravity). Conversely, when he
first encounters the quadrupedal Lunarians he believes them to
be beasts, and for them his bipedal stance is monstrous.
Monstrosity leads to rejection; rejection by the powerful leads to
expulsion or imprisonment. Cyrano's opus is built on an alterna­
tion of prisons—into which the unlucky narrator is continually
clapped by various superstitious or power-hungry authorities—
and escapes: ideologically and physically closed systems down
here ruled by priests of all stripes are transcended by fantastic
means in the direction of another world up there. The means
range from light-heartedly or irreverently burlesque to technical:
from dew, beef-marrow, and sacrificial odors (tending respec­
tively toward the Sun, the Moon, and God) to the first SF use of
multistage rockets. The flight from one world or existential situa­
tion to another is as a rule accompanied by some approximation
to death (unconsciousness, for example) and results in what
might be called debrutalization (rebirth, rejuvenation). Cyrano’s
characters are often interchangeable; like his protean matter,
they can split and recombine. The narrator himself changes roles
and situations. His situations are “metaphors realised by true
metamorphoses”,7 and his roles are manipulated with great skill
to expose all the possible nuances of false and true monstrosity,
as seen from both his side and that of the other races. This
satirical technique hits its goal—the oppressive relations prevail­
ing on Earth—both by direct invective and by sarcastic praise,
and it will be systematized to overwhelming effect by Swift. The
alternation of microscopic and telescopic vision (which will be
further developed in Gulliver’s Travels) is introduced by Cyrano
in his discussion of the two infinites of greatness and smallness:
. . . there are infinite worlds within an infinite world. Picture
the universe, therefore, as a vast organism. Within this vast
organism the stars, which are worlds, are like a further
series of vast organisms, each serving inversely as the worlds
of lesser populations such as ourselves, our horses, etc. We,

7. .Maurice Blanchot (Bibliography III D). p. 560.


106 THE ALTERNATIVE ISLAND

in our turn, are also worlds from the point of view of cer­
tain organisms incomparably smaller than ourselves, like
certain worms, lice, and mites. They are the Earths of others,
yet more imperceptible. . . . [trans. Geoffrey Strachan]
And Cyrano continues this humorous yet implicitly serious vi­
sion (it led Pascal to panic vertigo) bv ironically imagining a louse
circumnavigating the world of a human head, among the tides of
hair combed forward and backward.
Cyrano learned much from Rabelais, and the passage quoted
above stems from that of the peoples in Gargantua’s throat. But
repression weighed much more heavily on him, a member of the
small and isolated, if highly talented, circle of libertines, free­
thinkers, and burlesque poets permanently threatened by sword
and stake. In The States and Empires of the Sun his satire grows
more caustic with greater elevation from Earth, but also much
more allegorical and recondite. The Moon narrative kept the
mannered sensibility under uneas\‘ but effective control. But the
murderously unsettled times made impossible the sovereign en­
joyment signified bv Rabelais's giants; Cyrano himself died
young of a mysterious accident, probably a political murder by
clerical enemies. Even his charmingly whimsical yoking together
of elements from disparate fields (ranging from the Apocrypha
and folktales to the new philosophy understood as delight) and
his characteristic paradoxes and sallies of wit show the tenuous­
ness of his sociopolitical position between a browbeaten people
and a triumphant obscurantism. His narrative moves through
rapier flicks of ironic conceits or “points”; it is on a constant
offensive defense, in permanent acute denial. Innocently sensual
pleasure is forced to define itself as heresy, stressing what it is
against rather than—as in Rabelais—what it is for. Cyrano’s great
Epicurean tale, encompassing both sarcasm and tenderness, ac­
commodating the fantastic and the comical along with the ironi­
cally cognitive, was the culmination and swan-song of libertinism.
A monument of European mannerism and French prose, it is
also a forgotten masterpiece of SF.8

S. The thesis could be defended that only systematic repression has prevented
Cyrano's historical influence from being comparable to More's or Wells’s. What hap­
pened to his writing is representative of the fate of a whole tradition: the posthumous
THE ALTERNATIVE ISLAND 107

3.2. Jonathan Swift drew on the tradition of the imaginary


voyage—camouflaged into the newly popular form of real travel
accounts from the South Seas—to the point of making it the
basic form of Gulliver’s Travels (1726). After Lucian, More,
Rabelais, and Cyrano, the satirical-cum-utopian tradition also
had an offshoot in numerous contemporary pretended travels—
such as Foignv’s A New Discovery of Terra Incognita Australis or
Vairasse’s History of the Sevarambi, both dating from the 1670s
and of lasting fame though often officially persecuted—which
put political satire and reforms into the mouth of virtuous and
expository natives. Swift ironically manipulates many elements
found in this range of sober and tail-tale journeys, from isolated
borrowings to the matter-of-fact tone and a protagonist con-
stantly claiming travelog exactness. Into this framework he drew
the folktale wonders of giants and dwarfs, floating and magnetic
islands, monstrous or rational beasts, transforming them into
precisely observed, scientifically justified, Rationalist possible is­
lands; but simultaneously he used these islands for radical sub­
version of Rationalism, for direct and indirect ethicopolitical
satire of the English and European civilization. This makes
Gulliver’s Travels “at once science fiction and a witty parody of
science fiction.”9 Its basic concern is with the most radical an­
thropological question: What is Man? In order to suggest an
answer it destructively recapitulates the development of SF.
In Lilliput, it is the tradition of enchanted islands of human
dominion that is refuted. Where Gulliver—the average “gullible
worm” of his civilization—is physically superior, he does not
know how to control himself, his base vanity, and political snob­
bery, but grows as petty as his environment (which stands for the
English court and politics). In Brobdingnag Swift employs the

publication of The States and Dominions of the Moon in 1657 was heavily censored and
altered. An original MS was discovered only in 1861, another in 1908, and the first
critical edition published in 1910. The MS of The States and Dominions of the Sun was stolen
on his deathbed and never found: the published version is incomplete. The third part of
this trilogy. The Spark, has never been found. The first complete edition of the two novels
comprising The Other World, then, was published in 1921, the first popular edition only in
1959. In the meantime, Cyrano entered popular consciousness in Rostand’s crude
bourgeois falsification of a long-nosed Gascon sentimentalist.
9. Samuel Holt Monk. "The Pride of Lemuel Gulliver,” in Frank Brady, ed. (Bibliog­
raphy III E). p. 70.
108 THE ALTERNATIVE ISLAND

same basic device of materializing ethical qualities in order to


refute the tradition of enchanted islands of lusty and benevolent
nature. Where Gulliver tries to live up to lofty models, he is
prevented both by his prejudices and by crush ingly superior out­
side forces. Swift’s satire functions by always having it both ways:
whether Gulliver be subject or object of satire, Swift’s spokesman
or butt, the immoral European civilization is always subjected to
many-sided ridicule. As we share the narrator's terror when
faced with the colossal Brobdingnagian life-forms, we feel help­
lessly delivered into their power; as we side with the magnani­
mous Giant King (especially in the overwhelming gunpowder
discussion), we feel the preposterous, bloodthirsty vanity of this
bourgeois Everyman, representative of "the most pernicious race
of little odious vermin that Nature ever suffered to crawl upon
the surface of the Earth.” Further, the Lilliputians and Brob-
dingnagians are men viewed through the two ends of that “pock­
ets perspective" which turns up so often in Dr. Gulliver's pock­
ets. Vision too is both literal and cognitive: in the First Voyage,
Gulliver is myopic, in the second he possesses microscopic sight
and shrinks appalled from the craters in the breasts of giant
court-ladies. Swift inverts the cheerful relativism of Cyrano’s in­
finite series of mites (not only the techniques of shifting satirical
vision but also a number of situations in the Travels, including
Gulliver in his cage, owe much to Cvrano); he has us look with
disgust upon the corruption of the body politic in Lilliput, the
land of man as paltry political animal, and of the body physical
in Brobdingnag, the land of monstrous animals and dangerous
bodies. By what Gulliver discovers in them and what they reveal
in him, these terrifying islands function as a magnifying lens, a
spyglass for Xian's moral and material pettiness.
The post-Baconian descriptive precision is inscribed and yet
mocked in Gulliver’s very style. Right from the beginning it is
“peppered with citations of numbers, figures, dimensions . . .
[and] approximates an ideal of 17th Century scientists: ... ‘so
many things, almost in an equal number of words’.”10 The style
is itself a cognitive instrument turned against its Baconian orig­
inators—the middle class in alliance with the despots, the op­
timistic Rationalists, the projectors, the Royal Society scientists.

10. Robert C. Elliott (Bibliography III E>, p. 199.


the alternative island 109

Their irrationality is sarcastically revealed by reducing it to the ab­


surd in the most meticulously and pedantically rational manner.
In the Third Voyage Swift shows directly this dehumanized
science, its lifestyle and consequences. Working against life, it
sterilizes man’s relations to man and to nature. The Laputans—
the first “mad scientists” in SF—have one eye turned “up to the
zenith” of mathematical abstraction, and the other inward, as
befits a subjective Individualism. They know neither man’s place
in the world nor direct, sensual as well as cerebral, relations
between people: they have no poetry, their food and clothes
are shaped geometrically, women assiduously cuckold their hus­
bands with strangers, and their science is useless or—still
worse—useful to political tyranny (the marvelous flying island of
Laputa is used for bombing rebels, a maneuver which, however,
proves powerless against a united colonial people). To Swift,
Rationalism had—by accepting the quantified world view, the
bourgeois emphasis dn counting, weighing, and measuring, and
the resultant cosmology of mechanical balance, of bodies and
motions in a “value-free” space—betrayed the true, communal,
and value-imbued possibilities of human reason. The most comi­
cal examples in the Third Voyage, the Word Engine and the
Thing Language, are taken from the fundament of Newtonian
“natural philosophy,” mathematics, and from the social science
most intimately impinging on human consciousness, linguistics.
But the comedy is black; Swift’s critique is not simply moralistic
but also epistemological. The Newtonian model’s absolutist pre­
tension that its linguistic and mathematic formalizations have
finally revealed reality and are therefore ends unto themselves,
regardless of science’s practical incidence on human lives, is
treated by Swift as a logical tool or whore (la pula) of political
oppression. Indeed, the grotesque misery of Lagado—of a piece
with and in fact the final consequence of the capitalist mercan­
tilism directly stigmatized in book 4—shows that this science is a
road to ruin for the whole society. Such alienated knowledge is
sterile and obscene; it is symbolized by the project of extracting
food from human excrement as well as by the smooth, opaque,
and crushing “adamant” bottom of the flying island. Man as the
scientific master over (instead of partner with) nature is refuted
in these distorted islands of new knowledge-, they are a retraction of
Bacon’s New Atlantis, together with which they founded the tra­
110 THE ALTERNATIVE ISLAND

dition of modern SF as companion of modern science.


The frequent critical harpin gs on the incoherence of Swift’s
Third Voyage overlook such an ideological consistency and gra­
dation. The tour of the Academy of Lagado, for example, pro­
gresses from practical projectors through “advancers of specula­
tive learning” to “political projectors," who include experts in
taxing and finding out conspiracies. Beyond Lagado, the probing
into history and philosophy in Glubbdubdrib shows up modern
knowledge as mere fashion and modern politics as decadent
when compared to the ancient times, especially to the heroes of
freedom from Brutus to More. The dialectics of this Voyage
culminate in the nightmarish Immortals of Luggnagg: Swift’s
supreme value, human life, turns into an obscene malediction
when delivered to quantity (empty duration) without the qual­
ities of youth and health. In a sarcastic eversion of the Baconian
clerical elite assumed to constitute “a living treasury of knowl­
edge and wisdom," the reader is shown howf infinite quantity
becomes infinite disgust. The quantitative vision of nature, like
all knowledge that is not applied to the happiness of people—all
the science and politics that do not “make two ears of corn, or
two blades of grass grow upon a spot of ground where only one
grew before”—is thus a counsel of death, of the obscene death-
in-life of the Struldbrugs.
The satire of Man’s politics, his body, and his intellect in the
first three voyages leads to the great opposition of the disgust­
ing Yahoos and the rational Houyhnhnms in the Fourth Voyage.
The good old times of early Lilliput and the moderation of the
Brobdingnag King culminate in the Noble Horses; equally, the
abuse of reason for immoral ends in politics and science is offset
in their reasonable and virtuous country. But here the question
"What is Man?” becomes quite inescapable. The fashionable an­
swer for Swift’s time was “the rational animal.” Accordingly, the
European Everyman is placed between creatures who are not
just optically and ethically different humans: the Houyhnhnms
are rationality without humanity, and the Yahoos humanity
without a prideful pretense to reason. Gulliver (and his reader)
cannot in this voyage hold apart from the aliens encountered, he
is coinvoived in the confrontation of reasonable and anima!
species in Houyhnhnmland; shockingly, it is the horses who are
nobly reasonable and the Yahoo-tnen who are disgustingly igno­
THE ALTERNATIVE ISLAND 1

ble and brutal. No doubt such a confrontation was known from


Ovid through the sixteenth-century vogue of conversations with
temperate beasts that culminated in Cvrano’s quadrupeds and
Sun birds. But all those writers condemned an unnatural human
civilization; Swift takes to task human nature itself, the generic
pride of man as such. The Houyhnhnms, living without knowl­
edge of such abominations as money, government, war, laws, or
lies, are an (at rimes faintly comic) ideal within a moral fable
rather than a direct and perfect model. Indeed, being biologi­
cally different they can not be physically imitated by humans, as
Gulliver’s ludicrous attempts make plain. Yet at the same time
they possess a definite ethical superiority over a world where one
is exposed to “the sight of a lawyer, a pick-pocket, a colonel, a
fool, a lord, a gamester, a whore-monger, a physician, an evi­
dence [police informer], a suborner, an attorney, a traitor.” Gul­
liver is consequendy at the end, notwithstanding all his pitiful
Rationalist literalness, much like Plato’s escapee from the cave of
shadows: he has seen the truth but cannot communicate it to the
purblind. Overridinglv, the Noble Horses are the measure of
what the Yahoos and men—worshipping power, gold, and
excrements—are not.
Rejecting the constitution-mongering of so many middle-class
“State novels” from Harrington to Fenelon, Swift—who was in a
way the last of Renaissance humanists—looks back to More’s rad­
ical hostility against the encroaching capitalist and Individualist
civilization. But in a much more corrupted age Swift’s is an
integral monitory view which illuminates not only politics but
also science and ideology. His narrator is “More's Hythloday
[dressed up] to look like Defoe’s Robinson Crusoe,"11 whose fan­
tastic findings recoil back on everyday English life like More’s—
only more savagely. True, when he finally finds utopia, it is the
inimitable life of another species. But the fact of Gulliver’s Travels
being published and thus reinserted into social practice turns its
extreme anti-utopian despair into a critique of the anti-utopian
world which it mirros. The more passionate and precise Swift's
negation, the more clearly the necessity for new worlds of
humaneness appears before the reader. Swift was living in the

11. John Traugcu, "A Voyage to Nowhere with Thomas More and Jonathan Swift,"
in Ernes! Tuveson, ed. (Bibliography III E), p. 161.
112 THE ALTERNATIVE ISLAND

heyday of bourgeois ethics, of political arithmetics treating


people as computable economic atoms (see his Modest Proposal,
that unsurpassed masterpiece of fantastic essay as a radical pam­
phlet). At a time when capitalist empires begin to span the globe,
when “all that is holy is profaned, and man is at last compelled to
face with sober senses his real conditions of life and his relations
with his kind,”12 he defeated this totalitarianism by means of a
sarcasm just as total. He might have detested man as a species,
but parallel to such an ideology he also provided evidence for his
indictment springing from a civilization where “the rich enjoyed
the fruit of the poor man’s labour, and the latter were a
thousand to one in proportion to the former." What is seen
through both ends of Swift’s spyglass in the first two voyages, in
the distorted mirror of the third, and the inverted world of
ethicobiological absolutes of the Fourth Voyage, is our own civili­
zation, revealed as monstrous and inhuman, simultaneously
comic and pridefully bestial. The resulting horrifying comedy is
rendered in an apparendy icily emotionless style of empirical
realism which, turning the age’s basic vision against itself, gives
Swift’s lucid bitterness its quite exceptional corrosive power.
Using the parallelism of material and moral. Swift channels the
tremendous energies of idioms and metaphors to his purpose.
The rope-walking or crawling of politicians in Lilliput resurrects
a dead cliché into visual and connotative concreteness, so that the
inherent absurdity is imaginatively liberated to produce once
more an estranging shock of recognition. The ideological and
linguistic norms of European practice had glossed over, killed
these metaphors, but by an uncompromising insistence on their
plebeian sign-value Swift rediscovers their deep—political and
philosophical—truth.
Thus, if Swift is—quite literally—reactionary, his is a radically
conservative or “Tory anarchist”13 reaction against the shameless
perversions of knowledge, optimism, and dominion brought
about by Individualism. If he is the opposite of a didactic uto­
pian, he is a bitter ally of utopia. Though the reader leaves

12. Karl Marx and Friedrich Engels, Manifesto of the Communist Party, in their Selected
Works in One Volume (London, 1968), p. 38.
13. George Orwell. "Politics vs. Literature," in Denis Donoghue, ed. (Bibliography
III E). p. 354.
THE ALTERNATIVE ISLAND 113

Gulliver alienated in his stable within the larger alienation of


England, the values implicit in his travels remain. In such a
thorough review of the human condition, it is significant that
Swift’s sarcasm stopped short of the Giant King’s pacifism and
the fighting solidarity of Lindalino, of the women and illiterates
who prevented the introduction of “thing language" in
Balnibarbi—in brief, of the ethics and politics associated with
Brutus and More. One should not give to these glimpses of a
common body politic a significance which the Travels themselves
will not sustain; but historically that unscathed hope is a signpost
for subsequent SF. It will have to deal w-ith Swift’s discovery that
man’s body, this battlefield of vitality and putrefaction, is his
truth. After Gulliver’s Travels, it is impossible to believe in a
merely institutional, static utopia which does not face the nature
of man. The new Heavens and new Earth demand a new Man,
and the following age, from Blake and the Shelleys to Morris,
will explore this dialectical feedback. Swift himself remains the
great and desperate champion of an integral Man against the
terrible pressures of Individualist monstrosity. Only somebody
w’ho deeply cared about man’s potentials could have been so
outraged at his Yahoodom. By this utopian outrage, in his imag­
inary voyages and marvelous islands, Swift created the great
model for all subsequent SF. It is a wise interweaving of utopias
taking on anti-utopian functions and anti-utopias as allies of
utopianism; of satire using scientific language and technological
extrapolation as a grotesque; of adventures in SF countries, arti­
ficial satellites and aliens, immortals and monsters, all signifying
England and the gende reader. All the later protagonists of
SF, gradually piecing together their strange locales, are sons of
Gulliver, and all their more or less cognitive adventures the con­
tinuation of his Travels.
3.3. Swift (and Defoe’s Robinson Crusoe) stimulated an outburst
of fantastic voyages in England and France. The Moon had al­
ready been used for crude political and economic satire in De­
foe’s Consolidator and the pseudonvmous Voyage to Cacklogallinia.
For the first time since the Renaissance, muted utopian themes
appear on the stage, culminating in Marivaux’s plays The New
Colony (bp. as late as 1878), The Island of Slaves (1723), and The
Island of Reason (1727). The playful imagination of these pieces,
for all the compromises inherent in public harlequinades, at
114 THE ALTERNATIVE ISLAND

times touched the most sensitive nerve of the period between


Molière and Beaumarchais—the striving for class and sexual
equality. Holberg’s Niels Klim (1741) laicized the subterranean
voyage for satirical purposes. The voyages to planets culminated
in Voltaire's civilized irony of Micromégas (bp. 1752), which
brought the gigantic planetary visitors to Earth in order to
explode yet again, after Cyrano and Swift, more directly and
obviously than they did, the Rationalist notion of the great
heights mankind was supposed to have finally scaled. But the
writings of almost all the major French philosophes of the
eighteenth century—Montesquieu’s Persian Letters and Voltaire’s
Eldorado in Candide, Diderot's Supplement to Bougainville’s Voyage
(1772) and Sade’s island of Tamoé in Aline and Valcour—
incorporate some utopian passages for mental experiment, con­
trast, and satire; more than 1000 editions of such writings were
published before the French Revolution. The imaginary voyage
had become so popular that it could be parodied by "lying voy­
ages” such as Münchhausen’s. A recurrent figure tvas, from
Montaigne’s Cannibals to Rousseau’s primeval communists, the
savage whose natural nobility confounded the hypocrisy of
Christian Europeans; Denis Diderot’s Tahiti may be seen as only
the most consistent, complex, and charming indictment of the
bad European life seen against the sexually, ideologically, and
economically free island life far away. One of the most interest­
ing alliances of lower-class literature and such libertarian ideol­
ogy was Nicolas Restif's Flying Man (La Découverte australe, 1781),
in which the Rousseauist hero invents wings in order to fly away
with his upper-class beloved to some yet unspoiled Earthly Para­
dise, there to enjoy a new social and natural deal. He later flies
with her and their children to the Antipodes and settles down as
king of one of the marvelous countries (of giants, beast-men,
and the like) they visit. This naive, often crude blend of imagi­
nary voyage, technology, and utopianism, cosmic eroticism and
folk evolutionism, escapism and plebeian revolt, indicates the
vast possibilities of a popular romanticism. But its development
was unfortunately cut short by the collapse of the democratic
revolution in the nineteenth century.
6

The Shift to Anticipation: Radical Rhapsody


and Romantic Recoil

In futurity
1 prophetic see. . . .
William Blake
0.1. If SF is historically part of a submerged or plebeian “low­
er literature” expressing the yearnings of previously repressed or
at any rate nonhegemon ic social groups, it is understandable that
its major breakthroughs to the cultural surface should come
about in the periods of sudden social convulsion. Such was the
age of the bourgeois-democratic and the industrial revolutions,
incubating in western Europe from the time of More and Bacon,
breaking out at the end of the eighteenth and in the nineteenth
century. The high price of industrial revolution as a result of the
repeated failures of the political ones caused in SF too a shift
from the radical blueprints and rhapsodies of the revolutionary
Utopians in the epoch of the French Revolution to the Romantic
internalization of suffering. The inflection is visible in Blake and
Percy Bysshe Shelley, while Mary Wollstonecraft Shelley and the
US Romantics are already on the other side of this ideological
shift. The irresistible march of palaeo technic steam and iron
machinerv at the middle of the nineteenth century, along with
the concomitant growth of the proletariat, prompted SF to exam­
ine more directly the machine’s potentialities for human good
and evil. At last, at the Victorian peak of bourgeois exploitation
of man and nature, SF turned, more or less sanguinely, toward
the horizons of a new revolutionary dawn.
0.2. However, this age was not simply a major social convul­
sion comparable, say, to the Reformation. The instauration of
capitalist production as the dominant and finally all-pervasive
way of life engendered a fundamental reorientation of human
practice and imagination: a wished-for or feared future becomes
the new space of the cognitive (and increasingly of the everyday)
115
116 THE SHIFT TO ANTICIPATION

imagination, no doubt in intimate connection with and depen­


dence from the shift from the social power of land to the power
of capital (see chapter 4, 2.4.). In SF the horizon within which
the novum is developed was originally a space existing alongside
the author’s empirical environment, which is thrown into ques­
tion by the radical otherness and/or debunking parody conveyed
by die alternative location. As we saw in in chapter 5, the space
was an as yet unknown island beyond the fabulous seas; or
(probably an even earlier paradigm) a valley beyond the
mountain ranges, if not indeed a subterranean enclosure; or
finally an extrapolated planetary island in the ocean of ether
(Moon and Sun from mythological tales and Lucian to Cyrano
and Verne, other planets from the eighteenth century on). In its
Renaissance heyday, this ideal alternative was informed by the
wish that the “normal” space might, by a homeopathic magical
infection, begin metamorphosizing into a configuration of more
humane or humanized actuality, into—as More’s title said—the
best state of the commonwealth and the new island of Utopia.
The alternative space was aesthetically structured by a central
but static philosophical (rather than natural-science) cognition.
By the eighteenth century, in the increasingly activistic
dynamics of hope and fear, SF begins turning (first in the
technologically and ideologically most advanced bourgeois na­
tions, England and France) to a time into which the author’s age
might evolve. This turning, that cuts decisively across all other
national, political, and formal traditions in culture, has so far not
been adequately explained. The frequently articulated thesis that
it occurred faute de mieux, because the white patches on the map
of the globe which might validate a different microcosm were
fast disappearing, is unconvincing. Not only did many white
patches remain right up to the development of a viable aviation
(and they were abundantly used by SF as late as Wells and E.R.
Burroughs), but there was at hand the whole tradition of plane­
tary novels and subterranean descents. Clearly, the deeper rea­
sons have to do with the quantification of everyday, economically
based practice—the enthronement of commodity fetishism and
money as the universal yardstick for life values—as well as of the
“natural” sciences. For the first time, capitalist technology had
united the globe, though in a discordant unity undreamt of by
the reasonable cosmopolitanism of the Stoics or Renaissance
the shift to anticipation 117

humanists, and pregnant with the most destructive collisions of


nations and classes, The same technology had made mass social
change in one lifetime the rule rather than the exception. This
turning point in history is thus the one at which each succeeding
generation becomes itself a turning point. In SF as well, after an
interlude of revolutionary anticipation that was, from Condorcet
to Percy Shelley, focused on prophetic visions of immediately at­
tainable human possibilities and validated by a dynamic philoso­
phy of humanity, the alternative time came to be situated in an
anticipated future, and SF finally grew to be aesthetically struc­
tured by a “positive" scientific cognition. It is against this norm
that we must understand all subsequent nineteenth-century SF,
which gradually spread through more nations as capitalism,
technology, and die reacting expectations of a radically better or
at any rate different future themselves spread through the con­
tracting world, and as “world literature” in Goethe’s sense
loomed on the cultural horizon. This holds true even when the
anticipatory norm seems to be—but is in fact not wholly—
transgressed, as in Mary Shelley and Jules Verne.
1. Radical Rhapsody
[The poet] beholds the future in rhe present, and his thoughts
are the germs of the flower and the fruit of latest time.
P. B. Shelley

1,0. When Time is the ocean on whose farther shore the al­
ternative life is situated, Jerusalem can be latent in England:
I will not cease from mental fight,
Nor shall my sword sleep in my hand
Till we have built Jerusalem
In England’s green and pleasant land.
Blake’s preface to Milton fuses the stronger collective activism
and the Biblical tradition of such future horizons: “Jerusalem is
called Liberty among the children of Albion" {Jerusalem). In the
Bible, old Hebraic communism—the desert tradition of prizing
men above possessions—intermittently gave rise to expectations
of a time when everyone shall “buy wine and milk without
money and without price” (Isaiah) and when “nation shall not
life up sword against nation . . . but they shall sit every man
under his vine and under fig tree; and none shall make them
118 THE SHIFT TO ANTICIPATION

afraid” (Micah), even to “a new Heavens and a new Earth: and


the former shall not be remembered, nor come into inind”
(Isaiah). Christ’s communism of love was resolutely turned to­
ward such a millennium. Throughout the intervening centuries
heretic sects and plebeian revolts kept this longing alive. Joachim
Di Fiore announced a new age without church, state, or posses­
sions, when the flesh should again be sinless and Christ dissolved
in the community of friends. By way of the seventeen th-century
religious revolutionaries this tradition led to Blake. His age wit­
nesses a new, lay prophetic line from Babeuf and Shelley to
Marx, fusing poetry and politics and inveighing against the great
Babylon of class-state, "the merchants of the earth” and "the
kings of the earth who have committed fornication with her”
(Revelation). As of Blake’s time the future is a new existential
horizon corroding what he calls the “apparent surfaces” of the
present, etching it in as unsatisfactory. As in Virgil’s Fourth Ec­
logue, “the great succession of the ages begins anew.”
1.1. Except for some insignificant precursors, SF anticipation
began as an integral part of the French Enlightenment’s confi­
dence in cognitive and social progress. Its "drawing-room com­
munists” Mably and Morelly drew up blueprints transferring
Plato’s argument against private property from heavenly ideas
into nature’s moral laws. At the conservative end of the opposi­
tional political spectrum, Sébastien Mercier's hero, who wakes up
in Year 2440 (1770 and 1786), dwelt in the first full-fledged uto­
pian anticipation: in it, progress had led to constitutional govern­
ment, moral and technical advances (to wit: a phonograph play­
ing recorded cries of wounded is used to educate princes), and a
substitution of science for religion. The noblest expression of
such a horizon was Condorcet’s Esquisse d'un tableau historique des
progrès de l'esprit humain (Sketch ... of the Progress of Human Mind,
written in 1793), which envisaged a turning point in human
history—the advent of a new man arising out of the “limitless
perfectibility of the human faculties and the social order.” Pei-
fected institutions and scientific research would eradicate inhu­
manity, conquer nature and chance, extend human senses, and
lead in an infinite progression to an Elysium created by reason
and love for humanity. Condorcet tried to work toward such a
state within the Revolution, just as did François "Gracchus”
Babeuf, in whom culminates the century of utopian activism be­
THE SHIFT TO ANTICIPATION 19

fore Marx. Equality, claimed Babeuf, was a lie along with Liberty
and Fraternity so long as property (including education) is not
wholly equalized through gaining power for the starved against
the starvers. An association of men in a planned production and
distribution without money is the only way of “chaining destiny,"
of appeasing the “perpetual disquiet of each of us about our
tomorrows.” For a great hope was spreading among the lower
classes that the just City was only a resolute hand’s grasp away,
that—as Babeuf's fellow conspirators wrote in The Manifesto of the
Equals—“the French Revolution is merely the forerunner of
another Revolution, much greater and more solemn, that will be
the last." Even when Babeuf as well as Condorcet were perma­
nently silenced and the revolution was taken over by Napoleon,
even when anticipatory SF turned to blueprints of all-embracing
ideal systems eschewing politics, it remained wedded to the con­
cept of humanity as association. This may be seen in Blake as
well as in Saint-Simon and Fourier.
1.2. The latter two great system-builders of utopian anticipa­
tion can here be mentioned only insofar as their approaches are
found in and analogous to much SF. In a way, the whole sub­
sequent history of change within and against capitalism has oscil­
lated between Saint-Simon’s radical social engineering and
Fourier’s radical quest for harmonious happiness, which Hank
Marxism on either side. Henri de Saint-Simon anticipated that
only industry, “the industrial class" (from wage-earners to indus­
trialists), and its organizational method are pertinent in the new
age. The “monde renversé” where this "second nation" is
scorned must be righted by standing that world on its feet again.
This full reversal means in terms of temporal orientation “the
great moral, poetic, and scientific operation which will shift the
Earthly Paradise and transport it from the past into the future”
(Opinions littéraires, philosophiques et industrielles), constituting a
welfare state of increasing production and technological com­
mand of a whole globe by a united White civilization. This
“Golden Age of the human species” is to be attained by “a posi­
tive Science of Man” permitting predictive extrapolation. Saint-
Simon is the prophet of engineers and industrial productivity,
applicable equally to a regulated capitalism or an autocratic
socialism. The Suez Canal as well as Stalin, and all SF writings in
which the hero is the “ideologically neutral" engineering orga­
120 THE SHIFT TO ANTICIPATION

nizer—from Verne to Asimov or from Bellamy to the feebler,


utopian Wells—are saintsimonian.
For all his rational organizing, Saint-Simon had forsaken
eighteenth-century Rationalism by answering the great Swiftian
question “What is Man?” in terms of economic life rather than of
“nature” and “natural rights”—even if he then retreated to posit­
ing three separate human natures or psychophysiological classes
(rational, administrative, and emotive) whose representatives will
form the ruling “Council of Newton,” the college of cardinals of
his “New Christianity." Charles Fourier went much further, bas­
ing a radically humanized economy entirely upon a complex se­
ries of desires. Civilization “thwarted and falsified” them whereas
it could and should have increased gratification of all passions—
sensual, collective (desires for respect, friendship, love, and a
reconstituted family), or “serial” (desire for faction, variety, and
unity). It is a world turned inside out (monde a rebours) in which
the physician has to hope for “good fevers,” the builder for
“good fires,” and the priest for “the good dead;” where family
means adultery (and Fourier enumerates with witty glee 49 types
of cuckolds), riches mean bankruptcy, work is constraint, prop­
erty ruins the proprietor, abundance leads to unemployment,
and the machine to hunger. Against this Fourier elaborated a
method of “absolute deviation” which was to lead to a world
where both work and human relations would be a matter of
“passionate attraction." Men and their passions are not equal but
immensely varied, like notes in the harmonic scale, colors in the
spectrum, or dishes at a gastronomic banquet, and have to be
skilfully composed in a “calculus of the Destinies.” Correspond­
ing to the potential harmony of the “social movement” are series
of animal, vegetable, geometric, and cosmic relationships. There
will accordingly be 18 different creations on Earth in this
passional cosmology; ours is the first and worst, having to
traverse five horrible stages from Savagery down to Civilization
before ascending through “Guarantism” (the economicosexual
welfare state of federated productive associations or phalansteres)
to Harmony. At that point humanity will have cleansed the Earth
of sexual and economic repression, illnesses, nations, a produc­
tion sundered from consumption, and the struggle for existence;
and the Earth—itself a living being in love with another bisexual
planet—will respond by melting the polar ice, turning the oceans
THE SHIFT TO ANTICIPATION 121

into something like lemonade (all of this elaborately justified by


physics), and producing useful “anti-beasts,” such as the anti­
lion, as well as new senses for men. The blessed life of Harmony
and the succeeding 16 creations (the last one sees the end of the
globe) will turn inside out the procedures of class power; courts
and priests will be Courts of Love and priesthoods of sex. armies
will clean, plant, and reconstruct, work will become play and art,
and “abnormality” the mainspring of society. Fourier’s shattering
interplay of maniacal poetry and ironical dialectics, rooted in the
deep longings of the classes crushed by commerce and industry,
in a genuine folk imagination with its immense strengths and
foibles, will reappear in garden cities and kibbutzim, communes
and “retribalization.” In his exemplary scenes and characters—
witness Nero becoming a respected butcher in Harmony, much
like Rabelais’s King Anarch—he is himself writing warm SF. In
spite of the important modifications, this will be reproduced by
the rare but precious visions fusing relativist socio politics, erotics,
and cosmology in SF, from Blake and Shelley through Defon-
tenay and Stapledon to Le Guin and Delany.
1.3. Blake and Shelley too rejected the orthodox division of
man into body versus soul and of society into classes, as well as
the merely given “human form." Blake- championed Man’s in­
dividual and collective “imaginative body” rising as a Brobding-
nagian giant into a projected free fulfillment simultaneously
economic, sexual, and creative. The hypocritical and cruel civili­
zation of Church, Army, Palace, and Merchant, with its principle
of selfhood, brings about jealous possessiveness with regard to
children and women, shame in sexual love, and slavery to
hunger and toil. Money, the cement of this fallen society, mur­
ders the poor by stunting and the rich bv corrupting their imag­
inative needs, thus engendering sterility: the prophetic revolu­
tionary or “Reprobate" is the creative counterauthority to the
official “Elect," and his followers constitute the "Redeemed.”
Therefore, Blake sang the giant American and French Revo­
lutions in his Promethean “Orc cycle" of the 1790s—from The
French Revolution, The Marriage of Heaven and Hell, America, and
Europe to The Four Zoas—which announced the end of post­
Genesis history and the advent of a new divine Man in a realm of
freedom (a term Marx too would use). Revolution is identical
with imagination and life, and absolutely unavoidable; but if its
122 THE SHIFT TO ANTICIPATION

beginning is in politics, its end is in a joyous Joachimite


Jerusalem where the body personal and the body politic will have
been redeemed. The world as historical process is experienced
and indeed co-created by poetic vision. However, as the Ameri­
can and French experiences turned to bourgeois rule and ag­
gressive conquests, and as English repression grew virulent,
Blake’s earlier work remained unpublished and unfinished. Orc
aged into his Rationalist sky-god antagonist Urizen; Blake came
to stress timeless religious apocalypse and compensation through
art instead of the imminent passage through the Earthly
Paradise of sexuality and benevolent nature to the Eden of
creativity. His fantasies of cosmogonic history read like a gigantic
inventory of later “far-out” SF, from Stapledon and E.E. Smith
to Clarke and Van Vogt. But unlike their impoverished strain­
ings into cosmic sensations, his estranging of Newton’s world
model converts it into a richly (if confusingly) metamorphic crea­
tion. Even the most opaque pseudomythology in the later Blake
retains the estranging principle of multiple vision which sees the
unfallen world within the fallen one, and the cognitive orienta­
tion of an “Innocence [that] dwells with Wisdom.” In his last
year, in a time of bread riots, he persisted in his biblical com­
munism: “Give us the bread that is our due and right, by taking
away money or a price, or tax upon what is Common to all in thy
Kingdom.”
1.4. Percy Bysshe Shelley was separated from Blake by the
crucial impact of the French Revolution—visible in utopian liter­
ature, for example, in the vigorous political and agrarian democ­
racy of several works by Thomas Spence (Description of Spensonia,
1795)—and by an upper-class education. Both these factors set
him irrevocably against Christianity, which he identified as
tyranny; his poetry marks the gradual reorienting of the revolu­
tionary imagination toward political parable and historical vision
rather than religious myth, toward Hellenic, Gnostic, and scien­
tific rather than biblical and Miltonic traditions. From his youth
he had apparendy constructed a cosmic, scientific, and political
anticipation for himself in which chemical philosophy would syn­
thesize food as well as dot oceans with and transmute deserts into
gardens, electricity would unlock the secrets of nature, and bal­
loons ensure the abolition of slavery in Africa (all themes that
were to pass into SF through the adventure popularizations of
THE SHIFT TO ANTICIPATION 12S

Verne). His first major work, Queen Mab (1813), is an embattled


vision of humanity’s past, present, and future that draws on con­
temporary natural sciences as well as philosophes like Condorcet
and their English systematizer William Godwin for the future
ideally perfectible society. Godwin’s Political Justice—invoking
Plato, More, Mably, and Swift’s Houyhnhnms—pleaded for
property to be equalized so that men could change their charac­
ter, abandon war and the monogamous family, and finally be­
come immortal by control of mind over matter. Shelley fleshes
out such a Rationalist anarchism in his anticipation of a har­
monious Earth rejoicing in the perpetual Spring of a fertile and
gentle Nature, where “All things are recreated, and the flame /
Of consentaneous love inspires al) life.” In the notes to Queen
Mab, Shelley develops his views on labor (reducible to two hours
daily) as the sole source of wealth, as well as on the inexorable
change of the Earth's axis in “a perfect identity between the
moral and the physical improvement of the human species” and
the speeding up of the mind’s perception to vanquish time by
“an infinite number of ideas in a minute.” Such horizons, along
with the poem’s forceful attacks on the ruling political tyranny,
capitalist selfishness and corruption, church and religion, made
of it, despite legal persecution, the bible of English working-class
radicalism from the Owenites to the Chartists and beyond.
Queen Mab is the concluding chord in the great sequence of
societal and cosmic anticipations accompanying the democratic
revolutions in America and France. From Diderot and Condor­
cet to Blake and almost al) the European Romantics, two genera­
tions shared the expectation of an imminent millennium of
peace, freedom, and brotherhood:
Not in Utopia—subterranean fields—
Or some secreted island, Heaven knows where!
But in the very world, which is the world
Of all of us—the place where, in the end,
We find our happiness, or not at all!
[Wordsworth, The Prelude]
But the revulsion from the results of revolution “was terrible,”
observed Shelley in the preface to his Revolt of Islam (1818):
Thus, many of the most ardent and tender-hearted . . . have
been morally ruined by what a partial glimpse of the events
124 THE SHIFT TO ANTICIPATION

thev deplored appeared io show as the melancholy desola­


tion of all their cherished hopes. Hence gloom and misan­
thropy have become the characteristics of the age in which
we live, the solace of a disappointment that unconsciously
finds relief only in the wilful exaggeration of its own de­
spair.
The shift of SF location from space to the present or immediate
future—that is. to a radically alternative historical turning, fusing
the present with the future—was (we can now see) arrested by a
politically caused “moral ruin,” and rechanneled either back into
mythical timelessness or into the staking out of anticipation in
distant futures. These alternatives develop at that historical mo­
ment into different—twin but opposed—genres and atmos­
pheres. A fantasy more tenuous, internalized, and horrific than
the later Blake emerges as a new shudder and genre in Romantic
melodrama, tale, and narrative poem. In particular, Coleridge’s
Ancient Mariner, using scientific observations and the polar voy­
age as metaphor for the breakdown of human relationships in an
alienating society, had a profound effect on Mary Shelley and
Poe, and through them on much subsequent SF. On the other
hand, Percy Shelley is (together with Fourier) the great poetic
forerunner of the SF anticipation saved from arid Victorian
political or natural-science didacticism by also being a parabolic
analogy. In the hands of poets, whether in verse or prose, such
analogy, simultaneously collective and intimate, has cosmic pre­
tensions over and beyond sociopolitical (later also technological)
anticipation; rather than extrapolation, it is an alternative.
The Revolt of Islam itself is a not quite focused "alternative
history” about a loving pacifist-revolutionary couple who are de­
feated politically but not ruined morally because they keep faith
with their personal love as well as with the future vision of
“divine Equality.” Laon and Cythna must die in this "Winter of
the world,” but “Spring comes, though we must pass, who
made / The promise of its birth." Parallel to the satirical comedy
Swellfoot the Tyrant, a sarcastic political travesty of Oedipus Rex as
beast fable, Shelley’s culminating statement comes in Prometheus
Unbound (1820). This “lyrical drama” is a delicately tough para­
ble or dialectical allegory' in which the notions (whether lyrical
images or dramatic personae} flow into each other in iridescent
and eddying metamorphoses, aesthetically and philosophically
THE SHIFT TO ANTICIPATION 125

no less breathtakingly novel and daring than consistent. The


characters are therefore subversive, self-renewing processes
rather than fixed correspondences. Subject to this caveat, Pro­
metheus may be said to stand for the Humanity that created evil
in the shape of its oppressor Jupiter, and also for intellect and
intellectuals as champions of the oppressed. In order to escape
the fate of the French Revolution or of Blake’s Orc, he re­
nounces hate, in spite of torments by Furies, who stand for the
forces of court, church, war, commerce, and law and also for
ethical anguish and despondency: outer political and inner
psychological tenor are convertible in this multiply woven “fa­
ble.” Jupiter is thereupon toppled by Demogorgon (the subter­
ranean and plebeian titanic Necessity of nature and society, and
also subversive imagination, associated with volcanic and earth­
quake imagery), who has been contacted by Prometheus’s bride
Asia (Love or overriding human sympathy). Imagination, Love,
and Hercules (Force both as strength and as armed insurrec­
tion) liberate Prometheus and bring about a renewed peaceable
life on “Fortunate isles,” where evil and ugly masks have been
stripped off all nature and man remains
Sceptreless, free, uncircumscribed,—but man:
Equal, unclassed, tribeless, and nationless,
Exempt from awe, worship, degree; the king
Over himself; just, gentle, wise;—but man:
Passionless? no: . . .

Nor yet exempt, tho' ruling them like slaves.


From chance, and death, and mutability. . . .
[3.194-201]

In the final act even this Earthly Paradise is after “an hundred
ages” superseded by Time stopping in a full unfolding of human
psychic and cosmic potentiality. The universe too grows Prome­
thean, and the newly warmed and habitable Moon sings a paean
of loving praise to redeemed Earth in a lyrical finale of surpass­
ing power, imbued with the peculiar Shelleian “liquid splen­
dor,” often in images of vivifying electricity. The cosmic drama
ends in such a libertarian, gravityless, “uncircuinscribed’’ coun­
terpart and counterproject to Dante’s mystic rose of light and
musical harmony at the end of Paradiso.
126 THE SHIFT TO ANTICIPATION

Shelley’s expressionist lyricism, using poetic abstraction as an


“intelligible and beautiful analog}” with the most precise ap­
prehensions of mind and nature and their most sensitive histori­
cal oscillations, gives poetry the power to comprehend all knowl­
edge. Politics, cosmology, and natural sciences such as chemistry,
electricity, and astronomy are potential liberators of humanity.
They are equally based on labor and Promethean thought:
Our toil from thought all glorious forms shall cull,
To make this Earth, our home, more beautiful,
And Science, and her sister, Poesy,
Shall clothe in light the fields and cities of the free!
[The Revolt of Islam, canto 5]
And humanity cannot be made whole again (he resolutely agreed
with Mary Wollstonecraft) as long as “Woman as the bond-slave
dwells / Of man, a slave; and life is poisoned in its wells.” Loving
women are equal, if not indeed privileged, bearers of human
redemption in all of Shelley's major poems.
Parallel to such poetry of cognition, Shelley’s estrangement is
the most delicate yet vigorous personal emotion at the sight of
life enslaved, approaching it always “with a fresh wonder and an
insatiable indignation”1—a line such as "Hell is a city much like
London” (Peter Bell the Third) being quite Swiftian. Often at the
limits of the expressible, “With thoughts loo swift and strong for
one lone human breast,” his insight into scientific and political
thought as strife and sympathy between men, cosmic nature, and
time makes of Prometheus Unbound “one of the few great phil­
osophical poems in English.”1 2 In it, outwardly exploding love
overwhelms gravity, setting humanity off on its cosmic voyage,
world without end. This anti-gravity is “[in terms of space] the
pull of the void itself, in terms of time it is the future, which is
also an absolute emptiness, wailing for man to ‘invent’ it.”3 With
it culminate the tensions and resolutions of the cosmico-political
revolutionary utopianism in Shelley’s opus, strongly imbued with
political alternative, Lucretian cosmic and anthropological specu­
lation, humanized science, and indeed utopian romance from

1. H. X. Brailsford (Bibliography IV A), p. 158.


2. Car) H. Grabo (Bibliography IV A), p. 198.
3. Christopher Small (Bibliography IV A), p. 239.
THE SHIFT TO ANTICIPATION ¡27

works such as Pal lock’s Peter Wilkins (1750) and J. H. Lawrence’s


matriarchal Empire of the Nairs (1801). In their texture and struc­
ture, Shelley's significant poems “are microcosmic revolutions
which help ‘quicken’ the unborn worlds whose outlines they re­
flect and describe”4 Though the integral revolutionary and uto­
pian optimism of Shelley’s is a lone, soon-quenched blaze half­
way between the cosmic voyages of Rabelais or Cyrano and those
of the Leninist “storm and stress” of 1915-25 (Mayakovsky’s or
Krleza’s, for example),5 it is a proof that SF can be supreme
poetry: and vice versa.
2, Romantic Recoil
Forward, forward, ay and backward, downward too into the abysm. . . .
—Alfred Tennyson

2.1 Mary Shelley was the daughter of two prominent radical


wrriters, William Godwin and Mary Wollstonecraft, and wrote
Frankenstein, or The Modern Prometheus (1818) as her husband was
preparing to write Prometheus Unbound. Yet in this revealingly
flawed hybrid of horror tale and philosophical SF she indicated
with considerable force the widespread recoil from Promethean
utopianism, that “disappointment that unconsciously finds relief
only in the wilful exaggeration of its own despair” which was to
become a dominant tendency in subsequent English-language
SF. The novel's theme is twofold: the unfolding of Franken­
stein's hybris in creating artificial life is intertwined with a para­
ble on the fate of an alienated representative individual—his
Creature (called “monster” only twice, 1 think, in the text). A
series of paralogisms and contradictions emerges from the op­
position of these two themes and characters.
A comparison of Mary Shelley's stance to the radical Romanti­
cism of her husband can best identify the main contradictions.
Both Victor Frankenstein’s resolve to “pioneer a new way,
explore unknown powers, and unfold to the world the deepest
mysteries of creation” (chapter 3), and Walton's parallel resolve
to discover (for the Romantics practically a synonym of "invent”
and “create”) the unfrozen, warm geographic and magnetic pole,

4. Gerald McNiece (Bibliography IV A), p. 135.


5. See on Mavakovsky chapter 11, and on Krleza my essay “Voyage to ihe Stars and
Pannonian Mire,” Mosaic 6 (Summer 1973).
128 THE SHIFT TO ANTICIPATION

the hyperborean utopia of “a country of eternal light” and “a


land surpassing in wonders and beauty every region hitherto
discovered on the habitable globe” (Letter 1), represent favorite,
permanent, and passionately held ideals of Percy Shelley. True,
in such a philosophical romance, the discoverer-inventor’s desire
to learn “the secrets of heaven and earth . . . the outward sub­
stance of things, or the inner spirit of nature and the mysterious
soul of man” (chap. 2) might be punished by the Powers That
Be. But a suffering Prometheus would remain as unbowed as
Lucifer in Paradise Lost, romantically reinterpreted after the
image of all the subversive poets, philosophes, and scientists, all
the utopian enthusiasts in the spectrum that runs from Rous­
seau through the “philanthropic” revolutionists (notably the Il­
luminati who hailed from the same university of Ingolstadt
where Frankenstein studies and fashions his creature)6 and
through Condorcet to Byron. No doubt: the central grandiose
event of the age, the French Revolution—the awful course, con­
sequences, and lessons of which shaped both Shelleys and guided
all the other influences on them—was a burning disappointment.
But for Percy Shelley it had at the same time “created and
nourished hopes that could never die,” and his programmatic
passion became to discover the causes of and remedies to the
corruption within men and society that had led to its failure.7
Obversely, though sharing his anguish at that failure and his
belief that one major cause for it was hateful violence on both
sides, Mary Shelley was nearer to her father in stressing the
supreme necessity for civil order, and therefore the unac­
ceptability of sudden radical change and the proneness of the
lower classes and fanatic intellectuals to bloodshed. The Prome­
thean inventor was for her possible and impressive, but his in­
vention was from the outset doomed to failure. Such a “fit of en­
thusiastic madness” (chap. 24), transferred from the ideal realm
of artistic shaping and marvelous voyage of discovery to actual
philosophico-scientific intervention in everyday social life, grew
for her into a blasphemous horror tale, one related to Walton as
an awful warning not to pursue discovery by solitary imagina­

6. On the Illuminati and their importance for the Shelleys' understanding of the
French Revolution see the persuasive indications of McNiece, pp. 22-23 and 96-99.
7. McNiece, p. 41.
THE SHIFT TO ANTICIPATION 129

tion, which will inevitably sunder him from warm fellow-feeling.


As he was rendered friendless by Romantic poetry, science, ancb
utopian travel dreams, so too had Frankenstein spurned lan­
guages and politics, recapitulating in his personal history the
exclusion of human values from the “objective” post-Baconian
science. Just as Walton is ruthlessly prepared to sacrifice his crew
and his own life for the advancement of knowledge, which he
equates with dominion over nature in the name of an abstract
mankind, so had Frankenstein quite scientifically concluded that
“to examine the causes of life, we must first have recourse to
death” (chap. 4) and proudly gone about creating a quasi-human
being with the aid of a merely analytical science. Of the two
traditional Promethean pursuits of animating and shaping man,
he had succeeded in the first but failed in the second: Prome­
theus pyrphoros, the subversive thief of the “divine spark,” had
unaccountably become divorced from Prometheusplasticator, the
artist-molder of human clay. As in some horror tales of blas­
phemous alchemists and their elixir of life (Godwin’s St. Leon, for
example), the resulting creation is sterile and indeed demonically
destructive of all values. When the Promethean overreacher
finally acknowledges the rightness of Jovian power and its values,
he turns into a rightly punishable Faust. What Orwell would
expose as brainwashing, Mary Shelley shows as just expiation.
However, if Frankenstein’s Creature is sterile, it is living; if
botched, it is suffering. For Percy Shelley, electricity was vital
energy imbued with natural human sympathy, while the “cal­
culating faculty” or principle (Defence of Poetry)—the calculation
of mechanical and social power—informed both unimaginative
technology and the ideology of private profit. Mary’s Franken­
stein, on the contrary, used electricity precisely with mathematics
and charnel-house dissection, as a quantifying rather than qual­
itative tool. His theme is in the tradition of the Gothic story, in
which the universal horror and disgust at his creature would
simply prefigure its behavior and its hideous looks testify to its
corrupt essence. Yet the Creature’s pathetic story of awakening
to sentience and consciousness of his untenable position as a
subject (for whom “he” is used right from the moment of anima­
tion) provides an almost diametrically opposed point of view . His
theme is both the compositional core and the real SF novum that
lifts Frankenstein above the level of a grippingly mindless Gothic
130 THE SHIFT TO ANTICIPATION

thriller. The objective eye looking at empirical surfaces, that or­


thodox organ of things as they are, is balanced by the inward
sympathy with the Creature’s subjective feelings. Far from being
foul within, he sets out as an ideal "noble savage,” benevolent
and good, loving and yearning for love. His is not the Indi­
vidualistic quest of superior discoverer-geniuses like Franken­
stein and Walton, but a humbler and more basic search for
human solidarity and communion. His terrible disappointment
and alienation is that of the typical Romantic hero—as he himself
points out, of Goethe's Werther or the Romantic Lucifer—
wandering outcast through the icy landscape. Mary Shelley’s very
important contribution was to find an objective correlative for
her characteristic ideological oscillation between Shelleyan rebel­
lion and Godwinian protesting quietism by transferring this out­
cast status into the strategic halfway house between orthodox
theology and radical politics: into biological necessity. The Crea­
ture is caught between his vital spark of freedom and the iron
grip of scorn and persecution that arises from his racial or
species alienness.
The aspect in which the Creature is a representative of suffer­
ing Mankind oppressed by a hidden and at least indifferent (if
not evil) Creator is still astoundingly alive as well as directly on
the axis of the main, heretic SF tradition that links Swift to Wells.
We are back on the shores of Houyhnhnmland as seen by God­
win: a “sensitive and rational animal” (chap. 24), less guilty than
man, is again demystifying human history, politics, psychology,
and metaphysics, as, for example, in the Creature’s bizarre edu­
cation by proxy that recapitulates in brief the Romantic world
view:
The strange system of human society was explained to me.
... I learned that the possessions most esteemed by your
fellow-creatures were, high and unsullied descent united
with riches. A man might be respected with only one of
those advantages; but without either, he was considered . . .
as a vagabond and a slave, doomed to waste his powers for
the profits of the chosen few! And what was I? I knew that I
possessed no money, no friends, no kind of property. I was,
besides, endued with a figure hideously deformed and
loathsome; I was not even of the same nature as man. . . .
Was I then a monster . . . ? [chap. 13]
THE SHIFT TO ANTICIPATION 131

However, the More-to-Diderot tradition of "contrary’’ SF is not


only continued by the Romantics but also undergoes a metamor­
phosis at their hands. The addition of “sensitive” to the defini­
tion of man as a rational animal that (as is discussed in the
preceding chapter) dominated Swift’s whole epoch points to a
great shift across the watershed of the failed democratic and
costly industrial revolutions. Humanity is now being shown up
not only as irrational but also as cruel, in impassioned rather
than satirical accents, by a suffering and wronged creature who
wants to belong rather than by an enlightened and wondering
observer. This shift corresponds exactly to the shift from far-off
spaces to the present that should be radically transformed, from
More's or Swift’s static juxtaposition of islands and cities to the
dynamic mutual pursuit of Frankenstein and his Creature across
the extreme landscapes of lifeless cold and desolation, front be­
havioral to sentimental psychology', from universal human
nature to mutual relationships of men and women in society.
Life, the central category of the Romantics, “is opposed to being
in the same way as movement to immobility, as time to space, as
the secret wish to the visible expression.”8
This hallowed status of sentient life and its genesis was
threatened by a capitalist social practice—including ever more
prominently a use of physical sciences—that substituted “me­
chanical or two-way time for history, the dissected corpse for
the living body, dismantled units called ‘individuals’ for men-in-
groups, or in general the mechanically measurable or reproduci­
ble for the inaccessible and the complicated and the organically
whole.’’9 Among other consequences, this led to a growing
preoccupation and fascination with automata as puzzling “dou­
bles” of man. Before Mary Shelley, such a semi-alien twin had
either been treated as a wondrously ingenious toy (in the
eighteenth century ) or as an unclean demonic manifestation (in
most German Romantics). In the first case it belonged to
“naturalistic” literature, in the second to horror fantasy. The
nearest approximation to an artificial creature seen as perfect
human loveliness but later revealed to be a horrible mechanical
construct wFas provided by E.T.A. Hoffmann in The Sandman

8, Michel Foucault. The Order Things (London, 19"O), p. 278,


9. Lewis .Mumford (Bibliography III A), p. 50.
132 THE SHIFT TO ANTICIPATION

(1816). But even he oscillated between fiends and physics, and


his Olimpia is seen solely through the perceptions of a dazzled
observer. Mary Shelley’s Creature is not only undoubtedly alive
though alien, fashioned out of human material instead of the
inorganic wires of puppetry, and unmistakably this-worldly, he is
also allowed to gain our sympathy by being shown from the
inside, as a subject degradingly treated like an object. However,
because of the “exaggerated despair” which Percy Shelley accu­
rately diagnosed, it is not only human society that is monstrous
in its dealing with the Creature, but he too is “objectively” a
Monster: sentient and intelligent though inhuman, animated
creature without animating harmony, starting out like a newborn
baby yet from no woman born—the intolerable paradox, in brief,
of living and unnatural.
In a book pervaded by the pathetic if again paradoxical sym­
pathy between man and inorganic nature, both the thermo­
dynamic metaphors and the compositional metonymies meet
in the body and its psychophysiology (that is to say. again in
biology). The devastation of feeling has a correlative in the icy
landscape and isolated characters; the ever narrowing imagina­
tive vortex plunges through the three narrations from the North
Pole into the inner warmth of the Creature and his observation
of the family sentiments, at whose center lies the feminine and
Mediterranean warmth of the Safie story,10 only to rcascend back
into the killings, masculine loneliness, and final coldness of death
at the top of the world. The hidden marvelous voyage of (not
merely in) the text is a double inversion of Dante’s or Milton’s
descent in that, first, the chthonic warmth of the Earth’s center is
positive and vivifying, a deep and consoling “maternal nature”
(chap. 9), and second, the protagonists are barred and driven
away from this Earthly Paradise into the outer darkness of

10. 1 am indebted to Marc A. Rubenstein’s stimulating insights (Bibliography IV A)


into the “maternal’' metaphoric system hidden in the novel and opposed to its "masculine
creation,” even where I largely disagree with the uses he puts them to. See also James
Rieger (Bibliography IV A), pp. 79-Sl, 156, et passim, who rightly refers to Symmes,
Poe. and the Hollow Earth theory. In fact. Frankenstein is structured as a Maelstrom the
vivifying center of whose spiral can only be tantalizingly approached before the reader­
voyager is symmetrically spewed out. Jules Verne’s Journey to the Center of the Earth (see
chapter 7) is an optimistic counterproject belonging to the same morphological family,
and explicating (probably by way of Poe) a number of Frankenstein's structural implica­
tions.
THE SHIFf TO ANTICIPATION 133

mutual pursuit, misery, and commiseration, into the Coleridgean


hell of dazzling icefields; “For Mary Shelley, there are only lost
paradises.”11
Frankenstein’s relationship to his Creature remains thus un­
clear: their two themes and viewpoints contradict each other. If
there is a moral focus to this parable outside of vaguely Christian
melodramatics, then the Creature is that focus, so that the reader
cannot treat him as a Gothic Monster, merely validated by sci­
ence rather than by demonology. But conversely, ii one is to look
at this novel as SF, a whole cluster of fundamental but unresol­
ved cognitive questions appears: and centrally, why did the Crea­
ture have to be hideous or the Creation botched? Frankenstein’s
unmotivated creative haste might conceivably (though I do not
think so) be put down to Mary Shelley's technical clumsiness:
even then, why should alienness have to be equated with hide­
ousness? The tenor and the vehicle are startlingly discrepant—a
signal that strong psychic censorship is at work. Yet the vitality of
the parable shows that Marv Shelley’s personal history and im­
agination fused here with the passions and nightmares of a
whole social class—the intelligentsia in capitalism, oscillating be­
tween radical titanism and conservative recuperation. Both of
these positions are viewed in the wavering light of Mary Shelley’s
central ambiguity: the interfusion of an understanding sympathy
with a guilty horror at the subversive novelty, the radical Other.
As in Percy Shelley, “new chemistry is but old alchemy writ legi­
12 Victor Frankenstein and his
bly; both are ciphers for politics."11
startling creation are a scientific cipher for an overhasty radical
intellectual at the time of the French Revolution animating (like
the Ingolstadt Illuminati, so well known to the Shelleys) the
“[hardly adequate] materials” (chap. 4) of the broad popular
forces. The philosophy-scientist who awakens and animates these
victimized masses with “no kind of property” in the hope of a
new and glorious creation finds that persecution and injustice
exacerbate them to the point of indiscriminate slaughtering.
Such a hypothesis, in which the novel is the emblematic self-
awareness of a wavering and guilt-ridden rebellious intelligentsia
looking at the implications of the French Revolution, can solve

11. Jean de Palacio (Bibliography IV A). p. 41.


12. Rieger, p. 29.
134 THE SHIFT TO ANTICIPATION

the unexplained (and as far as I can see otherwise unexplainable)


cruces of the Creature’s unsuccessful fashioning and the univer­
sal revulsion felt for it. Mary Shelley’s reservations about the
effects of revolutionary animation went much further than her
husband’s and amounted almost to a guilty retraction. The fro­
zen whiteness, for example, which is for Percy Shelley the ele­
ment of a tyrannical Jupiter (as with Blake’s Urizen or
Nobodaddy), is in Mary returned to the scene where the over-
reachers who attempted to break through the “ideal bounds” of
natural and divine order “and pour a torrent of light into our
dark world” (chap. 4) get their pathetic but finally appropriate
deserts. The perversion of their utopian dreams results in a
gloom and misanthropy that, as Percy suggested in the preface
to The Revolt of Islam, was rooted in the moral ruin and revulsion
from the French Revolution, as a consequence of which “misery
has come home, and men appear ... as monsters thirsting for
each other’s blood” (chap. 9).
This also explains why both Frankenstein and the Creature
decline from great expectations and naive optimism to self­
devouring, mutually obsessed, and community-destroying loneli­
ness. If there is little logic of events in the plot, there is a logic of
feelings, which can alone unify the not quite compatible aspects
of the Frankenstein-Creature relationship. This relationship is,
incongruously, one of creator and creature, of two biological
aliens, and finally of a soon repentant intellectual animator and a
soon exasperated plebeian force. Frankenstein and the Creature
may also be in some ways comparable to Freud’s Ego and Id, but
they are not reducible to such a Jekyll-and-Hyde constellation.
Just as in Blake and Percy Shelley—or in The Tempest—the rela­
tionships in Frankenstein body forth a collective rather than pri­
vate psychology. The Creature is warmer and finally more intel­
ligent than his creator, like Milton’s Adam turned Satan; nor can
Freudianism explain why the lower class of Id, that plebs of the
psyche, must always be deemed destructive and lawless. Indeed,
a more revealing parallel is (with all due reservations) to be
found in the Creature’s exchanging an admiring, Miranda-like
naive and benevolent wonder at humanity for a Calibanic ram­
page of slaughter: the ideal Godwinian anarchist finds he can
become a social being only by perpetuating society’s most cruel
norms. H.G. Wells was to describe his novel of the indifferent
THE SHIFT TO ANTICIPATION 135

creator painfully fashioning monstrous creatures, The Island of


Dr. Moreau, as a theological grotesque (see chapter 9). Franken­
stein’s peculiar historical position and advantage is that on the
one hand its biology' partakes of a political as well as a theological
grotesque, while on the other its pioneering scientific horizons
proved more potent than Blake’s or even Percy Shelley’s heretic
but archaic abstractions. Biology, the Romantics' central science
in a spread that runs from Shelley’s electric fluid to Goethe’s
i/r-plant. can in its “objective” version disjoin ethical ideals, such
as compassion for the Creature, from living reality, such as that
of his crucial ugliness: "I compassioned him, and sometimes felt
a wish to console him; but when I looked upon him, when I saw
the filthy mass that moved and talked, inv heart sickened, and
my feelings were altered to those of horror and hatred” (chap.
17). Biology is thus the privileged form of pseudo-scientific
critique of revolutionary utopianism: “if Prometheus, in the
romantic tradition, is identified w’ith human revolt, is the
monster what that revolt looks like from the other side?"1314
Not that Mary Shelley w'as a Social Darwinist avant la lettre
(although once outside Percy’s magnetic field she soon reverted
to a staunch upholder of bourgeois law and order). In fact, the
paradoxes of a novel based on the principle of human sympathy
yet also guilty of a racism which betokens total failure of sym­
pathy” could find a resolution only in the peace of universal

13. .M. K. Joseph (Bibliography IV A), p. xiv.


14. A very curious feedback system between fiction and social history, that confirms
the position of Frankenstein's Creature within what one might call the Caliban Complex
of bourgeois imagination, with particular reference to England's tropical colonies and its
darker races, can be found in the fact (which I take from Me Niece, pp. 29 ff.) that Percy
Shelley read very carefully Bryan Edwards’s History of the West Indies and its lengthy
account of the savage Black and mulatto revolt in San Domingo in 1791, attributed by the
auLhor to incitement to "subversion and innovation” by visionary intellectuals and politi­
cians from Paris, and replete with strongly imagined scenes of mass murder and cruel
butchery as the inevitable result of “the monstrous folly of suddenly emancipating bar­
barous men” (both quotes from Edwards). Marv must have either read or at least known
of it. since Percy's reading is recorded in her journal for 1814-15. Reinforced by absorp­
tion into Frankenstein's Creature as seen by right-wing simplification, the same ideology
reappeared in Canning's 1824 speech in the House of Commons against freeing Black
slaves in the Antilles (which I take front Palacio, pp. 649-50). In spite of its length, the
pertinent fragment must be quoted: “In dealing with the negro. Sir, we must remember
that we are dealing with a being possessing the form and strength of a man. hut the
intellect only of a child. To turn him loose in the manhood of his physical strength, in the
maturity of his physical passions, but in the infancy of his uninstructed reason, would be
136 THE SHIFT TO ANTICIPATION

death. The Creature’s fiery self-immolation on the ice can finally


reconcile action and suffering, warmth and coldness, revolt and
consolation, and return the uncouth product of masculine crea­
tion into the womb of maternal nature, to the entropic rest of the
ultimate generic anti-utopia of Death. This final horizon will
recur in Wells’s Time Machine and, as I argue in chapter 10, a
whole wing of subsequent SF. In it, as in Mary Shelley’s revoca­
tion of the radical rhapsodies, it is possible to initiate a revolu­
tionary novum but not to curb its destructiveness. The pursuit of
life, liberty, and happiness ends in misery, bondage, and death as
the novum, in a supposedly inevitable Faustian hybris, oversteps
the familiar, “natural” bondaries of order. Mary Shelley’s faithful
transcription of this central antinomy of bourgeois practice is
much superior to any orthodox demonology, as well as to any
Panglossian optimism that (in SF, say, from Godwin to Asimov)
blandly denies the existence of such antinomic evil. But it is also
cognitively inferior to a dialectic which she herself adumbrates at
the end of the novel, when Frankenstein can acknowledge that
his Promethean creation of life has, in the best demonic tradi­
tion, boomeranged into death for all his dear ones, and yet can
still exclaim: “I have myself been blasted in these hopes, yet
another may succeed.” In this view he was an improper Pro­
metheus or bearer of the novum—a truly new one, with more
patience, love, and success, was to be presented in Percy Shelley’s
Prometheus Unbound.
Mary Shelley’s other SF novel, The Last Man (1826), is a re­
newed reversal of the perspectives in Prometheus Unbound. It first
fashions a somewhat rosewatery romance out of the political lib­
eration of Percy’s poem, and then reverses its cosmic optimism
by sending upon mankind a plague that leaves the sole survivor
finally even more isolated, but also more privatized, than Frank­
enstein or his Creature. The shift of the locale into the historical
to raise up a creature resembling the splendid fiction of a recent romance [i.e. Franken­
stein}-, the hero of which constructs a human form, with all the corporeal capabilities of
man, and with the thews and sinews of a giant; but being unable to impart to the work of
his hands a perception of right and wrong, he finds too late that he has only created a
more than mortal power of doing mischief, and himself recoils from the monster which
he has made."
For Mary Shelley’s political slide after the 1820s, see Palacio pp. 194 ff., 218 ff., 230 ff.,
et passim, and on Frankenstein's conservative aspects Christian Kreutz (Bibliography
IV A), pp. 144-52.
THE SHIFT TO ANTICIPATION 137

future (the “tale of the future" becomes six times more frequent
after 1800)15 both enlarges the loneliness of the desert island tale
to inescapable planetary proportions and translates the apoca­
lyptic or simply melodramatic fantasy tale into a black SF antici­
pation. Mary Shelley’s novel canonizes a tradition adumbrated in
several works which followed the debacle of eighteenth-century
hopes and often posited a new ice age (Cousin de Grainville’s
prose epic Le Dernier homme translated as the “romance in futur­
ity” Last Men, Byron’s poem “Darkness," and others) by impart­
ing a realistic believability to their topoi of lone landscapes and
ghosdv cities. This makes The Last Man a precursor of the SF
biophysics of alienation which extends from Poe and Flam mar-
ion to The Time Machine and beyond. But the more complex
Frankenstein remains her permanent contribution, claiming for
SF the concern for a personalized working out of overriding
sociopolitical and scientific dilemmas. It compromised with hor­
ror fantasy by treating them largely in terms of a humorless if not
hysterical biology, thus announcing the legions, of menacing
aliens and androids from Melville, Wells, and Capek on. Yet
even the inconsistent sympathy and responsibility for the Crea­
ture which are established in the novel transcend the contrived
coincidences, sensational murders, and purple patches of the
novel and indeed of most SF writing on this theme (not to speak
of Hollywood movies, which as a rule revert to one-dimensional
Gothic Monsters). The sense of urgency in Frankenstein, situated
in an exotic present, interweaves the characters’ intimate reac­
tions with their social destiny, an understanding for Promethean
science with a feeling for its human results, and marries the
exploratory SF parable with the (still somewhat shaky) tradition
of the novel. This indicated the way SF would go in meeting the
challenge of the cruel times and of Swift’s great question what
was human nature—to be answered in terms of the human body
and of social history.
2.2. However, the way proved long and thorny. A number of
scattered SF writings in Europe appeared in the second third of
the nineteenth century with the revival of utopian expectations
and Romantic dreams on both slopes of the watershed consti­
tuted by the failed 1848 revolutions. In Russia Odoevsky wrote a

)5. Mr calculation, based on data (o be found in Ian Clarke (Bibliography I).


138 THE SHIFT TO ANTICIPATION

mild anticipation, The Year 4338 (discussed in chapter 11). In


France Louis Geoifroy’s Napoleon apocryphe, 1812-1832 (1836)
and Charles Renouvier’s aptly titled Uchronie (1857 and 1876)
introduced into the novel the “alternative history’’ that was to
reach a bittersweet consummation in the twentieth century with
and after Anatole France’s Penguin Island and On the White Stone.
Emile Souvestre disguised a sermon on the immorality of
mechanical progress, which had destroyed the old pieties and
would therefore be destroyed by God, as possibly the first sys­
tematic anti-utopian anticipation in Le Monde tel qu’il sera (The
World as It Shall Be, 1845); and Etienne Cabet set a treatise ex­
pounding authoritarian collectivism and thinly disguised as fic­
tion in the fittingly regressive spatial location of Voyage en Icarie
(1840); both were only less insipid than la marline’s dream of a
petty-bourgeois European confederation translated as Prance and
England (1848). The most significant echo of warm. Fourierist
utopian enthusiasms was C.I. Defontenay’s Star (1854), which
used a revived interest in the planetary novel—marked already
by Restif’s Les Posthumes (1802)—for a vivid description, in prose
mixed with verse, of a whole solar system with different
humanoid species, their physics, politics, and ethics. A utopian
humanism and sensibility, which even supplied samples of Sta-
rian literature, vivifies Defontenay’s narration of their history,
which includes a cosmic exodus and return. This work is a lone
masterpiece not to be equaled before Stapledon and C. S. Lewis,
if not the 1960s. The publication dates of two books written in
the same period by exiled workers testify in mute eloquence to
the repressive reasons for Defontenay’s loneness and such cosmic
flights: the Chartist John Francis Bray’s A Voyage from Utopia (bp.
1957) attempted to merge Swiftian techniques with radical
socialist propaganda; and the Fourierist anarchist Joseph De-
jacque’s L’Humanisphere (partial bp. 1899, full 1971) gives a
vituperative and rhapsodic vent to his visions of sexual, religious,
and sociopolitical libertarianism in 2858. Charles Henningsen’s
voluminous romance Sixty Years Hence (1846), echoing Byron and
the Shelleys in its avenger-hero and critique of an extreme
plutocracy, though in prudence published anonymously, had a
somewhat better fate, probably because of the sentimental and
scientific melodrama it deftly fused with politics and economics.
2.3. In the United States, too, utopian writings—popular
THE SHIFT TO ANTICIPATION 139

since the secularization of (he first colonizing impulse—showed


some signs of reviving. But such attempts at utopian colonies as
Cabet’s Icaria or the Brook Farm venture failed, and the de­
tachment from—indeed hostility toward—the everyday world in­
creased among North American writers of the mid-nineteenth
century. Living in the country in which the bourgeois way of life
progressed most rapidly, these writers recoiled from its optimism
most thoroughly. Instead of treating the wondrous novelty in
terms of Prometheus, the revolutionary, they came to treat it in
terms of Faust, the overreacher who sold his soul to the Devil.
Already Goethe had adopted Faust as symbol of the permanent
dynamism borne by die bourgeois, and Mary Shelley had substi­
tuted him for the Greek Titan midway through Frankenstein. The
most prominent of the SF recoilers who followed them were
Hawthorne, Melville, and Poe. The first often used allegorical
fantasy, the second a more or less imaginary voyage, and the
third both. In some cases, admittedly marginal to the ensemble
of their work, such narratives bordered on or passed into SF.
One of the strong American literary traditions was that of the
world supplying moral symbols for the writer, and in particular
of the adventurous voyage as an inner quest. It flowed from
various updatings of Pilgrim's Progress, beginning with Joseph
Morgan’s Puritan allegory The History of the Kingdom of Basaruah
(1715), and into the Enlightenment world vision explicated in
Joel Barlow’s Columbiad (1787), a not quite felicitous precursor of
Queen Mab. This tradition approached SF in the degree in which
it adopted a consistently this-worldly novum, as in Brockden
Brown and Washington Irving (whose History of New York con­
tains a satirical SF sketch, midway between Voltaire and Wells, of
Lunarians dealing with Earthmen as Whites did with Indians).
Fenimore Cooper also wrote two crotchety and rather per­
functory novels satirizing upstart politics, the better of which,
The Monikins (1835), at times rises to bitter socioeconomic lucid­
ity. The tradition culminated in Nathaniel Hawthorne's writings
as the working out of hypotheses with a symbolically collective
rather than individualist character. In short, “there was no major
19th-century American writer of fiction, and indeed fewr of the
second rank, who did not write some SF or at least one utopian
romance. 6
16. H. Bruce Franklin, Future Perfect (Bibliography IV A), p. x.
140 THE SHIFT TO ANTICIPATION

Hawthorne usually equivocates between the natural and the


supernatural, so that the hypnotism and other controlling in­
fluences are never cognitively dominant in his major romances.
Even in the stories that hinge on the scientist-artist, the some­
what melodramatic allegory suggests that his Faustian urge is
unnatural—at worst criminal, as in “The Birthmark,” and at best
useless except for his inner satisfaction, as in “The Artist of the
Beautiful.” Only in “Rappaccini’s Daughter” (1846) is Haw­
thorne momentarily prepared to envisage an alternative world
and person on their own merits. Though Beatrice is not given
as spirited a defense as Frankenstein’s Creature, she is at least
an innocent and wronged Alien and exercizes considerable
passionate attraction (analogous to and probably as a parable for
the Fourierist ideas which Hawthorne was to renounce as sense­
less and wicked after his Brook Farm experience, itself compara­
ble to a poisoned Eden). But finally, her father’s revolutionary
countercreation is dismissed in an ending more akin to exorcism
than to SF.
2.4. On the contrarv Edgar Allan Poe took to an exemplary
extreme both the autonomy of his imaginary worlds and the
isolation of the individual who does not relate to a coherent
community but to some metaphysical principle. Poe was eco­
nomically more exposed to a consistently capitalist society that
was finding the artist unnecessary except as a leisure-time enter­
tainer for marginal social strata. History and community meant
to him merely a rapidly expanding “dollar-manufacture," a hate­
ful democracy or mob rule, so that his typical protagonist—
raising the stakes in comparison with the revolts of the first
Romantic generation—ignores almost all human interactions, not
only in politics and work but also in sex and knowledge. Science,
technology, and all knowledge have become Mephistophelean
instead of Promethean powers, fascinating but leading only to
dead-ends and destruction; “Poe confronts and represents, as
few authors before him, the alienated and alienating quality of
the technological environment.”17 Therefore he constructed a
compensatory fantasy world connecting an exacerbated inner re­
ality directly to the universe. But this fantasy is a kind of photo­
graphic negative of his environment. Feeling is dissociated from

17. David Halliburton (Bibliography IV A), p. 247.


THE SHIFT TO ANTICIPATION 141

the intelligence and will that had normally acted upon a socially
recognizable reality, and a subjective timelessness (indeed a
dream time or nightmare time) or instant apprehension of hor­
ror efface any objectively measurable or progressing duration:
personality and consciousness are here disintegrating. In the ac­
tuality “time-keeping had merged with record-keeping in the art
of communication.”18 Poe was with Mars' Shelley the first sig­
nificant figure in this tradition to make a living by writing for
periodicals (both of them even wrote stories to fit an illustration
in a yearbook or magazine, as did many authors of later SF);
accordingly, he concentrated on the obstacles to communication.
Communication is for Poe a maze of masks, hoaxes, and cryp­
tograms, typified in the manuscript put into a bottle, falsely sent
or mysteriously received, revealing truth ambiguously if at all.
Most of Poe’s tales exist within the horizons of terror, of Flight
“out of space—out of time” (“Dream-land”): they are horror­
fantasies pretending to a private supernatural reality that is in
fact based upon prescientific lore. In this light, Poe is the
originator of what is least mature in the writing commercially
[Middled as SF—an adolescent combination of hysterical sensibil­
ity and sensational violence, a dissociation of symbol from imagi­
native consistency of any (however imaginary) world, a vague
intensity of style used for creepy incantation. His protagonist is
often “the perpetual American boy-man” with a somewhat hys­
terical urge “to express himself . . . above, or away from, or be­
yond our commoner range of experience”: T.S. Eliot, acknowl­
edging his “very exceptional mind and sensibility," has even
suggested that Poe’s intellect was that “of a highly gifted young
person before puberty.”19 Though this may not be fair to Poe,
who at his best knew how to present his limitations with ironic
distancing, it accurately pinpoints the emotional age of his
imitators in the no-man’s-land of fantasy passed off as SF, from
the work of Haggard and Lovecraft to Bradbury and Beyond.
Three groups of Poe’s works have a more direct claim to atten­
tion in this overview: those marginally using some SF conven­

ts. Mumford, p. 136.


19. First quotation Edward H. Davidson (Bibliography IV A), p. 214; second quota­
tion T.S. Eliot. "From Poe to Valéry,’’ in Eric W. Carlson, ed. (Bibliography IVA), pp.
212-13.
142 THE SHIFT TO ANTICIPATION

tions, those using SF for comic comment or ideological revela­


tion, and those dealing with cosmological speculations. The first
group comprises the poem “Al Aaraaf,” the dialogues “Eiros and
Charmion” (which mentions for the first time the destruction of
Earth in a conflagration caused by a comet) and “The Power
of Words,” and the tales of oceanic descent culminating in The
Narrative of Arthur Gordon Pym (1838). Pym appropriates the
extraordinary-voyage tradition for a metaphysical (and, in the
Tsalal episode, passably racist) quest for purity in the unknown,
presents an interesting use of correspondences between the
world and the protagonist, and possibly ends with the Pole being
an entrance to the hollow^ Earth popularized in the pseudony­
mous Symzonia (1820). The second group used contemporary
popular SF interests for anticipations like balloon flights across
the Atlantic or—in the wake of George Tucker’s A Voyage to the
Moon and Richard Locke Adams’s celebrated “Moon Hoax”—
to the Moon, and suspended animation (in “The Unparallelled
Adventure of One Hans Pfaall,” “The Balloon Hoax,” "Some
W’ords With a Mummy,” “Mellonta Tauta,” “Von Kempelen and
His Discovery”); but again, it transmuted them into hoaxes and
satires of present-day certainties of progress. “The Man That
W’as Used Up” (1840) in this group is the first tale about a man
almost totally composed of artificial organs. The most substantial
among these stories, “Pfaall” (1840) and “Mellonta Tauta” (bp.
1850), are most strongly science-fictional. The interplanetary
flight prepared by an amateur inventor in his backyard, the
verisimilar flight perils and observations, and the glimpses of
grotesque yet kindred Aliens in “Pfaall” gave the cue to much
later space-travel SF. More subtly, so did the future inventions,
political satire, and barriers to understanding of the reader’s
times in “Mellonta Tauta” (as also, retrospectively, in “The Thou­
sand and Second Tale of Scheherazade”) to later time-travel
SF. The three “mesmeric tales” culminating in the scientifically
motivated horrors of “The Facts in the Case of M. Valdemar”
(bp. 1850), whether used for revelation of Poe’s cosmology or
tongue-in-cheek sensationalism, are ancillary to his fantastic
system of correspondences. This third group is subsumed under
Eureka (1848), Poe’s crowning piece of essayistic SF, which ex­
plicates the highly heretical, complex web of analogies and con­
versions by which, in Poe, life does not end with death, sen­
THE SHIFT TO ANTICIPATION IIS
lienee is not confined to organic matter, cosmogony is analogous
to individual sensibility and creativity (as in "The Power of
Words”), and the universe is God’s coded monologue. Such
mechanistic metaphysics leads finally to solipsism: whatever the
writer can imagine is as good as created, and conversely all that is
created is imagined. No wonder Poe appealed to later lonely
writers.
In fact, Poe's influence has been immense in both Anglo-
American and French SF (the latter has yet to recover from it).
Though his ideology and time-horizon tend to horror-fantasy,
the pioneering incompleteness of his work provided SF too with
a wealth of hints for fusing the rational with the symbolical, such
as his techniques of gradual domestication of the extraordinary
and of the “half-dosed eye” estrangement just glimpsing the ex­
traordinary. With Poe, the tradition of the moral quest became
urbanized, escapist, and unorthodox. His influence encompasses
on the one hand the mechanical marvels of Verne and the
dime-novels, and on the other the escapist strain in some of the
“hardest" U.S. SF, for example, Robert A. Heinlein’s time­
traveling solipsism. Both are blended in the Wellsian grotesque
tradition, from some of Wells's cumulations of believable terrors
to, say, the symbolical tales of James Blish or Damon Knight.
Poe’s notes stressing verisimilitude, analog)’, and probability for
the wondrous story made him also the first theoretician of SF.
2.5. Herman Melville’s whole opus is "a major contribution to
the literature of created societies,"20 for he took the Faustian
quest more seriously than Hawthorne and less necrophilically
than Mary Shelley and Poe. Mardi (1849), though somewhat
formless, is an iconoclastic “extraordinary voyage" among islands
of unsatisfactory mythologies, politics, and philosophies that
blends memories of Polynesia with elements from Rabelais. “The
Tartarus of Maids," a revulsion against sexual physiology
masked as a burlesque alternative, is on the margins of SF by
virtue of its sustained parallel between organic creation and
paper production (just as Frankenstein’s uncouth creation is in
some ways analogous to the novel. Frankenstein). Most interest­
ingly, in “The Bell-Tower" (1856), the "practical materialist"
merchant-mechanician protagonist "enriched through commerce
20. Franklin, p. 135; see also his stimitlaung discussion of “The Bell-Tower." ibidem.
144 THE SHIFT TO ANTICIPATION

with the Levant,” rising as a new force in a feudal society, and


raising his tower with the clock and the “state-bell", is a potent
symbol for rising capitalism and the emblematic U.S. Liberty
Bell. But his bell has been cast with an admixture of workman’s
blood, and the automaton created by him to be the bell’s ringer,
the “iron slave” who represents the servitude of Negroes and all
workers, finally slays his master. The complex—even if not al­
ways congruous—religious, sexual, and political symbolism
makes this the nearest that mid-nineteenth-century narrative
prose SF came to a Blakean approach. The American SF story
continued to be well represented into the second half of the
century, especially by some of Fitz-James O’Brien’s tales, which
culminated in the somber story of microscopic fatality and elec­
tive affinity “The Diamond Lens” (bp. 1881). But he was killed in
the Civil War, and the ensuing Gilded Age was not propitious to
sustained SF, which revived only with Bellamy.
3. And so the period that opened with universal anticipa­
tions of liberation, with Blake’s and Percy Shelley's rhapsodies,
found its central expression in the anguished immediacy of
Frankenstein’s costly failure and ended in the symbolic gloom of
representative writers from what began as liberty’s first and last
frontier but turned out to be a Liberty Bell fracturing because
it was cast with an admixture of toilers’ blood. As Wordsworth
precisely noted: “We Poets in our youth begin in gladness; / But
thereof comes despondency and madness” (“Resolution and
Independence”). These words can be seen as a characterization
of the age more than of the poets it molded, turning them from
Shelley’s unacknowledged legislating to Melville’s passionate
witnessing.
7

Liberalism Mutes the Anticipation:


The Space-Binding Machines

Bring out number, weight and measure


in a year of dearth.
William Blake

0. After experiencing the first railroad from Liverpool to


Manchester, and either thinking that the train ran in grooves or
being dazzled by the seemingly absolute stability and preor­
dained course of wheels on rails, Alfred Tennyson incorporated
this new industrial imagery into some significant contrasts in
Locksley Hall:
. . . Summer isles of Eden lying in dark-purple spheres of
sea.
There methinks would be enjoyment more than in this
march of mind,
In the steamship, in the railway, in the thoughts that shake
mankind.

Fool, again the dream, the fancy! but I know my words are
wild,

Not in vain the distance beacons. Forward, forward let us


range,
Let the great world spin for ever down the ringing grooves
of change.
[164-82]
These lines embody and explicate a very interesting intimate
debate between, on the one hand, the personal, painful, escapist,
and timeless dream of Edenic, half-Greek and half-Oriental
islands—Tennyson's recurring temptation of the Lotos Eaters—
in a Homeric wine-dark sea; and on the other hand, the public
and official beliefs of the “great world” of Victorian industrial
145
146 LIBERALISM MUTES THE ANTICIPATION

capitalism exporting not only the products of Manchester,


Pittsburgh, and the Ruhr but also the concomitant ideology of
linear liberal progress. Tennyson’s references to the “march of
mind," to the spacious ranging “forward, forward" (or “excel­
sior,” as Ixingfellow said), to the thoughts that truly shook man­
kind (we have not stopped shaking since) are a pregnant formu­
lation of the orthodox liberal optimism of progress radiating by
way of the steamship and the railway. His earlier lines, “Saw the
heavens fill with commerce, argosies of magic sails / Pilots of the
purple twilight, dropping down with the costly bales,” could have
been taken from the leading missionary of free trade, Richard
Cobden:
Commerce is the grand panacea, which like a beneficent
medical discovery, will serve to inoculate with the healthy
and saving taste for civilization all the nations of the world.
Not a bale of merchandise leaves our shores, but it bears the
seeds of intelligence and fruitful thought to the members of
some less enlightened community; not a merchant visits our
seats of manufacturing industry, but he returns to his own
country the missionary of freedom, peace and good
government—while our steam boats, that now visit every
port of Europe, and our miraculous railroads, that are the
talk of all nations, are the advertisements and vouchers for
the value of our enlightened institutions.1
Yet Tennyson’s concluding image of a linear, forward-going
progress spinning down the grooves of change is ultimately am­
biguous. Spinning is, after all, a cyclical motion, either round
and round (as a top) or to and fro (as a distaff), which always
returns to the initial situation and point. Tennyson was in all
probability, as witness the whole poem and his reference to “the
great world,” thinking here of Earth’s motion simultaneously
around its axis and “forward”—but since even this “forward” is a
seasonal motion around the Sun, the ambiguity is only shelved,
not resolved. We shall perhaps find the proper due if we re­
member that Earth’s spinning round the Sun is a measure of
time—the true space of liberal progress.

1. Richard Cobden, quoted in David Thomson, England in the Nineteenth Century


(1815-1914) (London, 1964), p. 29.
LIBERALISM MUTES THE ANTICIPATION 147

Tennyson’s lines are thus an especially compressed and apt


introduction to the convertibility of quantified space and time for the
Victorian liberal mind. However, a prolific novelist will naturally
be able to show the implications of these ambiguous and an­
tinomic historical horizons more fully. In the case of an SF
novelist, who operates by definition at the “outer limits of de­
sire,”2 these implications can be shown in a magnified and ex­
plicated form, seen in a parabolic mirror—-as happened for the
culmination of palaeotechnic liberalism in the work of Jules
Verne.

1. Communication in Quantified Space:


Verne’s roman scientifique
It would be instructive to compare two excellent reports on
Vernean studies, written 13 years apart by Mark R. Hillegas and
Marc Angenot, to see the extraordinary, qualitative jump in
Verne's reputation as a writer not only symptomatic but—for all
his drawbacks—aesthetically worthwhile as well.3 The major
names and currents of French criticism—Michel Butor and Ro­
land Barthes, structuralists and neo-Marxists, psychoanalysts and
archetype hunters—all discovered him more or (ess simultane­
ously, independendy, and with equal enthusiasm after 1960. He
was of course always known as one of the founding fathers of SF.
He created a specific early and basic variant of it, the roman
scientifique (novel of science), and gained a permanent popularity
for the genre among a mass readership, mainly but not exclu­
sively juvenile. As the overall title of his shelf-ful of novels—
Extraordinary Voyages: Known and Unknown Worlds—indicates, he
refurbished the oldest tradition of SF, that of the marvelous
voyages of tribal legends, antiquity, and the Middle Ages, for
new purposes in the age of industrial adventure. However, pre­
cisely because the French Second Empire, in its increasingly des­
perate attempts at adventure in Italy, the Crimea, and Mexico,
collapsed much sooner than its British (and in our days the US)
parallel, a sensitive French writer like Verne can be discussed in

2. Northrop Frye. Anatomy of Criticism, p. 136.


3. Mark R. Hillegas. "A Bibliography of Secondary Materials on Jules Verne," Ex­
trapolation 2 (December I960):5—16; Marc Angenot. "Jules Verne and French Literary
Criticism," Science-Fiction Studies 1 (Spring 1973}: 33-37, supplemented by his second
survey of the same title. Science-Fiction Studies 3 (March 1976): 46-49.
148 LIBERALISM MUTE5 THE ANTICIPATION

terms of a changing cognition of historical horizons for such


industrialized promenades around the map—in terms of the
paradoxically unstable yet appealing mirage that I propose to
call “utopian liberalism.”
The utopian aspect of Verne is an echo and deformation of
several strong French traditions, mainly the saintsimonian one.
Saint-Simon’s crucial place in the development of “forward­
going” horizons has been indicated in the preceding chapter.4
Rather than adopting his orientation toward “[shifting] the
Earthly Paradise from the past into the future,” however, Verne
in his exemplary microcosms developed the saintsimonian uni­
versal communication involving large human collectives that is
the obverse and complement of the quantified convertibility be­
tween time and space:
The symbols and instruments of the saintsimonians’ collec­
tive will to power will be that which physically breaks down
the barriers between the peoples . . . and permits their
quicker linking. . . . The “utopia” of physical communication
bringing about the internationalization of ideas will take a
most tangible form: ships, vehicles, locomotives.5
Indeed, in spite of saintsimonian ascendencies, it is significant
that Verne, often referred to as the prophet of future gadgetry,
did not in fact write any anticipations (except for a very few late
stories to which I shall return). His works are not extrapolations
in time but ostensibly factual, newspaper-style reports about
parallel universes or alternate time-tracks in which Professor
Lidenbrock had just a few months earlier journeyed under the
Earth, Nemo under the sea, and the Columbiad trio around the
Moon. These reports are neither a Swiftian open satirical conspir­
acy calculated to estrange the reader from his environment, nor
a Poean hoax playing upon his gullibility toward a magically
omnipotent science. Verne transferred Walter Scott’s, Dumas
pere's, and particularly Fenimore Cooper’s exotic otherwhen into

4. For Saint-Simon, see Ernst Bloch (Bibliography II), Max Beer (Bibliography III A),
and works by Ansart, Cole, Desanti, Durkheim, Engels, Leroy, Manuel, and Volgin in
Bibliography IV A. For the parallels between Saint-Simon and Verne all students arc
indebted to the pioneering hints of Kirill Andreev and the study by Jean Chesneaux
(Bibliography IV B).
5. Dominique Desanti (Bibliography IV A), p. 56.
LIBERALISM MUTES THE ANTICIPATION 149

an alternative and extraordinary but strictly natural “other­


where”—the voyage, equally as believable as but more glam­
orous than the everyday Europe or North America where it
begins and ends. Its time is exactly measured and wholly filled
by the traversing and mapping of space. ‘The whole history of
the Carboniferous period was inscribed on these walls” (chap.
20), comments the narrator of his first SF novel Journey to the
Center of the Earth (1864), which is also a descent into the depths
of geological past. Later, the subterranean travelers encounter
an immense “plain of bones” constituted by 20 centuries of ani­
mal generations: "There, on three square miles perhaps, the
whole history of animal life was gathered, scarcely written on the
too recent grounds of the inhabited world” (chap. 37). Quantified
time translated into quantified space constitutes the book of Nature,
which is decoded and claimed for knowledge by the act of mo­
tion through it that permits the reading of its hidden informa­
tion. The key for decoding, the instrument through which
human imagination seizes upon Nature, is pre-Darwinian mea­
suring and classificatory natural science: geology, geography,
astronomy, or zoology. That is why “novels of science” can be
written, but also why they do not contain any new principles or
theories. Only that can be discovered which is already known to
be there—as for example the trail to the center of the Earth or
the poles—and has now to be verified by physical proximity and
scanning, conducive to imaginative absorption by the reader.
Only the possession of a sure compass and guide permits the
basic Vernean pursuit of orientation and mapmaking, so that his
enduring fascination with the magnetic pole reactualizes Sinbad’s
magnetic island in terms of a nineteen th-century metaphor for
human cognition. Two of the major plot entanglements of the
subterranean voyage are caused by the loss of the guiding thread
(the Hans-Brook) in the granite labyrinth and by the reversal of
the compass in one of Verne's recurrent electromagnetic storms.
But all the numerous scientific dilemmas (that, for instance,
about Earth’s inner heat), red herrings, puzzles, and crypto­
grams are—as in Poe’s ratioci native tales but not as in Pym—fi­
nally clarified and solved. The voyage is in fact a spatial equiv­
alent of the process of reasoning necessary for solving the ini­
tial riddle. In that sense, Verne's SF draws its excitement from
the prestige of the mid-nineteenth-century scientific method by
150 LIBERALISM MUTES THE ANTICIPATION

which Cuvier reconstituted a mastodon from one bone, and in


turn popularizes it.
The world of Verne's early books is, accordingly, more inter­
polated. into than extrapolated from the imaginative space of
textbooks of exotic geography, zoology, mineralogy and similar,
which he quotes at great (and by now wearisome) length. Yet the
voyagers are not only verifying the plenitude and solidity of this
“positive” material universe—which is in Verne identical with
what pertains to the Earth. They are also voyaging toward one
of its privileged points—the center of the Earth, the poles, the
Moon—or on the privileged circular line of Fogg’s civilized and
Nemo’s subversive circuit. Verne's SF is in a way the triumph
of imaginative cartography, of the great measuring adventure
of mankind which succeeded in quantifying the planisphere,
the flow of time, and human relationships. All his heroes, from
Nemo through Cyrus Smith to Kaw Djer, assume at moments
of crucial conflict the characteristic Napoleonic (perhaps one
should say Byronic) stance of suneying the battlefield with
folded arms and fixed gaze.
This geometric imagination has clear limitations. Verne’s voy­
ages fill in the white spots of already sketched space. In From the
Earth to the Moon (1865) and All Around the Moon (1870) the Moon
is never reached; the same is true, with one exception, of all the
privileged cartographic points in other novels. Verne’s innova­
tions (the subterranean Mediteranean and “lava lift,” the Moon
projectile and the Nautilus, the human time-binding machine
Phileas Fogg, Robur's airplane and all-purpose vehicle, Ser-
vadac’s comet) may be stimulating technical dreams but they are
infirm scientific extrapolation, on the one hand just one step
beyond existing blueprints and on the other tending toward the
inexact or even the grossly unscientific (humans could survive
neither the lava lift nor the firing of the Columbiad). But all
these innovations are vehicles of an epic of communication for the
age of industrial liberalism. In the two Moon novels, the four
logical stages of such a new epic—the conceiving and the creat­
ing of technical means for the extraordinary locomotion, its
vicissitudes, and the delirious discussions fostered by it—are sys­
tematically orchestrated to culminate each in one strong set
scene: Barbican’s speech, the casting of the Columbiad, the space
and time point of zero gravity, and the oxygen intoxication (al­
LIBERALISM MUTES THE ANTICIPATION 151

ready used in Verne’s story “Dr. Ox” to symbolize the accelera­


tion of life by scientific progress). Though technology is used to
verify the Moon map, its main function is to induce enthusiasm
for the extraordinary locomotive adventure which vanquishes
measurable space and time. The movement at a breathless pace
is the soul, the exhilaration of covering ground the supreme
passion, and the various vehicles practically the heroes of Verne's
plots.
However, all his plots describe the neutralized trajectory of the
Moon “double novel”: they are a momentary escape from and
final return to bourgeois normality. The whirl around the globe
is the obverse of a longing for the still point of repose; even the
limited novum of the wondrous means of locomotion is de­
stroyed or otherwise repudiated at the end of each story. Verne’s
voyages and plots approximate a circle, the figure which recon­
ciles dynamics and statics, the geometrical locus of a movement
that never violates a preestablished track. Hvthloday, Gulliver,
Frankenstein, or the l ime Traveller are profoundly changed by
what they learn during their travels. For Verne, space does not
harbor a hierarchy of values but a quantified grid convertible to
quantified time through speed (see Around the World in 80 Days).
His voyagers are swept up in a movement where, as in Cartesian
analytic geometry and kinetics, only bodies, forces, and obstacles
to motion (which also serve for narrative retardation) exist. Con­
versely, his world is a sum of discrete points or objects the only
possible relationship of which is distance or a collision course.
People too have become equivalent to physical molecules and
energies, able to communicate only by movement through space
and time. As in Robinson Crusoe, Verne’s great model, his charac­
ters are constantly menaced by the doom of dehumanizing sol­
itude on their individual psychic islands, as it wrere. “Only con­
nect" could have been Verne’s slogan—for example in the
emblematic episode of Axel’s losing his way in Journey to the
Center of the Earth. In this individualistic world it is impossible to
find the unexpected, the fundamentally different—say a new
mode of life—even under the globe, on the Moon, or in time.
The estrangement of Verne’s early voyages is limited to a tran­
sient pleasure in adventure, and the cognition to adding one
technical innovation or bit of locomotive know-how' (as the Moon
projectile) to an unchanged world. His “novel of science” can be
152 LIBERALISM MUTES THE ANTICIPATION

compared to a pool after a stone has been thrown into it: there is
a ripple of excitement on the surface, the waves go to the
periphery and back to their point of origin, and everything set­
tles down as it was, with the addition of one discrete fact—the
stone at the bottom of the pool. Both the pleasure in adventure
as such and the pedagogic addition of one new bit of informa­
tion at a time are suitable for—and were aimed at—a childish or
juvenile audience of pre-teens. As an introduction to SF in an
industrial age, Verne’s best stories work very well at that boy­
scout level of a group of male friends in an exciting mapping
venture.
And yet there is more to Verne than a closed world validating
its own certainties: there is also a longing to escape from it. The
distant spaces, and especially the sea, allow his characters to man­
ifest their individualities far from the regulated dullness of
bourgeois respectability. Verne’s vivid eccentrics are individ­
ualists escaping the Individualistic metropolis. In utopian and
indeed folktale fashion he wants the privileges of industrial pro­
ductivity without the relationships of production and the political
institutions in which it came about. He accepts the tenet of the
steam, iron, and coal “palaeotechnic” age, that value lies in
movement; but instead of its orientation toward a future of
infinitely expanding Manchesters, Pittsburghs, and Ruhrs, he in­
clines toward clean electricity and movement in an ultimately
circular space, toward traveling rather than arriving. He wants
the power of marvelous machines but only for a kind of ship
with a crew of friends, or at least loyal followers, which leaves the
sooty factory and its class divisions behind—exactly as the Moon
projectile, escaping social as well as physical gravity, leaves the
explosive Columbiad on Earth. Verne's furthest venture into
such waters, where escapism blends with subversion, is 20,000
Leagues Under the Sea (1870):
The world, so to speak, began with the sea, and who knows
but that it will also end in the sea! There lies supreme tran­
quillity. The sea does not belong to tyrants. On its surface,
they can still exercise their iniquitous rights, fighting, de­
stroying one another, indulging in all other earthly horrors.
But thirty feet below its surface their power ceases, their
influence dies out, their domination disappears! Ah, Mon­
sieur, one must live—live within the ocean! Only there can
LIBERALISM MUTES THE ANTICIPATION 153

one be independent! There I acknowledge no master!


There 1 am free! [chap. 10; trans. A. Bonner]
The frequent Romantic identification of sea with freedom is not
only explicitly recalled in this outburst of Captain Nemo’s but
incarnated in him. This “Nobody,” a disdainful Byronic political
pirate, scientist, and visionary hoists a black flag with a capital N:
he is a Napoleon turned heroic Unknown Avenger from popular
literature, alone as Prometheus against the whole civilized order.
He lives sheltered and supplied by the sea, a shepherd of its
flocks and guardian of its treasures, the tutelary genius of a fully
furnished world which is simultaneously on Earth and as foreign
to terrestrial life as a different planet would be. In Nemo’s—and
in Lidenbrock's—travels Verne comes much nearer to outer­
space SF than in the two Moon novels. In both cases the exciting
and yet almost dreamlike ease of travel is allied to electricity,
which is for Verne a Shelleyan, libidinous “soul of the industrial
world" (The Clipper of the Chuds, chap. 6). Nemo and his electrical
submarine Nautilus are perfectly adapted to the sea, “mobile in
the mobile." Within it, this great anarchoindividualist ’s violent de­
sire has for the nonce produced a parallel and rival microcosm—
a potent weapon against the oppressors but also museum, gal­
lery, concert-hall, and library stocked with the spacious works of
freedom: “poetry, fiction, and science from Homer to Hugo, from
Xenophon to Michelet, from Rabelais to George Sand.” Civiliza­
tion is the Frankenstein of the Nemo/Nautilus “monster,” that
ally and avenger of the Third World, of the national liberation
movements in the wake of the French Revolution, against the
imperial powers. The strong Romantic leanings of Verne, a lover
of caverns, tempests, volcanoes, polar zones, and old castles,
come in this novel very close to a revolutionary liberalism. Within
his politically ambiguous opus, Nemo is an exceptionally sym­
pathetic and lucid achievement, an Odysseus with both superior
technology and liberating aims, redeeming the novel’s boring
ichthyological passages. Nemo the "superman” is Verne’s only
hero to plant his flag on the pole, though even he comes peril­
ously close to immobilization and asphyxiation (the contrary of
oxygen inebriation) as punishment for this encroachment on the
still point of the whirling globe, the Faustian moment of blissful
arrest. Verne’s major voyages are either directed toward a
privileged point, or they approximate a girdle around the Earth
154 LIBERALISM MUTES THE ANTICIPATION

as in The Children of Captain Grant and Around the World in 80 Days


(which became his most popular book because it presented a safe
encyclopedia of means and adventures of speeding locomotion).
Only Nemo—even if he is at the end sucked into the
Maelstrom—manages to combine this great circle with the at­
tainment of Earth's axis and navel, for which the heroes ofJour­
ney to the Center of the Earth and The Adventures of Captain Hat­
teras—like those of Frankenstein—had striven in vain.
Nemo’s rich character also combines the traveler, rescuer, sci­
entist, and explorer monomaniacs of Verne’s above four novels.
In each of Verne’s tales, his protagonist is a passionate incarna­
tion of the theme. Lidenbrock. the energetic German professor
of transcendent crystallography, is a Hoffmannesque incarnation
of geology; Barbican, Nicholl, and Ardan are embodiments, we
are told, of Science, Industry, and Art, but (one would have to
add) also of projectile-making, iron plate, and aerial voyaging in
high spirits, and of Yankee, Scotsman, and Frenchman; Fogg of
Anglo-Saxon coolness and chronometric precision in traversing
time and space. Verne’s “humours” characterization and his al­
ternation of thrill and exposition is straight out of the operas or
boulevard vaudevilles of Second Empire Paris, and so is his scen­
ery, be it the outdoors of electric tempests or the indoors of the
upholstered Moon projectile or the Nautilus. The trio of main
characters, Verne’s “three musketeers,” usually takes the roles of
resolute explorer, loyal companion, and more or less comic ser­
vant (Lidenbrock, Axel, and Hans; Nemo, the not wholly loyal
Arronax, and Conseil with Land for added tension; Barbican,
Ardan, who becomes a second protagonist, and Nicholl usurping
the place of Marston). But as important as any persons are the
machine-vehicles that often steal the limelight as objectivizations
of the theme and of its protagonist. The protagonist and the SF
concept (the machine-vehicle or some other islandlike micro­
cosm) are therefore linked by the strongest secret sympathies
within a cluster of correspondences at the center of which is the
story’s theme or “element.”6 The Moon projectile, the elevation
out of gravity, and the ejected trio; Nautilus, the sea, and Nemo;
Fogg, chronometry, and the spectrum of means for swift
locomotion; Robur, the air, and The Albatross- Schulze, the as­

6. Michel Butor (Bibliography IV’ B). pp. 48 ff.


liberalism mutes the anticipation 155

phyxiating cold, and the supercannon factory—they all form


homogeneous symbolic systems. Around them are distributed
the supernumeraries: crew, wayside acquaintances, dastardly
enemies (usually dark-skinned, but this too is reversed in Nemo’s
system). Finally, there is the "sublime father," the representative
of providence (Nemo in The Mysterious Island, Antekirtt in
Mathias Sandorf ).
Table I
Theme, "Element,’’ SF Concept = Micro-
Book or Semantic Field cosmic Novum Protagonist
Journey Geolog}' Underground world Lidenbrock
to Center
of the Earth
Moon Novels Sublation of gravity Moon projectile Trio
20,000 Leagues Sea — freedom Nautilus Nemo
Around the Chronometry Ensemble of palaeo­ F°gg
World in 80 technic means of
Days locomotion
Mysterious Colonization of Emblematic island Cyrus Smith
Island nature
500 Million Asphyxiating cold Superfactory for Schulze
of the Begum supercannons
Clipper of Flight as scientific Airship Robur
the Clouds power
With unimportant exceptions, the cast is an all-male one. The
whole libido in Verne’s ultimately sterile world is invested in
machines instead of in women. The phallic connotations of
Nautilus or the Columbiad ejection are unmistakable, but so is
their fruitlessness. Finally, in The Carpathian Castle (1892), the
opticoelectrical machine resurrects or replaces the image of the
woman. The lively machine integrates man into space, it allows
him to be in harmony with nature, to move his individual micro­
cosm through it and closer to other people, and thus to com­
municate with them. These “clean” machines or mechanisms “do
not produce surplus value"1 nor consume human labor, since they
tap the miraculous electricity. Along with women, the working

7. Chesneaux, p. 43 (his italics).


156 LIBERALISM MUTES THE ANTICIPATION

class is also absent from Verne’s world of boyish innocence. Dis­


coverers and dastards face each other like noble and ignoble
savages, translated from Cooper’s forests into a space validated
by science.
A clean technology, worldwide communications, science and
art ruling the world—the whole outlook represented by Barbican
plus Ardan or by Nemo has a strong kinship with saintsimonist
utopianism. In his most optimistic parable, The Mysterious Island
(1875), Verne presents the rise of a fraternal community fertiliz­
ing nature by applying scientific knowledge. The voyage has
been reduced to a rudimentary framework, the favored micro­
cosm is an island instead of a ship (but the two are in Verne, as in
SF generally, largely interchangeable, both being analogs to the
author’s world), the SF concept is resolutely sociological as well as
technological, and the all-embracing theme is man’s scientific
colonization of Nature, including but superseding simple com­
munication. This novel is Verne’s “recapitulation of race his­
tory”8 and a culmination of his opus. Its mysterious island is
privileged because, like Crusoe’s, it is a figure of our globe, to
the point of possessing mutually incompatible geographic zones.
Cyrus Smith, Verne's supreme “knower” and source of energy, is
that saintsimonian ideal: an engineer, communicator, and or­
ganizer who is “man of action at the same time as man of
thought” (part 1, chap. 1), in fact “a microcosm, a composite of
all human knowledge and intelligence!” (part 1, chap. 9). He is
flanked by two loyal seconds, a hunter-cum-artist and a teenage
heir gifted for recognizing the classes of Nature, and by two
comic senants, one white and one black. To complete this verti­
cal chain of being, there is also a dog and a tamed orangoutan
(loyal companion and comic servant in the animal realm); and at
its demonic and angelic ends a repented criminal and the dying
Nemo are eventually discovered. This island crew starts out with
its know-how, two watches (which Cyrus uses for the Promethean
gift of fire by means of their curved glasses, and for surveying
space), a single sliver of steel, and a single grain of wheat. It
progresses through gathering and hunting to pottery, metal­
lurgy, and a series of increasingly sophisticated tools and tech­
niques. With the help of a hidden and providential Nemo, and in
8. Kenneth Allott (Bibliography IV B), p. 77.
LIBERALISM MUTES THE ANTICIPATION 157

a cooperative, though strictly graded, exploitation of Nature,


they increase and multiply their possessions, eventually attaining
a wholly cultivated island home, truly civilized since it can boast
roads, bridges, a lift, boat, and electric telegraph line.
But this program of—as the saintsimonists said—all by steam
and electricity, “[substituting] for the exploitation of man by man
. . . the harmonious action of man on nature,”9 has problematic
blind spots and ambiguities for a parable on history. First ol all,
it is a colonization unhampered by aborigines, which allows it to
progress, in Vernean ease, as a cross between a holiday and a
utopian colony. Yet the colonists’ ambition (like Ardan’s more
lighthearted reason for traveling to the Moon) is to valorize and
then annex rhe island to the USA. The building of the Colum-
biad had proceeded on territory recently “cleared” of Indians in
the Seminole wars; equally, after the volcanic earthquake which
destroys the island, Cyrus Smith (the Yankee Everyman as impe­
rial conqueror of space) and his companions use Nemo’s saved
treasure to found a vague utopian colony in Iowa (equally
cleared of the Sioux). Sympathy with enslaved peoples is, in
Verne as in Saint-Simon, limited to Whiles—from the Quebecois
and the Irish to the Greeks; the colored races are either blood­
thirsty beasts or natural inferiors, so that Nab in The Mysterious
Island is even given an instinctive affinity to the orangoutan!
Nemo, Verne's sympathetic rebel, is in this novel retracted: he
dies alone and mellow'ed, and Smith’s judgment on him is that
for all his heroic qualities he was wrong in “fighting the neces­
sary progress” (part 3, chap. 16). Second, this is a human history
without the lower class: of the two manual workers in the novel,
the White is a seaman and the Black devotion personified.
Third—even more strictly than usual—not a single woman ap­
pears in the novel. This history has no future, and Verne had to
employ a whole series of somewhat weary plot tricks to destroy
the colony without destroying the sense of the colonists’ work.
For this thematic culmination is also the point at which Verne's
enthusiasm flags and his writing starts to slip badly. The adven­
turous Second Empire had ended in ignominious defeat; the
ensuing Third Republic had begun with the bloody suppression

9, The Doctrine of Seinl-Simon (New York, 1972), p- 29; the quote is considered to be by
B* P. Enfantin*
158 LIBERALISM MLTES THE ANTICIPATION

of the Paris Commune and continued in a welter of corrupt


factions; free competition in the major bourgeois countries was
giving way to trusts and monopolies; the boom in colonial an­
nexations dividing the globe among imperial powers was on: the
precariousness of liberal enthusiasm was becoming quite man­
ifest. At this very time finance capital was fast ascending to
power at the expense of Saint-Simon’s privileged industry and
industrial capital, and was inaugurating the full panoply of im­
perialism as “the highest stage of capitalism.”10 With startlingly
close parallelism, Verne’s horizon grows more and more gloomy
after the mid-1870s, and his marvelous inventions—both vehicles
and communities—more and more malignant and destructive,
prisons instead of harmonies. The menacing potential of sci­
ence—seen already in Dr. Ox, the mutilated artillerists of the
Gun Club, or Nemo—is no longer neutralized by, respectively,
farce, peaceful international cooperation in exploring the uni­
verse, or political justice. Instead, the irascible but selfless eccen­
trics and explorers change into power-mad inventors or—
historically more farsighted—into willing scientific tools of mad
militarists. The petty feuds of Barbican and Nicholl or Florida
and Texas, once soothed by the dreams of communication and
peaceful colonization, explode into nightmares of world domin­
ion and a total war of faction against faction, each against each.
In The Begum’s Fortune (1879) the asphyxiating Teutonic hell of
Steel-City with its ballistic MIRVs is still vaguely balanced by a
roseate, hygienic France-Ville. But the formerly exemplary
America of freedom and progress is now' seen as a plutocratic
microcosm tearing itself apart (Milliard-City in Propeller Island,
1895). Equally, Robur, the conqueror of the air, instead of
founding an “aerial Icaria” has to recognize that “science should
not overtake the morals” in a civilization of selfish and opposed
interests (The Clipper of the Clouds, 1886); and when he reappears
in the still shriller and feebler Master of the World (1904) he has
changed from “the science of the future” to a madman, w'hose
Promethean vehicle for all elements is—as all other novelties in
the later Verne—destroyed by Providence. Verne’s racism nar­
rows to French chauvinism, and his confident alliance of science
with commerce and finance changes to a condemnation of the

10. See V. I. Lenin’s 1917 booklet of that optimistic title.


LIBERALISM MUTES THE ANTICIPATION 159

sterility of gold and money. Both fuse in the cosmic flight of


Hector Servadac, in which Anglophobia and anti-Semitism have to
figure as substitutes for the euphoria of the Moon voyage. The
fraternal exploitation of nature by men has turned to a discord
which amounts to the end of Verne’s world. In three interesting
posthumous works, The Survivors of the “Jonathan” (bp. 1909), The
Eternal Adam (bp. 1910), and The Barsac Mission (bp. 1919), he
retracts the saintsimonian optimism of The Mysterious Island. The
Survivors of the “Jonathan” (translated in two books as The Master­
less Man and The Unwilling Dictator) is Verne’s most explicit politi­
cal parable, and it is situated symbolically on the furthest shore
of the world, an island halfway between the lost freedom and
progress of the Americas and the privileged but dangerous point
of the South Pole. There, a cargo of suffering and ill-adjusted
humanity from a wrecked ship all officers of which have been
swept overboard is subjected to the equally pernicious entice­
ments of state socialism, egalitarian communism (both professed
by failed intellectuals), and usurious capitalism. The childlike
workers would have gone under but for the intervention of a
mysterious and noble an archoindiv ¡dualist, the Kaw-djer, who
ultimately discovers that political leadership demands violence
and retires embittered to a lighthouse at the end of the inhabited
world. Though full of clichés, this novel at least manifests an
interest in different political horizons together with a disbelief in
their success. The Eternal Adam, Verne’s only significant anticipa­
tion, opts therefore for a cycle of eternal return. The leading
scientist of a future civilization discovers to his horror a receding
vista of lost civilizations, including an account of the end of ours
in cataclysm and savagery'. The story conveys the realization that,
as Valéry will put it, “we civilizations, we also know now that we
are mortal”—quite a feat for the erstwhile bard of technologi­
cally conquered space. Finally, Blackland in Mission Barsac (also
translated as City in the Sahara) is Verne’s most developed anti-
utopian city (though the text might have been rewritten by
Verne’s son), astoundingly similar to a Nazi concentration camp
plus war factory, with its segregated quarters, slavery, super­
weapons, and megalomania, even to a rival SA and SS. As in
Steel-City, there is a rare appearance of the cowed industrial
proletariat, which here even participates in an uprising, and a
strong stress on the scientist’s social responsibility. All three
160 LIBERALISM ML'TES THE ANTICIPATION

works look thus into a threatening future rather than a cheery


present—a fitting final chord to Verne’s SF at the beginning of
the twentieth century.
Thus Verne’s initial dream of space can be seen as a flight
from uncertain time, and his fascination with filling in the in­
terstices of geography and science as leading to a subtle reifica­
tion. Time and Nature have to be strictly mastered, for they
threaten to run down to the cold immobility of those poles and
interplanetary spaces which attract and imperil Verne’s voyagers.
But the obsessive control over time, which enabled Fogg to van­
quish space as well as the aptly named threat of Fix and to gain
the warmth of Aouda. has in Schulze (in 500 Million . . .) become
a bearer of thermodynamic death. In Verne’s first phase the
energetic hero always taps a saving electric or volcanic energy; in
the second, Prometheus turns into Luciferian blasphemer and
energy into destruction. Verne’s world is not quite mechanical, it
is thermodynamic. Both communication and colonization mean
civilized conquest of space, which is a decrease of entropy. On
the contrary, the destruction or perversion to destructive ends of
the vehicles and embodiments of civilization, from means of
locomotion to cities and colonies, is in his post-1875 writing the
mark of entropy, of insidious Time, culminating in Time’s de­
adly reign in The Eternal Adam. From Journey to the Center of the
Earth to The Mysterious Island, decoding or understanding space
led to a happy end; understanding time leads to a dead end.
Though Verne is not a writer of anticipations, he lived in the age
of anticipation and could turn his back on it only for a brief
historical moment and at the price of seeing time as the deadly
enemy. In that he was a representative writer of the positivist
epoch. Rejecting the radical rhapsodists and the introverted
Romantics, the Victorians thought of themselves as realists. The
great scenes of Verne's SF—Axel’s dream or the vision of prehis­
toric man in Journey to the Center of the Earth, the saintsimonian
poetry of rapid motion (as in the Moon novels and the Nautilus),
of creative labor (as in the casting of the Coluinbiad), or of dis­
covery (as in Nemo’s underwater forest or in the congealing of
the supercooled sea even in the second-rate Seruadac)—all fuse
science or nature with the mysterious liberating excitement
brought by a “real” novelty. Yet his work is pervaded first by a
rhapsodic Romanticism channeled toward exotic space, and later
LIBERALISM MUTES THE ANTICIPATION 161

by a dread of time that breaks out from beneath the positive


certainty. Verne accepts from the enthusiasts of “utopian
socialism" what can be accepted by a Christian liberal—a com­
mon denominator of individual affirmation within social en­
gineering that one must paradoxically call “utopian liberalism.”
The fact that this contains something for everybody who shares
the dreams and fears of an industrial society explains Verne’s
wide popularity. But it also situates his “novel of science” half­
way between the science-oriented middle class’s11 saintsimonian
utopianism at the beginning and anti-utopian gloom at the end
of the nineteenth century. Creatively, this precarious balance
lasted only a dozen years, so long as that class could still conceive
of science and liberal capitalism as wholly concordant. After
Verne, in his imitators all around the world that balance disin­
tegrated into its components of subliterary adventure, gadget
popularization, and pretentious ideology. But while it lasted
Verne’s work itself, in spite of its long slack stretches, gave shape
to some of the most persistent because most threatened mirages
of his age: the joy of human contact by way of mastery over
nature, its mapping, and intimate penetration; the binding of
time and its translation into energetic motion and civilized ex­
pansion by means of wondrous machines. In the second phase,
he substituted for spatial wonder some sensitive préfigurations of
the dark forces menacing the liberal society. Verne’s steady vi­
sion made him the Balzac of space-machines, as his vehicles and
colonies can be called. With his twenty-odd SF books, he is the
first systematic novelist initiating dynamic plotting and the over­
view of a fully furnished world in SF—“what a style; nothing but
nouns,” remarked Apollinaire admiringly of him.11 Verne re­
vived the subterranean and interplanetary journey and intro­
duced technology into the heart of utopianism. He turned SF
toward a juvenile audience, but drew into this audience readers
of serious periodicals interested in scientific puzzles, thus setting
up the basic equation for SF consumption in the ensuing 100
years (including fan mail and fan visitors). Perhaps most impor-11 12

11. A homologous analysis for twenueth-century Anglo-American SF by Gérard Klein,


“Discontent in American Sdence Fiction," Scimee-Fietum Studies 4 (March 1977): 3-13,
offers interesting parallels to Verne's trajectory as weli as the best sociological hypothesis
for its causes so far to be found in SF criticism.
12, Quoted in Marguerite Alloue de ta Fuÿe (Bibliography IV B), p. 66.
162 LIBERALISM MVTES THE ANTICIPATION

tantly, he presented this specialized reading public with the joy


of free movement outside the compartments of Victorian society.
The bard of palaeo tech nies was ambiguously—as every signifi­
cant writer—-also the bard of its alternatives: electricity, wide-
open spaces, peace, fraternal utopian colonies.
In all these respects Verne is one of the shapers of modern SF,
and an important link in as well as modifier of the chain of its
development. He was himself acutely aware of that chain and
how he was continuing it. Apart from his wide though superficial
scientific gleanings, the allusions or outright lectures in many of
his novels show that he knew’ practically the whole tradition of
significant SF before him, his taste in books being similar to
Nemo’s. In the two Moon novels, for example, he details the
entire tradition of both beliefs and romances about the Moon,
from Thales through Cyrano to Poe, in whose honor the scene is
set in Baltimore. Poe, of whom he wrote a study, was an om­
nipresent though superficial influence on his puzzles and their
solutions, his polar voyages, and his characterization (compare
the Gun Club members and Poe’s Man Who Was Used Up).
Another major influence was Cooper—not only his Leatherstock­
ing 'I'ales but also the sea stories and the utopian romances with
their black pattern of discovery and retraction. Biblical associa­
tions are common. Sw’ift can be felt in the conflict of the Starbor-
dians and the Larbordians of Propeller Island, and Verne seems
to have known not only German Romantics but also Hawthorne.
He followed attentively his contemporaries, even those who were
learning from him, and appears to have used E. E. Hale’s Brick
Moon (1872) satellite story for Servadac, Villiers and Robida for
The Carpathian Castle, and Wells’s Imhsible Man for a later attempt
on the same theme. As for his direct influence, a whole school of
Verneans lasted well into the twentieth-century—from his col­
laborators Laurie-Grousset and Michel Verne to writers like
D’Ivoi, Graffigny, Le Rouge, or the ingenious writer-designer
Robida (La Guerre au 20e siècle, 1883, and La Vie électrique, 1893)
in France, Kraft and Dominik in Germany, Jokai in Hungary,
Salgari in Italy, and countless other imitators as far away as Rus­
sia and Japan. Also evident are the debts to Verne of many
inferior English-language works of adventure SF, from the early
1870s through E. D. Fawcett, Max Pemberton, George Griffith,
Conan Doyle, and the dime novels to E. R. Burroughs and the
LIBERALISM MUTES THE ANTICIPATION 163

Gernsbackians. Perhaps most curious is the case of Wells, who


fought so furiously against being "the English Verne" that in
some cases he apparently undertook to go Verne one better—so,
for example, in the underground life of Cavor’s Moon and in the
Time Traveller’s depths of time as compared to the descents in
Journey to the Center of the Earth or Les hides Not res, but also in his
whole program of meeting a different life as opposed to Verne’s
shving away from it. Directly and through Wells, Verne is thus
behind all modern SF dealing with the conquest of space and
social engineering.
2. Communication Breaks Down
2.0. Verne's popularity stands in startling contrast to the con­
spicuous lack of popularity of other SF between Frankenstein and
the 1870s. despite the quality of some texts. Powerful social pres­
sures on writers, running the full gamut from exile through lack
of social recognition and finances to internalization of despair
and resignation, “impeded any success for forms with an intense
speculative drift or a strong utopian and social-satire element."
Verne’s genius made a virtue of the resulting very limited “nar­
rative recipe," with its elements of verisimilitude, positivistic
popularization, closed world with circular plots, and taboos on
such radical novums as extraterrestrials, mutations, and different
sexual practices. But such limitations were in effect an interdict
by the bourgeois aesthetic practice of the times against a simul­
taneously far-reaching and hopeful imagination, against “dreams
of expanding universes ... or of the happiness of metamor­
phoses.”13 As the lines from Tennyson quoted at the beginning
of this chapter imply, only fools were supposed to believe in
the "dream” or the “fancy” of a change that would not be iden­
tical with the "ringing grooves” of railways carrying Victorian
gendemen and the products of capitalist industry into the world.
2.1. It is understandable, then, that the main body of SF
published in the age of Verne shows an equal uneasiness about
strange novelties which might imply a radically different horizon

13. For the quotes and the general argument in this paragraph I am indebted to
Marc Angenot’s allowing me to read in manuscript his pioneering essay "Science Fiction
in France Before Verne," now in Saence-Fiction Studies 5 (March 1978): 5&-66—all quotes
in this paragraph are from ibidem, 64-65.
164 LIBERALISM MUTES THE ANTICIPATION

of human relationships in the future, or indeed a radical alterna­


tive in the present. On the outskirts of capitalism, in Russia, two
seeming exceptions (which will be discussed more fully in chap­
ter 11) confirm the rule. Reacting against the overt pressures of
blatant police repression, Chernyshevsky’s What Is To Be Done
(bp. 1905, 42 years after magazine publication) was a préfigura­
tion set in an alternative present but issuing into a utopian, class­
less future that is its fulfillment. This imaginative but not escapist
attempt of bridging time-horizons or diverting the flow of time is
a revealing contrast to Verne’s only momentarily diverging pres­
ent, bereft of any such future and reduced to finding its pathos
in the actual adventures of momentary deviation, validated by
“positive” knowledge, while they were happening and so far as
they went. A confirmation of the correspondence between the
location of an SF tale and its historical horizon is to be found in
Dostoevsky’s “The Dream of a Ridiculous Man” (1877), which
acts out a fruitful tension between his skepticism about the feasi­
bility of, and his heartfelt longing after, a salvation within his­
tory. The dream of this tale is a framework for what is clearly
spelled out as a parallel universe standing for ours. The ensuing
complex balance permits Dostoevsky for the nonce to use simul­
taneously a twin, parallel history and an orientation toward the
future recovery of lost innocence.
2.2. Edward Bulwer-Lytton’s The Coming Race (1871) opts for
a subterranean location. Yet the Romantic system of corre­
spondences between depths, warmth, energy, and femininity is
in this apostate from radicalism and admirer of Frankenstein pre­
sented with mingled fascination and horror. Sympathy and will
came for Bulwer, because of their associations with revolution
and communism, to represent a power for evil as well as for
good. The addition of a banalized version of the occult, quasi-
magnetic fluid permeating men and all nature—the vri)—
completes the basic givens of and explains his ambiguous at­
titude toward the incompatibly different mode of life in the
novel. While the novel contains incidental sallies into sub-
Swiftian satire on Darwinism and American democracy, Bulwer
is mainly preoccupied with and uneasily indecisive about the sci­
entific power, the collective social organization using technics in­
stead of a “separate working class” (chap. 26), and the sexual
emancipation of the Vril-ya. His device of a menaced protagonist
LIBERALISM MUTES THE ANTICIPATION 165

was adopted by Wells and has since become the staple of anti-
utopian SF. However, the subterranean place is neither the clas­
sical Hell, nor Holberg’s excuse for satire, nor Verne’s exhilarat­
ing Mediterranean, but an omnium gatherum of demonic
menace, neutral device, and matriarchal womb—or Owen (and
even Fourier), Paltock, and Tory occultism in mindless admix­
ture. Wanting to touch all bases, such demagoguery ends finally
with no score. But obversely, at the time of publication it had a
great success with the mid-Victorian reading public and—
together with Verne, Erewhon, and the “future war” vogue—
ushered in a revival of publishing interest in an SF suggesting
but also warning against the significant novum and presenting an
individualistic—usually sentimental and horrific—melodrama
alongside new gadgetry. One of the most interesting variants of a
Bulwerian ambiguity toward the horrible but fascinating topsy­
turvy country is James De Mille’s A Strange Manuscript Found in a
Copper Cylinder (bp. 1888), which self-consciously fused it with the
marvelous-voyage tradition from More to Poe’s Pym. A positive
evaluation of the radical break implicit in a reasonable country of
female parthenogenesis is allied to an equal indecisiveness about
relating to and communicating with it in Mary Bradley lane's
Mizora (bp. 1890), located in a warm, hyperborean “Symmesian
hole." The same location is used in William Bradshaw’s The God­
dess of Atvalabar (1891) more in the vein of Haggard (and of
some Bellamy) than of Bulwer. The “lost race” tale is here
turned into a semioccultist yoking together of supergadgetry,
feverish sentimentality, and spiritual “magnicity." The baroque
exuberance of this fine piece of eccentric Victoriana, with its
holy locomotives, zoophytes, mass eroticism, and mass slaughter,
teeters between the naive and the ridiculous, and reveals more
frankly than Bulwer some of the central libidinal wishdreams
impelling these writings.
2.3. Samuel Butler’s Erewhon (1872) is a somewhat more last­
ing text because it at least approaches a sketch of the country
where ulterior motives of Victorian society are explictly unveiled.
However, Butler too uses a Vernean yet undiscovered country
“over the range,” on the traditionally upside-down antipodes.
The diverse cognitive discussions in the text—the interchanging
of illness and crime, Unreason and Reason, or religion and
banking—are not only mutually incompatible expositions, they
166 LIBERALISM MUTES THE ANTICIPATION

also hesitate between Swiftian bite and middle-class propriety,


mildly diverting paradox and cynical justification. The fable of
the Unborn has a certain Platonic charm, but what survives today
is the application of Darwinist evolution to machines that could
enslave man (beginning with the time-machine of the watch),
prefiguring as il does the discussion of reification and of
machine consciousness in cybernetics. A generation later, But­
ler’s sequel Erewhon Revisited retracted even such partial es­
trangement, as well as his own satire on the rise of religion, by its
final horizon of a saving annexation to England. It is as if the
anxious limitations of Trollope’s intervening venture into SF, The
Fixed Period, had mediated between Buller’s two novels.
2.4. Finally, under the immediate stimulus of 1871, the
momentous year of the German victory over France and the
Paris Commune, the “future war” tale blossomed forth. Though
precursors can be found as far back as the anonymous The Reign
of George VI 1900-1925 (1763), Mary Shelley's Last Man, Louis
Geoffroy, and the pseudonymous “Herrmann Lang” ’s The Air
Bailie (1859) in which the southern hemisphere and dark races
rule Europe, il was immediate political anxiety that prompted
George T. Chesney’s The Battle of Dorking (1871). The sensational
echo, the imitations, rejoinders, and alternatives it provoked,
first from other military and political gendemen and then from
popular writers jumping on a rolling bandwagon, resulted in the
publication of hundreds of books before 1914, ringing changes
upon political lineups, the ruthlessness of the enemy, and the
unpreparedness but speedy victory of one’s own gullible coun­
try.14 Their hallmark was crass Anglo-Saxon, Gallic, or Teutonic
chauvinism, escalating to the “yellow danger" racism, and loca­
tion in a shockingly imminent future well within the implied
middle-aged reader’s life expectancy. Though decisive super­
weapons were often resorted to, these tales evidence a general
inability to imagine the real economic, technological, and psy­
chological aspects of the coming world war.15 The rise of the “fu­

tí. The fundamental but certainly incomplete bibliography appended to 1. F.


Clarke’s Voices Prophesying War, 1763-1984 (Bibliography IV B) lists about 300 book titles
in English, French, and German. Including titles from the United States, and with fuller
data on European titles, the number, I would guess, would easily surpass 500. This does
not even attempt to take into account the numerous serials in boys’ magazines, etc.
15, Clarke, p. 90, notes that the only writers who came anywhere near seeing the
LIBERALISM MUTES THE ANTICIPATION 167

lure war" tale demonstrates how politics can directly bring about
a new literary form, how SF can be effectively used as a factor in
international and domestic politics, and how bourgeois expertise
could imagine a genuine future location only as awful warning
—here (as different from Verne’s Eternal Adam) meretriciously
combined with uplifting finale. The only justification of the sub­
genre is that Wells transmuted it in The War of the Worlds—by
fusing it with the "fall of civilization” subgenre—into a reflection
on the whole historical epoch of liberalism and thus into signifi­
cant SF.
2.5. As we saw in considering Verne, any significant novum,
in space as well as in time, grew untenable within liberal hori­
zons. The hesitant groping toward new horizons that ensued will
be discussed in the next two chapters. What remained was mostly
subliterature, popularizing past writing paradigms a generation
or two after they were exhausted; examples include the US
post-Civil-War dime-novel series of the “Frank Reade Jr.” or the
“Tom Swift” variety, set mostly in a Never-never Far West and
destined to have a strong influence on modern SF. There also
remained a few eccentrics swimming against the current. Among
those who are the true progenitors of significant modern SF is
Edwin A. Abbott. In his tale Flatland (1884) the location in other
geometrical dimensions is the bearer of analogies to human class
perception, conceiving, and behavior; its brief sketches show how
strictly scientific cognition and even popularization can ascend to
philosophical parable. Some other writers of the time—among
them Edward Page Mitchell in the United States, whose stories
touched upon a menacing android, a future of racial equality,
visiting the past, invisibility, and a gracious plant intelligence—
dealt with themes which were to flower in Wells and after him.* 16
The most interesting of such marginal people was Auguste
Villiers de 1’Isle Adam, a symbolist whose Cruel Tales had already

possible scientific changes in a future war were Robida, Doyle, and Wells (who foresaw
not only die tank but also poison gas and the atom bomb). The latter is the only one who,
to my knowledge, identified the psychological correlative of total warfare, and nobody
saw the economic ones.
16. Though two of Mitchell's SF stories were published in an anthology in 1884, their
first book collection seems to be the one by Sam Moskowitz as The Crystal Maa (Garden
City, NY, 1973). who also supplied a pioneering but hyperbolic introduction on the man
and the themes of his work.
168 LIBERALISM MUTES THE ANTICIPATION

sometimes hovered close to SF. His L’Evefuture (1886) grafts onto


a defiant Romanticism of the Hoffmann and Poe kind a concern
with wondrous possibilities of modern science, personified in
Edison and the '‘electrohuman” or “android” woman he fashions
for a disenchanted Byronic or Baudelairean Lord. The Shelleys’
and Verne's electricity has here literally become the soul or
Promethean spark animating an artificial, metal and plastic cre­
ation. But for all its intelligent guesses about radiant matter in
vacuum, “photosculpture,” color movies, and human electro­
magnetism, the interest of Villiers’s novel lies in its twofold
theme: how to find a lion-philistine Beauty in the world and howr
to determine the difference between a “real" but insensitive
woman and an “illusion" whose behavior is more intelligent and
generically more human than that of the original “bourgeois
goddess." The possibility of androids is suggested by the
stereotyped behavior and ideas of people in the world of propri­
ety and self-interest. “Chimaera for chimaera,” why should not
an openly and purely artificial being, incarnating the best of
human knowledge and genius, constant, sexless, and immortal,
be preferable and indeed more real than such people are? And
could not an ideal love be felt for such a humanized being, a
“machine for manufacturing the Ideal,” rather than for the false
humans?
In a divergence from Verne, this alternative present of pessi­
mistic idealism does not evade but passionately hates the future
in which the steam-boiler and egotistical aridity have suppressed
the old values. Hadaly, the “future Eve,” is both the logical end­
product of bourgeois reification or standardization (satirized also
in some of the “cruel stories,” for example, "The Heavenly Ads”)
and a rich countercreation named after the Ideal. A strong dose
of misogyny, amounting at times to a kind of sexual racism, leads
paradoxically to the concept of a Platonic, superhuman yet
woman-shaped, soul-sister as the worthy companion of man in
his terrestrial exile. The novel teeters between the sublime and
the ridiculous, with many a passage of exclamative sentimentality
and shrill preaching, and ends with the usual twist into timeless—
supernatural and even occult—fantasy and the destruction of the
novum. Yet it not only contains splendid scenes of a Decadent,
Salammbo-like otherwhere, like Edison’s underground realm of
electricity with its artificial birds and flowers, it also uncovers the
LIBERALISM MUTES THE -ANTICIPATION 169

psychological source of many seemingly erotic but in fact asexual


robot tales; and at its nucleus there is an incipient modern dis­
cussion of free will, creations with different types of logic, the
burden of consciousness, and mass production of “ideals.” Puz­
zles of identity, role playing, and semantics appear in Villiers
amid long stretches of philosophical and technical monologues.
Thus, for all its impurities, The Future Eve carries the discussions
of Frankenstein or “Rappaccmi’s Daughter” into the age of Verne
and Butler, envisaging their fascination with machines and final
anti-utopian distrust of bourgeois progress as an existential prob­
lem of human sensitivity and intelligence. More than the
Nemo/Nautilus centaur, or even Frankenstein’s Creature, the
melancholy Hadaly, a union of spirit and scientific hardware, is
the liveliest space-machine of the century.
3. Muted and transferred from time to space, the novum
turned malevolent there too. Verne’s spinning world of progress
spun finally into a cheerless cycle of eternal return. Communica­
tion was turning into a nightmare of Butler's grinning statues
and rigidifying orthodoxies, Bulwer’s collective superwomen,
or the exchange of high explosives in future wars. In “Locksley
Hall Sixty Years After" the hesitant Tennyson swung toward the
disenchanting result of the century’s original if ambiguous
promise:
Gone the cry of “Forward, Forward,” lost within a growing
gloom;
Lost, or only heard in silence from the silence of a tomb.
Half the marvels of my morning, triumphs over time and
space,
“Forward" rang the voices then, and of the many mine was
one.
Let us hush this cry of “Forward” till ten thousand years
have gone.
[73-78]
But the denial of the future led to such stifling closed worlds
and so acute menaces that it eventually provided the basis for
rehabilitating the possibility of a brighter anticipation in the
dawning of socialist hope with Bellamy and Morris.
8

Anticipating the Sunburst:


Dream, Vision—or Nightmare?

The great cry that rises from all our manufacturing cities, louder
than the furnace blast, is all tn eery deed for this,—that we man­
ufacture there everything except men; we blanch cotton, and
strengthen steel, and refine sugar, and shape pottery; but to bright­
en, to strengthen, to refine, or to form a single living spirit, never
enters into our estimate of advantages.
John Ruskin
Is the Earth so?
Let her change then.
Let the Earth quicken.
Search until you know.
Bertolt Brecht
Was it a vision, or a waking dream?
Fled is that music:—do I wake or sleep?
John Keats
0. The gloom and recantation of SF—including utopian or
social-science-fiction—writers from Mary Shelley and Herman
Melville to Jules Verne and Villiers de l’lsle Adam was part of
the increasing closure of liberal bourgeois horizons. Yet at the
same time the thirst for anticipations—fictional pictures of an
excitingly different future—rose sharply (one assessment puts
their frequency from 1871 to 1916 at 35 times the pre-1870
rate of publication).1 SF is as a genre potentially and even intrin­
sically oriented toward humanity's furthest horizons, and there­
fore in quite aesthetic terms (that are, of course, inseparable
from ethical and cognitive ones) not fully developed in the
timeless, cyclical, or merely catastrophic realizations discussed in
the last two chapters. Consequently, the radical alternative of a
socialist dawn found an even more congenial soil and erupted

1. Calculated by me from Ian Clarke (Bibliography 1).

170
ANTICIPATING THE SUNBURST 171

even more strongly in it than in the contemporary political surge


in Germany, Britain, the United States, and elsewhere. In addi­
tion to its thematic and ideological appeal this alternative had
the merit of solving the racking dilemmas brought into existence
by the time of the radical Romantics—of movement forward ver­
sus the closed circle, wish versus realization, freedom versus
brotherhood, skepticism versus belief, individual versus society.
A whole century had dealt with these dilemmas by ingenious or
feeble evasions within a spatial symbolism, and had in plot­
endings washed its hands of the cognitive reason for the story's
existence. Therefore the SF narrations from Frankenstein to
Villiers and most of Verne culminated in destructions and mur­
ders as the logical end and outcome of the quantitative, indi­
vidually anguished Faustian quest—as opposed to the qualitative,
collectively subversive Promethean quest of earlier utopian and
SF writers, from More and Cyrano to Percy Shelley. Even Goethe
felt _hejx>uld avoid such an outcome only by tacking on to his
Faust a religious happy-ending incompatible with_ the initial
wager that validates the plot The socialist vision of a classless
millennium on Earth was thus a solution to both the ideational
and the formal problems of nineteenth-century SF. It flourished
for a brief time in Bellamy and Morris, the absence of its open
horizon explains Mark Twain's impatience and despairing fail­
ure in A Connecticut Yankee, and at the end of the century it
provided one of the basic ingredients for Wells's ambiguous
synthesis.
1.1. In Looking Backward 2000-1887 (1888) Edward Bellamy
started not only from the widespread Victorian observation that,
as Disraeli put it, the rich and the poor were “two nations,” but
also from the observation that “[the] working classes had quite
suddenly and very generally become infected with a profound
discontent with their condition, and an idea that it could be
greatly bettered if they only knew how to go about it.” Bellamy
was willing to show them how, for it was “[not] only the toilers
of the world who are engaged in something like a world-wide
insurrection, but true and humane men and women, of every
degree, are in a mood of exasperation, verging on absolute re­
volt, against social conditions that reduce life to a brutal strug­
gle for existence . . (chap. 1). In Equality (1897), a sequel w hich
set out to plug the gaps left by the first novel, he added to these
172 ANTICIPATING THE SUNBURST

sources of discontent the ruin of prairie farmers by capitalist


mortgages, the degradation of women fostered by economic ex­
ploitation, the recurrent economic crises of the last third of the
nineteenth century, and the concentration of three-quarters of
all national wealth into the hands of 10 percent of the popula­
tion. Bellamy’s utopianism was the point at which all these deep
discontents (which in the decade of the Chicago Haymarket trial
ran the whole gamut between bankrupt smaller businessmen and
the industrial workers who participated in the almost 6,000
strikes per year) intersected with the earlier utopian-socialist tra­
dition of American religious and lay associationism and with the
experiences of the nineteen th-century socialist movement. As a
spokesman of the American “immense average of villagers, of
small-town-dwellers,” Bellamy believed in “modern inventions,
modern conveniences, modern facilities”2—in Yankee gadgetry
as a white magic for overcoming drudgery. This perspective dif­
fered from the Populist revolt, which inveighed in the name of
the smallholder against the financial trusts of Wall Street enslav­
ing the countryside by means of railways. Bellamy accepted
the trusts as more efficient and—following their own logic—
condemned only their private character as economically too
wasteful and politically too dangerous to tolerate. Instead of this
corporate tyranny, his practical streak of “Yankee communism,
or, to be more precise, ‘Associationism' ”3 led him to envisage
“the nation . . . organized as the one great corporation . . . , the
sole employer, the final monopoly ... in the profits and
economies of which all citizens shared" (chap. 5). Bellamy’s new
frontier, replacing the West traversed by the irreversible rails, is
the future. It offers not only better railways, motor carriages,
air-cars, telephones, and TV, but also a classless social brother­
hood of affluence which will make these means of communica­
tion generally accessible and will socialize all other upper-class
privileges such as culture. Comfort and security are the ends of
Bellamy’s utopia, and economic reorganization the means. In
this pragmatic socialism, unhappiness is ethical waste: Looking
Backward shows forth “the economy of happiness.”4 That is

2. W. D. Howells (Bibliography IV C). p. ix and vii.


3. Daniel Aaron (Bibliography IV C), p. 95.
4. Unlocated quotation, apparently from Bellamy's diary, in Aaron, p. 97.
ANTICIPATING THE SUNBURST ITS

brought about by universal high education, universal industrial


service from the twenty-first to the forty-fifth year, equal and
guaranteed income (in nontransferable yearly credits) for every
citizen including the old, children, and the sick, a flexible plan­
ning adjusting workloads and production according to demand,
and a highly developed system of public bestowal of honors.
Government is reduced to the operations of the Great Trust
or—since the economy is run on the lines of universal civic ser­
vice analogous to the military service—the Industrial Army. In it,
every citizen rises through the ranks as far as his capacity will
carry him. The generals of each guild or industrial branch are,
however, not appointed from above but chosen by all the retired
members or alumni of the guild, and so on up to the head of the
army who is president of the United States. Doctors and teachers
have their own guilds and regents outside this army, and a
writer, artist journal editor, or inventor can be exempted from it
if a sufficient number of buyers sign over a part of their credit to
support him. The sequel in Equality clarifies that economic
equality gives free play to the greatest possible expansion of in­
dividuality, that there is a reservation for Thoreau-like objectors
to “work out a better solution of the problem of existence than
our society offers” (chap. 5—possibly the first use of this recur­
ring escape hatch of later utopias), that the population of the
cities has drastically shrunk, that all tools are electrically powered
and garments made from disposable paper, and so on.
Bellamy’s economic blueprint is integrated into the story of
Julian West, who wakes from a mesmeric sleep begun in 1887
into the year 2000, is given information about the new order by
Dr. Leete, and falls in love with Leete’s daughter Edith. Further,
all this is a “romantic narrative” (Preface) by an anonymous his­
torian writing in the festive year 2000 to instruct his readers in
the contrasts of past and present, by looking backward. This
system of mirrors and receding vistas in time is memorably reac­
tualized in the nightmarish ending when Julian dreams of awak­
ening back in the capitalist society of 1887. He meets its folly
and moral repulsiveness with an anguished eye which supplies to
each spectral place and person a counterpossibility. This utopian
estrangement culminates in the hallucination about “the possible
face that would have been actual if mind and soul had lived”
which he sees superimposed upon the living dead of the poor
¡74 ANTICIPATING THE SLNBLRST

quarter; the lesson is (hat living in this nightmare and “pleading


for crucified humanity” might yet be better than reawakening
into the golden twenty-first century—as, in a final twist, Julian
does (chap. 28). Looking Backward—intimately informed by Bel­
lamy’s constant preoccupation with human plasticity, with mem­
ory’ and identity (concerns of his Dr. Heidenhoff’s Process [1880]
and Miss Ludington’s Sister as well as a number of his short
stories), with brute reality and ideal possibility—reposes on a
symbolic balance of time horizons. Its plot is, in fact, Julian’s
change of identity. In two of Bellamy's later SF stories, “The
Blindman’s World” and “To Whom This May Come" (1898), the
improvident Earthmen, sundered from their neighbors and
self-knowledge, are contrasted to worlds of brotherhood and
transparency where men are "lords of themselves.” Julian West,
the idealistic and insomniac rich idler with a revealing name,
becomes an apostate from such a life in the West of 1887
through his education into a citizen of 2000, which is effected
through a healer’s reasonable lectures and his daughter’s healing
sympathy and intercession. The construction of a social system for the
reader is also the reconstruction of the hero. This innovation in the
utopian and SF tradition uses the radical-democratic paradigm
which (as was noted in chapter 6) began with the American Revo­
lution, for a positive answer to Shakespeare’s and Swift’s chal­
lenge that changing the world entails changing (he “nature” of
men. On the other extreme from Frankenstein, such a dialectical
malleability is also an epoch-making pointer to future SF.
On the other hand, Bellamy immediately retreated from this
discovery. Just as Julian is the mediator between two social sys­
tems for the reader, so Edith Leete—the descendant and, as it
were, reincarnation of Julian’s fiancee from 1887—is the steady­
ing emotional mediator in his passage to a new world, a personal
female Christ of earthly love and brotherhood. Bellamy’s is an
ethical socialism, abhorring violence and haired. The “sunburst”
of a new order, “when heroes burst the barred gate of the fu­
ture” (chap. 20), is validated equally by economics, ethical evolu­
tion, and Christian love; unethical economics was for him un­
workable. Such a millenary horizon makes for a fundamental,
qualitative change; the future brings a different, purified space
and man. The friendly house of Dr. Leete stands on the burnt-
down remnants of Julian’s house and on top of his underground
ANTICIPATING THE SUNBURST 175

shelter, which have to be excavated as a feat of archeology for


the twenty-first century. For Julian, the Leete household is the
hearth of spacious, reasonable, clean, classless Boston of the year
2000, and Edith is not too far from an image of his favorite
writer Dickens, the cricket on the hearth. The hard-headed civic
pragmatism is only the obverse of a soft-hearted petty-bourgeois
romance or “fairy tale of social felicity.”5
This fairytale character is most evident in Bellamy’s sanguine
expectation of a nonviolent, imminent, and instantaneous aban­
donment of private capitalism by universal recognition of its
folly. With telling effect he extrapolated bourgeois rationality,
ethics, and institutions to a logical end-product of universal pub­
lic ownership. But this consistent pedagogic starting from the
known signifies a sanitizing of capitalism to ensure the freedom,
equality, brotherhood, and abundance of the Rationalist or Jef­
fersonian dreams. Bellamy remained limited by such ideals,
wfhich form an important part but by no means the final horizon
of a socialist future, it is perhaps unfair to judge his fascination
with the army as a model of rational organization by the norma­
tive ethical reaction toward armies today, since he acquired it in
Lincoln's days and translated it into peaceful and constructive
terms, just as Fourier did. Further, any self-respecting utopia
before automation had to ensure its working by a certain harsh­
ness for recalcitrants (and Bellamy—possibly learning from
Morris—clearly evolved toward greater openness and participa­
tory democracy in Equality, where all officials are subject to recall).
Nonetheless, even there he continued co stress a hybrid of state
mobilization and “public capitalism" (chap. 22); neither did
he modify his patronizing dismissal from Looking Backward of
“the more backward races” (chap. 13) and political efforts by nar­
row-minded workingmen, nor, above all, his faith in techno­
cratic regimentation within economic production as opposed to
ideal classless relations outside them, all of which strike an
alienating note in the tradition of Saint-Simon and Cabet rather
than that of Fourier or Marx. That note is out of harmony with
his basic libertarian preoccupations, and introduces into his ro­
mance a cold and static element.

5. Edward Bellamy, "How I Came to Write Looking backward" The Nationalist (May
1889): I; reprinted tn Scitnee-Fution Studies 4 (July 1977): 194.
176 ANTICIPATING THE SUNBURST

But if Bellamy is a pragmatist who is not comfortable when


depicting sweeping processes of change, he is at his strongest in
the shrewd treatment of the economics of everyday life—of dress­
ing and love, the distribution of goods, the cultural activities—
and in the brilliant passages on making democratic supply and
demand work outside a capitalist framework, for example, in
organizing a journal or in solving brain-drain between countries.
On such occasions, Bellamy is quite free from a State Socialism
regulating everything from above. When contrasting such warm
possibilities with the irrationality and dead-end character of pri­
vate competition, his clear and attractive, though sometimes
pedestrian, style rises to little parabolic inserts of great force, as
the initial allegory of the Coach, the parables of the Collective
Umbrella and of the Rosebush, or (in Equality) the parables of
the Water-Tank and of the Masters of the Bread. All such
apologues, exempla, and parables come from a laicized and radi­
cal pulpit style, openly displayed in the sermon on the sunburst
from Looking Backward. It is within this New England oral and
public tradition, from the Bible and the Platonic dialogues and
not from the genteel literature of Gilded Age mandarins, that
Bellamy’s rhetoric arises as an imposing and sometimes splendid
accomplishment of its kind. Such addresses were primarily
meant for middle-class women, and Julian’s sentimental intrigue
as well as the whole ethical tone of Looking Backward addressed
itself to them, and generally to that part of the educated classes
which felt insecure and unfree in bourgeois society. Thus Bel­
lamy's homely lucidity made his romance, with all its limitations,
the first authentically American socialist anticipation tale.
Bellamy's success can—as always in significant SF—be ex­
pressed in terms of a creative fusion of various strands and tra­
ditions. These were not only literary, but reached back to the
hundreds of religious or lay utopian communities which had
been cried in the young United States. Though all of them finally
collapsed as utopian communities under the violent pressures of
an inimical environment, their legacy to American horizons from
Hawthorne to our day has been larger than commonly assumed.
An attenuated lay vision of the glorious City had now and then
crossed from the oral and hortatory into the written fictional
tradition, in works such as Mary Griffith’s feminist, abolitionist,
and technological anticipation “Three Hundred Years Hence"
ANTICIPATING THE St'NBLRST 77

(1836), Edward Kent's and Jane S. Appleton’s future city of


Bangor in Voices from the Kenduskeag (1848), and several descrip­
tions by Edward Everett Hale culminating in "My Visit to
Sybaris” (1869) and How They Lived in Hampton (1888). Though
Howells exaggerated when he claimed for Bellamy “a romantic
imagination surpassed only by that of Hawthorne,”8 Bellamy did
interfuse such narratively helpless precursors—to whom one
should add Fénelon, Cabet, and contemporary British anticipa­
tions by E. Maitland {By and By, 1873), H.C.M. Watson (Ercho-
menon, 1879), W.D. Hay (Three Hundred Years Hence, 1881), or
P.C. Wise (Harkness and Dawn, 1884)—with an effective Roman­
tic system of correspondences. In particular, he seems to have
drawn on a number of important elements from John Macnie’s
The Diothas (1883), such as a utopia with an industrial army, love
with a descendant of the nineteenth-century sweetheart named
Edith like her ancestress, or the use of radio.67 But, most impor­
tantly, Bellamy was the first to go all the way with such a lay
millenarianism. Therefore, his ending, which refuses the easy
alibi of it all being a dream—a norm from Mercier and Griffith
to Macnie—marks the historical moment when this tradition
came of age and changed from defensive to self-confident. The
new vision achieves, within the text, a reality equal to that of
the author’s empirical actuality. This claim translates into his­
torical cognition Hawthorne’s psychological fantasy and, es­
pecially, the long sleep of Irving’s Rip van Winkle, itself cognate
to folktales such as the Sleeping Beauty or the Seven Sleepers
(Hawthorne and Irving are the only U.S. authors in Dr. Leete’s
library). Bellamy links thus two strong American traditions: the
fantastic one of unknown worlds and potentialities, and the
practical one of organizing a new world—both of which avail
themselves of powerful biblical parallels while translating them
from religion to economics. His materialist view of history as a
coherent succession of changing human relationships and social
structures was continued by Morris and Wells, and thence was

6. Howells, p. xiii.
7. Arthur E. Morgan's refutation, in his Plagiarism in Utopia (Yellow Springs, OH,
1944), of Bellamy's supposed plagiarism from The Diothas seems both unconvincing and
unnecessarily fond of the shibboleth of "originality," exactly as pertinent to literary value
as bourgeois copyright law. and particularly inapplicable to SF, including utopian fiction.
178 ANTICIPATING THE SUNBURST

built into the fundaments of subsequent SF. The same holds


for the plot that educates the reader into acceptance of the
strange locus and its values by following the puzzled education
of a representative protagonist. Modern SF, though it has for­
gotten this one among its ancestors, builds on Looking Backward
much as Dr. Leete’s house was built on Julian’s ruins and on top
of the hermetically sealed sleeping chamber under its founda­
tions.
Particular traits from Bellamy’s other works also drew from
and returned into the SF tradition. The Flam mar ion-like cosmi­
cally exceptional blindness of C.S. Lewis’s Earthmen and E.R.
Burroughs’s transferral by spirit to Mars are also found in “The
Blindman’s World,” and the despotic oligarchy as the alternative
to revolution in Wells and London has a direct precursor in
Equality. Most immediately, the immense ideológico-political echo
of Looking Backward reverberated around the globe through
numerous writers of sequels, rebuttals, and parallels. Bellamy
had hit exactly the right note at a time of widespread search for
alternatives to ruthless plutocracy, and between 100 and 200
utopian tales expounding or satirizing social democracy, state
regulation of economy, a Populist capitalism, or various uncouth
combinations thereof were published in the United States from
1888 to the first World War. Though none of them approached
Bellamy’s coherence, the most notable were Ignatius Donnelly's
melodramatic Caesar’s Column (1890) and Howells's politely satir­
ical discussions in^ Traveller From Altruria (1894). In Britain the
echo was to be felt down to Wells, and in Germany it re­
sulted in at least two dozen utopian and anti-utopian tales. But
the perfect complement to Looking Backward was written by Wil­
liam Morris.
1.2. As so many other utopias, Morris’s News From Nowhere
(1890) was a direct reply to Looking Backward. Reviewing it, he
had denounced Bellamy’s “unhistoric and unartistic” tempera­
ment which “makes its owner (if a socialist) perfectly satisfied
with modern civilization, if only the injustice, misery and waste
of class society could be got rid of” and whose ideal of life “is
that of the industrious professional middle-class man of today,
purified from [the] crime of complicity tvith the monopolist
class"; whence it follows “that he conceives of the change to
socialism as taking place without any breakdown of that life, or
ANTICIPATING THE Sl'NBCRST 179
indeed disturbance of it.” Morris especially objected to Bellamy’s
stress on both technological and social machinery that leaves the
impression “of a huge standing army, tightly drilled," to the
corresponding “State Communism” as opposed to direct par­
ticipatory democracy, and to the reduction of labor instead of its
change to work as pleasure, work blended with an art which “is
not a mere adjunct of life . . . but the necessary and indispens­
able instrument of human happiness.”8
Accordingly, it is direct, sensual relationships of men to each
other and to nature, a different civilization where useful work is
pleasure, that provide the fundament of News From Nowhere. It
adopts the frame of Looking Backward, which begins with the
narrator falling asleep and waking up in the future house built
on the place of his own, and ends with his terrible return to his
own time. But from the vert' beginning, Morris’s story is a coun­
terproject to Bellamy’s. It is presented neither as a safe retros­
pective from the year 2000 nor as the voice of a lone member of
the upper class, but as one privileged voice and vision of the
future among several others possible and held within the socialist
party of which Morris was a member, and in whose periodical
News From Nowhere was published serially. The whole story is
informed by the tone of a man displaying his personal vision for
consent to potential comrades in bringing it about, and yet very
aware of its distance in the future. This approach blends collec­
tive validity and personal heartbreak. It is much richer than the
easy Christian Socialist resolution of Julian West’s private an­
guish by means of a resurrected bride, for it takes into full con­
sideration both the collective difficulty of arriving at and the
personal impossibility of setting up an abode in the promised
classless land; the narrator William Guest—Morris’s persona—is
in the position of Moses walking through a vision of Canaan.
Therefore, the story does not, as does that of West, progress
through sallies from a safe individual hearth. It retains the obli­
gatory outlining of the future in the Mercier-to-Bellamy tradi­
tion—here a ride from Hammersmith to the British Museum,
that repository of collective memory; but it adds two further

8. William Morris, “Looking Backward," The Commonweal, 22 June 1889; reprinted in


Science.Ficdon Studies 3 (November 1978): 287-90, together with Morris's also pertinenl
introduction io his Ke Im scon Press edition of More's Utopia.
ISO ANTICIPATING THE SUNBURST

and historically new elements. First, it introduces an account


of the revolution that brought the future about; though today
this account may seem still too naive and optimistic, it is of a dif­
ferent order of credibility than the sudden wholesale social con­
versions depicted by previous writers up to and including Bel­
lamy. Second, the bittersweet rowing up the Thames shows
what the future might have meant to the author-narrator per­
sonally. Together with the ubiquitous guide Dick, the average
Nowherean, Guest’s main partner in the first part of the story is
Old Hammond, the custodian of history, and in the second part
Ellen, the incarnation of the “pleasure of life” of the future
present.
The narrator’s vision is also a dream. Not only can it naturalis-
tically be considered a dream in his nineteenth-century Ham­
mersmith bed, it is also a wish-dream. Reacting against the
capitalist use of machinery that polluted the life of man and the
Earth and created ugliness and misery, Morris began with the
Pre-Raphaelite tradition of art as daydreaming. However, in its
refusal to look deeper into the basic problems of reality such an
art became nonetheless the complement of reality, as green
complements red, and thus directly dependent upon it—pretty
where actuality is ugly, sweet where it is bitter, brightly shaped
where it is amorphous and sooty, a pastoral where it is an ulcer
of slums:
Forget six counties overhung with smoke,
Forget the snorting steam and piston stroke,
Forget the spreading of the hideous town;
Think rather of the pack-horse on the down,
And dream of London, small, and white, and clean,
The clear Thames bordered by its gardens green. . . .
[Morris, The Earthly Paradise]
Steam-age capitalism was ruthlessly transforming towns into “an
extension of the coal mine”9 and the countryside into the spoils
for the railway that already Thoreau complained about, and de­
veloping the war of each against each into global imperialisms.
In a technique similar to More’s, News From Nowhere is primarily
a counter project to that life:

9. Lewis Mumford (Bibliography IV A), p, 159—but see his whole chapter 4.


ANTICIPATING THE SL’NBLRST 181

The hope of the past times was gone. ... Was it all to
end in a counting-house on the top of a cinder-heap,
with Podsnap’s drawing-room in the offing, an a Whig
committee dealing champagne to the rich and mar­
garine to the poor in such convenient proportions as
would make all men contended together, though the plea­
sure of the eyes was gone from the world . . . ?
[Morris, “How I Became a Socialist”1011
]
At this high point of the paleotechnic world, any sensitive artist
might have wished with Guest “for days of peace and rest, and
cleanness and smiling goodwill” (chap. 1). For its “realities were
money, price, capital, shares: the environment itself, like most of
human existence, was treated as an abstraction. Air and sunlight,
because of their deplorable lack of value in exchange, had no
reality at all."11 On the contrary, News From Nowhere presents an
airy and sunny environment, where only direct interhuman rela­
tions are clearly envisaged. In contrast to the capitalist gospel of
toil, work as playful human necessity stands at the novel's moral
center. In contrast to the Victorian starvation of the mind and
die senses, the novel’s figures are perhaps the fullest and least
self-conscious Epicureans in modern English literature. And in
contrast to the terrible anxieties of blood-and-iron progress, the
novel’s subtitle is “An Epoch of Rest.”
There is accordingly a strong element of mere escape in News
From Nowhere. With disturbing implications for a utopian ro­
mance, Morris overreacted into a total refusal to envisage any
machinery, technological or societal. This amounts to leaving his
future society without any economic or organizational basis. As
to the economy, a “force-barge” with an undisclosed new energy
is the only exception to a turning away from and indeed dis­
mantling of technology in Nowhere. People who are so minded
can collect together to draw on the universally available "power”
(energy) in workshops, but this is used only for handicrafts; for
the rest, England is now a garden. Morris makes some telling

10. All of Morris's essays cited in this chapter in parenthesis by title are to be found in
Political Writings of William Morris, ed. A. L, Morton (New York, 1973), an indispensable
companion to his novel; in particular, the essay "The Society of the Future" (1887) is the
nearest thing to an ideological nucleus of Xeuis From \’owhtre.
11. Mumford, p. 168.
182 ANTICIPATING THE SUNBURST

points about the "never-ending series [of] sham or artificial


necessaries’’ (chap. 15) imposed by the capitalist world market
that necessarily enslaved colonial countries as a counterpart to
the corruption of consumer taste. But to reject resolutely not
only useless forms of technology and industrial organization but
technical productivity and inventiveness in general while keeping
the population stable and affluent, as tVews From Nowhere does, is
self-defeating. Any utopographer has the right to fashion his
Lit nd of Heart’s Desire; but he has a corresponding obligation to
make it an at least arguable alternative possibility.
It could be argued that the gap left by Morris’s disgust with
modern economics might today be filled in bv an imaginative
reader supplying his own economics, based on the possibilities
of automation and other ways of “post-industrial” productivity
which Morris could not know about—though his vision assumed
that humanity would somehow evolve “immensely improved
machinery“ (chap. 15) for irksome tasks if only basic problems of
social organization were solved. Unfortunately, the absence of
sociopolitical organization in Nowhere is a gap that cannot be
argued away and denies it the status of a utopia. True, there is a
classical Marxist glimpse of “communes and colleges” (chap. 5)
run by participators' democracy. However, overcom pen sating for
Bellamy’s state apparatus and clear lines of power, News From
Nowhere omits all machinery for determining priorities between
communes or any other basic units. Yet all production, includ­
ing very much an automated one, requires—as long as it is not
simply magical—coordination and a (however truly participa­
tory) system of vertical decision making. As Bellamy astutely
countered in his review of Morris, “[no] degree of moral im­
provement [will] lessen the necessity of a strictly economic ad­
ministration for the directing of the productive and distributive
machinery.”12 News From Nowhere sacrifices human productivity
in order to get rid of Statism and technocracy.
But if it is not a utopia, and much less prophetic anticipation,
News From Nowhere is the finest specimen of Earthly Paradise
story in modern literature. As I argued in chapters 3 and 5, the
Terrestrial Paradise—a place of this-worldly fleshly contentment,

12. [Edward Bella mv.) “News from Nowhere," New Nation. 14 February 1891, p, 47,
quoted in Sylvia E. Bowman et al. (Bibliography IV C), p. 94.
ANTICIPATING THE SUNBURST 183

magical fertility, happiness, health, and youth—is a wish-dream


that does not focus on economic and sociopolitical organization;
it is a magical parallel world akin to folktale and pastoral yet of
collective import as an alternative community to be striven for.
Morris’s tale has almost all of these elements. The weather and
the people are (perhaps a shade monotonously) perpetually
summen, the salmon are back in the Thames, and Shelleyan
consentaneous love inspires all breasts (though there is just
enough exception to keep it from being too saccharine). Liber­
ated from grim capitalism, the world has entered upon its “sec­
ond childhood” (chap. 19), very similar to an idealized version
of the fourteenth century and characterized by a childishly un­
spoiled enjoyment in artful work not sundered from play: all its
people look younger than they would in our civilization. Above
all, the dryness of the usual utopian panoramic sweep is avoided
by Morris’s fashioning the second part of the story as a personal
working out of the new country, as a glimpse of the narrator’s
alternative—happy and wholesome—life. His journey up the
magical waters of the fertile Thames, signposted with references
to a range of legends from the Grimm Brothers tales “from the
childhood of the world” (chap. 16) to Tennyson's Lotos Eaters,
shows Morris’s rich and contrapuntal use of the Romantic fairy
tale. The newly fertile land and happy relationships in the future
England are the result of a metamorphosis from the ugly
Victorian past—still inscribed in Guest’s clothes, looks, and
memories—to the clear and colorful beauty of the Nowherean
present, a metamorphosis analogous to Andersen's ugly duckling
reborn as a beautiful swan. Under the spell of his rejuvenating
journey toward the sources, the narrator also moves toward the
happiness felt in his childhood.
Nonetheless, “shades of the prison-house” are inescapably
upon the narrator-protagonist: he personally can only testify to,
not accomplish for himself, the metamorphosis that brings hap­
piness. The tension between the report about collective happi­
ness and the personal melancholy of the guest-reporter in that
Earthly Paradise—for him truly a Nowhere—-refuses a Bellamy-
type sentimental happy ending. The crucially more mature reso­
lution is not one of ethical salvation, as in Looking Backward, but
one of political strife. We are back at Blake’s great oath not to let
“the sword sleep in my hand” until Jerusalem is built “in Eng­
184 ANTICIPATING THE SUNBURST

land’s green and pleasant land.” But here such a strife is trans­
lated from Blake's arena of a single mind to public political
struggle, as personal compensation for and collective justification
of Guest’s visit and departure;
. . . Ellen’s last mournful look seemed to say, “No, it will not
do; you cannot be of us; you belong so entirely to the un­
happiness of the past that our happiness even would weary
you. ... Go back and be the happier for having seen us, for
having added a little hope to your struggle. Go on living
while you may, striving, with whatsoever pain and labour
needs must be, to build up little by little the new' day of
fellowship, and rest, and happiness.”
Yes, surely! and if others can see it as I have seen it, then
it may be called a vision rather than a dream, [chap. 32]
For this dream is, finally, to be understood in the tradition of
the medieval genre of the same name, in which the convention,
as in Langland or Chaucer, is that the author relates the dream
as a non-naturalistic analogy—often using the fable or other al­
legorical means—to public problems of great personal import.
Morris had already used this convention in his SF story about the
peasant revolt in the Middle Ages,d Dream ofJohn Ball (1888). Its
narrator, double horizon of defeat and yet victory, historical as­
sumptions, and lime scheme combined w’ith color imagery (night
and moon opposed to “the east crimson with sunrise”) prefigure
the fuller use in News From Nowhere. But just as A Dream ofJohn
Ball was not an Individualistic historical novel, so the later work
is not to be taken for positivistic prophecy but for the figure or
type of a fulfilment that could or should come. In that round­
about, dialectical way News From Nowhere and its “ideal of the old
pastoral poets” can, through its nucleus of frank and beautiful
human relationships to other humans and to nature, be reinte­
grated into anticipatory utopianism. Its Earthly Paradise is an
analogy to the classless socialist day. Its collective dream, “if
others can see it," will finally also be a vision reinserted into
history. Staying within the bourgeois—or indeed WASP—
existential horizons, Bellamy had pursued the everyday need for
security to its logical conclusion and ended up with the socialist
dawn as an order of things, a societas rerum. Reneging on the
bourgeois existential horizons but opposing to them unrealisti­
ANTICIPATING THE SUNBURST 185

cally idealized pre industrial—indeed bohemian—horizons,


Morris pursued the arrested timeless moment, the visionary
dream (in all the above senses) of Earthly Paradise to its logical
conclusion and ended up with another aspect of that same dawn:
creative and therefore beautiful human relations, a sodetas
haminum. Between them, they covered the technical premises and
sensual horizons of that dawn: each lacks what the other has. For
a brief but still significant historical moment—which extended to
Wells, London, and Zamyatin—the discussion about darkness
and dawn became one inside the international socialist move­
ment.
Dawn or sunburst, a favorite image of that whole movement, is
here particularly appropriate because of close correspondences
among people, vegetation, and the seasons of day and year.
Morris's narrator went to sleep in a wintry night, when the young
moon portended renewal; he wakes up, "by witness of the river­
side trees,” in a bright morning of spring or summer. The sun­
light denotes happiness, as the moonlight throughout the story
reminds die narrator of the past times. Colors too are connected
with the opposed tempers and historical epochs, “the sombre
greyness, or rather brownness, of the nineteenth century"
versus “the gaiety and brightness” of the twenty-first (chap. 19).
Mankind has again become a part of nature, “men and women
[are] worthy of the sweet abundance of midsummer” (chap. 21),
and the river-side trees are emblems as much as witnesses to it.
The representative denizen of the future, Ellen, sums it up in
her cry: “O me! O me! How I love the earth, and the seasons,
and weather, and all things that deal with it, and all that grows
out of it . . .” (chap. 31). She is the ideal partner or anima of
Morris, who characterized himself in “How I Became a Socialist”
as a man "careless of metaphysics and religion, as well as of
scientific analysis, but with a deep love of the earth and the life
on it, and a passion for the history of the past of mankind.”
The arrested moment of Earthly Paradise is conveyed by a
series of pictures, one of Morris's basic stylistic devices. The vi­
sion in News From Nowhere sharply etches in colors and shades,
rime and place—especially the topography of London and the
Thames valley—but it is most attentive to the sensual nuances of
behavior and movement of humans through a nature produced
by their hands’ work. The beautiful bridges, the gardenlike
186 ANTICIPATING THE SUNBURST

banks of the Thames, the hayseis, and the old house that grows
out of the earth are of the same stuff as the nut-brown maids
“born out of the summer day itself” (chap. 21), Howers in the
green countryside. Yet this pictorial, at times somewhat pictur­
esque vision is ever and again clouded by the dreamlike melan­
choly and alienation of the beholder. The bemused and never
quite sunny narrator does not fully fit into the bright day of the
pictorial narration. He comes from the wrong, moony or night
side of the dawn, and he finally has to step outside the picture
frame and fade from the Earthly Paradise. Yet, in their turn,
the translucent characters, scenery, and style all harmonize with
the yearning of the narrator in an “identity of situation and
feeling."1314Nowhere and William Guest are two polar aspects
of Morris the author—the healing, achieved hope and the
wounded, hoping subject. Both the subject and his hope are in
some ways marked by Pre-Raphaelite narcissism and thus very
much at odds with modern taste. But the sensual immediacy and
clarity of their interaction render with great fidelity and
economy a genuine poetry of human beauty and transience. The
characters are ranged along a graduated spectrum which ex­
tends from the clouded narrator to Ellen, the personification of
sunshine loveliness. Nearest to Guest are Old Hammond with his
knowledge of, and the "grumbler” with his eccentric penchant
for, the old-time unhappiness, while the fulcrum of the nar­
ration is occupied by Dick and Clara. Since this is “a land of fel­
lowship rather than authority, there are no fathers: a genera­
tion is always skipped.But all characters are mirror-images
of the narrator (Old Hammond) or of the landscape, and all
elements of the story a system of stylistic mirrors which would
easily become tedious were it not for the fundamental existential
estrangement and opposition between Nowhere and England,
the twenty-first and the nineteenth century, light and soot, sum­
mer and winter. The narration glides in a leisurely manner
among these clarifying mirrors, progressing from Guest’s first
immersions into the Thames of the future and the deurbanized
London to the explanation of history, the beauty of the river
journey, and—since he cannot be in at the fruition—his final

13. Raymond Williams, Culture and Society 1780-1950 (Bibliography IV C), p. 234.
14. Tom Middlebro' (Bibliography IV C), p. 10.
ANTICIPATING THE SL'NBL'RST 187

expulsion from the harvest celebration.


The horizons of News From Nowhere are a variant of Marxism,
with a bias toward Fourier’s passionate attraction for work and
pleasure but without his systematization. Human history is seen
as a dialectical development from tribal communism, or from
Morris's beloved Middle Ages, through capitalism to classless
society, “from the older imperfect communal period, through
the time of the confused struggle and tyranny of the rights
of property, into the present rest and happiness of complete
Communism” (chap. 27). The chapter “How the Change Came”
extrapolates from the experience of French revolutions and Eng­
lish working-class agitation, such as the Bloody Sunday demon­
stration of 1887, a first approximation to realistic revolution in
SF. There are also shrewd hints about the transitional period
after the revolution. True, the resulting life, in which mathemat­
ics’ is an eccentric foible on a par with antiquarian novels and
education is left to haphazard communion with the society of
people and things, is in many important aspects a multiplication
ad infinitum of ideals from Morris's arts and crafts circle. How­
ever, if he sees life somewhat too exclusively as a Pre-Raphaelite
work of art, at least Morris went to the logical end of his genera­
tion’s demand that life should become a work of art. He took it
seriously, that is. literally and collectively, and tried to depict its
realization. If the attempt was not wholly successful because of
Morris’s well-founded but one-sided distrust of science, still the
further horizons of such life are open-ended. Like any Golden
Age or Eden in or after Morris’s favorites Homer, Hesiod, and
the Bible, this is a static society. But in Morris’s scheme of history
it is explicitly an epoch of cleansing rest, which might well evolve
further.
1.3, News From Nowhere is an alternative not only to what
Morris felt as mawkish bourgeois novels and as the technocracy
of Looking Backward. In the spectrum of SF it is also an alter­
native to both of the contemporary basic variants of the “post­
catastrophe tale”—W. H. Hudson's merely escapist and idio­
syncratic pastoral and Richard Jefferies’s embracing of "hard”
barbarism and primitivism. Jefferies’s After London (1885) intro­
duced a fall of civilization patterned after that of the Roman
Empire, which results in a return to the barbarian—part feudal
and part slave-owning—social order. In his novel the reasons
188 ANTICIPATING THE SUNBURST

for the catastrophe are unclear—possibly economic, perhaps


cosmic—but in such works reasons are as a rule secondary to the
middle classes’ sense of impending doom and wish-dream of a
new start. The reassertion of wild life which Jefferies renders
with peculiar intensity, the legends about “the ancients” and
their knowledge, the poisonous site of the old metropolis, and
the new geopolitics superimposed on old maps subsequently be­
came the staple of a whole range of SF. Hudson’s mellifluous
A Crystal Age (1887) introduced the other timeless simplification of
pastoral, here recomplicated by changed sexual behavior in a
beehive-type matriarchal family. In a reversal of Bulwer, the
fever of sexual passion is equated with the individualist civiliza­
tion, but it is again both a sign of sickness and indispensably
sweet to the protagonist. Both of these works have some ele­
ments similar to Morris’s, since they totally reject the present and
the city. Morris read After London before he wrote his book and
was stimulated to “absurd hopes”15 by its picture of de urbaniza­
tion. But his romance is a third way, transcending the opposition
between Bellamy’s ethicoreligious pacification and Jefferies’s
politicogeological devolution, as well as that between escapism a
la Hudson and naturalistic sentimentality. Guest and his hosts
are obscurely conscious of meeting “as if I were a being from
another planet” (chap. 9); but he is also the link between the
universes of darkness and sunlight, and Morris overcomes the
one-sidedness of these various traditions by a blend of veri­
similitude and Earthly Paradise, by a future sunlight constantly
contrasting with our darkness—as befits a dawm.
Finally, Morris’s underlying view' of world and man is simply
and beautifully but inexorably materialistic. Though there is no
immortality in Nowhere—the only feature of Earthly Paradise
not incorporated—death and sorrow, as in the episode of the
jealousy killing, do not destroy but confirm the paradise. For in
this view the individual is bound up with his fellow's and nature
in an existence that has wholly eradicated the social and cosmic
alienation of man. Morris “seems to retire far from the real
world and to build a world out of his wishes; but when he has
finished the result stands out as a picture of experience inelucta­

15. Quoted (from Morris’s letter to Mrs. Burne-Jones of 28 April 1885) in A. L.


Morton (Bibliography II), p, 204,
ANTICIPATING THE SUNBURST 189

bly true.”16 In Bellamy’s romance the new vision evolved sys­


tematically out of facers of the old; his “colder” political stance is
accompanied bv a closed and often oppressive narrative struc­
ture. In Morris's romance the new vision as a whole is incompat­
ible with the old; its open and airier structure is homologous to
his warmer, nonregimenting politics. As we have seen, Bellamy’s
vision achieved therefore within his book a reality equal to that of
the author's empirical environment; but Morris’s achieved an
“ineluctable” reality superior to that of the old civilization. That is
why his narrator, tragically marked by the old, must in the end
be extruded from the vision.
Let us compare one representative feature: Dr. Leete’s private
room in the communal dining house stands for Bellamy’s gen­
eral treatment of the public whole as a sum of rationalized and
sanitized private elements, no doubt spatially transposed and re­
grouped but qualitatively unchanged. It is a dining room for a
monogamous family and its private guests, just as the speech,
furniture, dress, maidenly blushes, and the like—in short, the
whole lifestyle of the future Bostonians—is for all practical pur­
poses simply extrapolated from the style of “their cultured an­
cestors of the nineteenth [century]" (chap. 4). On the contrary,
Morris's dining rooms harbor truly communal feasts, open to the
abundant fertility of nature, and with a large cast of erotically
sympathetic and open “neighbours" who transmit information to
the narrator by asking him curious questions rather than simply
lecturing him, and us, as affable but omniscient teachers. The
Hammersmith Guest House, likewise erected on the site of the
narrator's nineteenth-century dwelling, is its temporal extension
not into a safer private home, as in Bellamy, but into a collective
entity that has done away with the pernicious sundering of pri­
vate and public, indoors and outdoors, beauty and utility. Mor­
ris’s dining place is a fusion of an idealized fourteenth century
and classical antiquity, open to the garden and river glimpsed
beyond: but the improved architecture, food, and flowers
have a counterpart in men and women who age at less than half
the Victorian rate. For all the borrowings from the past, such a
sweeping biological improvement is the measure of the qualita­

16. C. S. Lewis (Bibliography IV A), p. 54.


190 ANTICIPATING THE SUNBURST

tive difference between Nowhere and any, however “cultured,”


nineteenth-century lifestyle.
Furthermore, the whole narrative of Looking Backward pro­
gresses as a retrospective series of West’s topographical and
ideological sallies into the new Boston from the individual,
monogamous hearth of the Leetes and under their reassuring
guidance. Any unaccompanied personal venture front this safe
cocoon immediately provokes in West a “horror of strangeness”
(chap. 5), an existential or indeed existentialist nausea that is—
most revealingly—quite as violent in the supposedly safe new
Boston of the future as in the nightmare of returning to the
competitive old Boston of the past. This microcosm, consisting of
a very restricted number of spaces and characters, imparts a
strong agoraphobic aura to Bellamy's millennium: it harbors a
panic fear, for which only the closure of space, of ideas (State
Socialism), and of the narration itself can provide a remedy. The
underlying metaphoric cluster of his book is one of static heal­
ing, whereas in Morris’s book it is one of dynamic observation
during a journey. That is why, though Bellamy came w'ithin an
ace of returning his narrator to the nineteenth century to work
in his own epoch for his new vision, and furthermore made it
clear that this would have been the ethically proper course to fol­
low, it was left for the libertarian communist Morris, with his less
hidebound readership, to actually effect this large step. The su­
preme sacrament of acceptance into Bellamy's society is a mysti­
cally subromautic marriage into which the narrator once and for
all escapes, in a sentimental happy ending of ethical—rather
than political—salvation. Quite homologously, Bellamy’s fear of
existential openness unshielded by a personal savior or vertical
hierarchy is also the motivation for his ideological stance, for
example, that in favor of strict industrial organization and
against a forcible political revolution: in utopian writings, politics
are based on the author’s simultaneously deeply personal and
deeply classbound psychology. Thus Morris’s novel not only
more than doubles the number of characters (two main women
and two main guides instead of one each in Bellamy—plus a
great number of subsidiary characters instead of the lone Mrs.
Leete and some disembodied voices and faces); it also enriches
the times, spaces, and overall complexity of their relationships.
In brief, Morris transcends Bellamy's model of fraternity under
ANTICIPATING THE SUNBURST 191

the “fatherhood of God” (chap. 26) and of lay elders (the alumni,
the father-figure of Dr. Leete) in favor of the youthful, self-
governing, and as it were parthenogenetic model of potential
lovers. Where Bellamy opts for a psychological repression of
self-determination, equally of the workers at their working place
and of sexual relationships (demurely identical to those in the
contemporary sentimental novel), Morris opts for an extension
of sympathy or libido to the whole of the gardenlike nature, a
sinless Earthly Paradise. The supreme sacrament of acceptance
into his society is, therefore, not sentimentality but the actual
journeying and working together, as far as is realistically feasible.
Phis is not to belittle the achievements of Bellamy or to ignore
the gaps in Morris. Both writers are deeply committed to an
anguished distancing from nineteenth-century capitalism and to
a different life. However, following the main US tradition, Bel­
lamy's “scheme was arithmetic and comfort,”*7 and it resulted
mainly in a sentimental dream and a tight and earnest embracing
of security, where anguish is discharged upon a series of personal
mediators, w’hereas Morris’s journeying results mainly in a
painterly vision and an attempt at direct creativity, which, being
open-ended, is inseparable from a possible anguish to be re­
solved only in self-determined practice or praxis. Yet, in other
ways, the dreams or visions of Bellamy and Morris can also be
treated as complementary: there is, finally, no need to make an
exclusive choice between them. The paradox of Looking Backward
being both more limited than and yet complementary to News
From Nowhere is finally the paradox of Christian Socialism itself,
simultaneously committed to the patriarchal vertical of the
“fatherhood of God" and to the libertarian horizontal of the
“brotherhood of Man.” Such conflicting Protestant and middle­
class abstractions are resolved by Moms: radically careless of the
fatherhood, he explores the meaning and price of brotherhood
in terms of an intimate neighborliness.
Accordingly, it is not discrete scenes of estrangement and par­
ables that stand out in News From Nowhere, as they do in Bellamy.
Learning from him, Morris also provides a few such scenes: the
phantasmagoric vision of Bloody Sunday superimposed upon

J 7. Emerson’s note (in Journals 5: 473-74, quoted in Vernon Louis Parrington


[Bibliography IV Aj, 2:349) on rhe Brook Farm project.
192 ANTICIPATING the sunburst

the sight of the orchard leading to “the Parliament House, or


Dung Market" (chap. 7); the shocking final recognition of the
dark cloud and the servile men of the nineteenth century. But it
is the tone of the whole vision-dream—the book-length parable
of new human relations in a society of "wealthy freemen” (chap.
14), beauty, and "free exercise of the senses and passions of a
healthy human animal, so far as this [does] not injure the other
individuals of the community” (Morris, “The Society of the
Future”)—that remains with the reader. It is the historical hori­
zon, the spectacle of people who "pass [their] lives in a reason­
able strife with nature, exercising not one side of [themselves]
only, but all sides, taking the keenest pleasure in all the life of
the world” (chap. 9), in counterpoint with a Marxist “optimistic
tragedy” borne bv the narrator bereft of such a life, that gives
News From Nowhere its bittersweet, tensile strength. Within a
well-defined, deep but narrow sensibility, its dialectics of con­
sciousness and unconsciousness establishes an Earthly Paradise
more real and more human than the reader’s tawdry actuality. If
Morris’s romance harkens back to almost animistic elements, it
does so as the crown of a plebeian tradition of legends and
folktales. Morris could have claimed for himself, in Fourier’s
phrase, that unlike the best political economists who wanted to
throw light on the chaos, he wanted to lead out of it. Though
News From Nowhere only partly escapes the weakness of utopias—
their abstractness—it fully shares their strength, which lies “in the
ineluctable and the absolute.’’1B And even the abstractness is over­
come in Morris’s late essays, his crowning and truly utopian
works such as “How We Live and How We Might Live,” “The
Society of the Future," "How I Became a Socialist,” and "Com­
munism." In them he even accepted the cognitive necessity of
(as he wrote in “Communism”) “time [teaching] us what new
machinery may be necessary to the new life.” This is Morris’s
final marriage of art with history.
Morris bequeathed to SF several key elements. He endowed
Bellamy’s suffering narrator in the new country with philosophi­
cal and poetic value. He transferred a believable revolution from
political tracts into fiction, fathering a line that stretches from

18. Ernst Bloch (Bibliography It), 1:679: from here. 1:677, is also the paraphrase
of Fourier’s dictum.
ANTICIPATING THE St'N BURST 193

Wells’s The Sleeper Wakes through Jack London, Alexei Tolstoy,


G. B. Shaw and Robert A. Heinlein to the flood of SF revolts in
the last 40 years. His utopian pastoral or Earthly Paradise has
had less success than Jefferies’s neobarbarism or Hudson’s titil­
lating escapism, though it can be felt as the endangered alterna­
tive from Wells’s Eloi to C. S. Lewis’s Venus or Le Guin’s New
Tahiti (in The Word for World Is Forest). But his dialectical, tragic,
and victorious Epicurean socialism remains the mature horizon
of all SF drawing upon hopes of an open future for human
beings and for the Earth. No one has yet surpassed Morris in his
intimate understanding that “times of change, disruption, and
revolution are naturally times of hope also” (“The Hopes of
Civilization”). No one in nineteenth-century SF, and few outside
it, conveyed this understanding in such lucid and warm prose.
2.1. Mark Twain's A Connecticut Yankee in King Arthurs Court
(1889) can be compared in very revealing ways with Looking
Backward (and News From Nowhere) responding as it does to the
same acute social dilemma which powered the anticipation
stories. It too is "a philosophical fable which sets forth a theory
of capitalism and an interpretation of the historical process that
has brought it into being.”’9 Twain too confronts two historical
epochs by means of a narrator-commentator from the author’s
epoch, a stranger to the epoch he is strangely projected into and
which finally ejects him (in Bellamy this ejection is then reversed,
proving to be a dream, as it is in a way in “Twain’s” postscript).
But instead of finding himself in a bright future, Twain’s Hank
Morgan arrives in the darkest Middle Ages. Bellamy’s story was
presented as a new type of hortative historical romance written
in and in praise of an estranged utopian future; Twain’s is pre­
sented as a donated manuscript, but it uses the old historical and
exotic novel for subversive SF purposes, which include a bitter
debunking of Scott's, Cooper’s, and Tennyson’s sentimentality
toward feudalism and savagery. Shifting the standpoint from
Bellamy’s utopian future to a common-sense Yankee present,
this new type of epoch-collision looks backward at the author’s
past instead of at his present. But the backward glance discovers
in both cases a dark dystopia: “In Bellamy’s analysis contempo­
rary America, a Yankee phenomenon, was as benighted and

19. Henry X’ash Smith. Stark Twain'; Fablr of Progress (Bibliography IV C), p. 39.
194 ANTICIPATING THE SUNBURST

brutalized as Athur’s England. The American labourer was


scarcely better off than the chained slaves, in A Connecticut Yan­
kee, driven to market in London."®0 However, Twain’s Hank has
to live in the midst of that brutalized epoch, not merely to judge
it at leisure. Unlike Bellamy’s hero, who is accepted into the
security of a changed and better world, or Morris’s hero, who
finds in the new, future world enough tragic optimism to return
fortified for his struggle inside and against the old, past world,
Hank sets about changing his adoptive bad old world. An out*
sider activist, he intervenes in the affairs of the sixth century in
the name of the nineteenth.
However, the nineteenth-century values in the name of which
he intervenes are deeply contradictory and finally frustrating.
On the one hand Hank is an engineering foreman, a convinced
democratic ideologue, radical to the point of Jacobin terror:
“When he snatched up the banners under which the middleclass
was forcing the nobility to disgorge, he was eloquently sincere;
his flaming calls to revolt against self-appointed masters are great
statements of that right . . . [to] self-respecting manliness and
political equality."20
21 On the other, he is a thoroughgoing bour­
geois Individualist and businessman—as he says, “just another
Robinson Crusoe cast away on an uninhabited island, with no
society but some more or less tame animals, and if I wanted to
make life bearable I must do as he did—invent, contrive, create
.. (chap. 7). Hank starts his reforms as “the Baconian utilitarian
and progressive, the Whig bourgeois ’22—by opening a patent
office; his idea of creation is patterned after his patron saints
“Gutenberg, Watt, Arkwright, Whitney, Morse, Stephenson,
Bell," the patent-office giants of the capitalist industrial and
communicational revolution (chap. 33). His next enterprises are
special schools, somewhat chillingly called Man-Factories, and
frontier-type sensational newspapers accepting and dispensing
political and economic patronage; then come factories, “iron and
steel missionaries of my future civilization” (chap. 10). Hank
Morgan reenacts thus the historical ascent of the bourgeois class
both politically and paleo technically, and he memorably typifies

20. Justin Kaplan (Bibliography IV C), p. 16.


21. Louis Budd (Bibliography IV C}, p. 144.
22. Roger B. Salomon (Bibliography IV C), pp. 30-31.
ANTICIPATING TH£ SUNBURST 195

its Yankee variant—a blend of political radicalism, go-getting


commercialism, profitable showmanship, and technological
gadgetry, issuing finally in ruinous stock market speculation. But
he also shares the historical fate of the Yankee bourgeoisie as felt
by Mark Twain in the Gilded Age: a separation from the labor­
ing people.
The first part of A Connecticut Yankee is taken up with Hank’s
orientation at the court of Camelot. It concentrates on hearty
lampooning of Malory’s and Tennyson’s Arthurian styles and
ideologies and on Hank’s use of scientific knowledge as a bur­
lesque magic superior to that of his rival Merlin. Having become
on the strength of that magic “Sir Boss,” a reformist prime
minister to Arthur, Hank travels through the country on an
obligatory knightly mission. In this second part there is added to
the burlesque of knight-errantry and miraculous religion and to
technical spoofs a democratic indignation at the oppressive laws
of feudalism. In the third part the Yankee and the king travel
incognito in the style of Harun-al-Rashid, plumbing the depths
of slavery and demoralization in the country. This is the crucial
juncture for a supposedly radical and a quasi-historical novel.
Wanting to overthrow the Dark Ages, the Yankee has to find out
whether the people would follow him. This hegemony had in
fact been the historical achievement of the bourgeoisie up to “the
ever-memorable and blessed [French] Revolution.” But Hank
cannot achieve this alliance, because Twain quite realistically
does not believe in it any more: “I knew that the Jack Cade or
Wat Tyler who tries [a revolution] without first educating his
materials up to revolution-grade is almost absolutely certain to
gel left” (chap. 13). It is difficult to see how Hank could proceed
with this education since he is continually envisaging the people
as commodities or animals, a view endorsed by the author as he
focuses on the snobbery, pusillanimity, and finally mob lynchings
of an environment reminiscent of the antebellum South with its
chain gangs and poor whites. The Yankee had fulminated
against a class training that blinds the people. Yet he comes to act
as a manufacturer, tamer-trainer, or indeed a being of superior
race among “human muck" (chap. 43). These are hardly becom­
ing roles for a democrat setting the stage for freemen. Faithfully
the author shows us that, in spite of all his sententiousness, the
Yankee in the end “gets left” high and dry by a receding tide of
196 ANTICIPATING THE SUNBURST

history. Not seeing beyond the ebb of bourgeois democracy and


revolutionism, it is Mark Twain as much as Hank Morgan who
concludes that "there are times when one would like to hang the
whole human race and finish the farce” (chap. 31).
The indignation of the first half of A Connecticut Yankee against
feudal and church oppression was closer to Swift than to the
lukewarm iconoclasm of Erewhon. But Swift’s satire had been so
savagely efficient because he kept the authorial distance from
Gulliver subtly flexible and yet precisely controlled, correlative to
a radical pessimism about what Twain himself was to call the
damned human race. In this second half of the novel, Twain’s
attempt to do the same in relation to Hank fails. This narrative
cannot sustain Swift’s singleminded control over a comicosatirical
narrator, who travels within a didactic, estranged country, be­
cause A Connecticut Yankee oscillates between commitment to a
historical ideal and horror at its workings in history. Early on the
novel affirms the progressive theory of history that emerged
from the heyday of bourgeois revolutions and a positive hero
upholding it. But this came to clash with Twain’s increasing
alienation from the effects of the industrial revolution as ap­
propriated by the bourgeoisie and his consequent pessimistic
theory of human nature, and the book was left without a moral
and political core—which is fatal equally for satire and utopia.
Sundered from the people, all the Yankee can do is train a small
band of elite technocratic enthusiasts, whose program is opposed
to the Arthurian Age as the Yankee Gilded Age exploitation of
the frontier, or of any new market, is to Southern slavery: “Look
at the opportunities here for a man of knowledge, brains, pluck,
and enterprise to sail in and grow with the country. The grand­
est field that ever was; and all my own; not a competitor . . .”
(chap. 8). Their main trump—historically exact again—is science
and technology.
Twain’s attitude to technological progress oscillates into confu­
sion and despair in strict parallel to his attitudes toward its his­
torical bearer, the middle class, and especially the technics-
committed Yankee bourgeoisie. Hank’s superior know-how is
shown first as spoofs, ambiguously magical tricks used as a joyous
means to power. The degradation of science and technology to
the role of magic, a fundamentally different attempt at control­
ling nature, leads to their becoming a juggernaut and, finally,
ANTICIPATING THE SUNBURST 197

means which turn into their own end—the end being mass car­
nage. The Yankee’s societal counterconscruction ends in an Ar­
mageddon, prefigured in the book’s volcanic and explosive im­
agery.
Though Twain's stance as outsider does not do justice to the
liberating possibilities of science, it enables him to pass a shrewd
judgment on its historical sociopolitical uses. Sundered from the
artisans and peasants, “Jack Cade or Wat Tyler,’’ Baconian sci­
ence is able only to destroy impartially the upper classes and its
wielders. Finally, even this potent means is only a theatrical “ef­
fect” of Barnumian proportions, effecting no social change. The
new Crusoe-type, Individualistic civilization collapses under the
interdict of an omnipotent Catholic Church that we have never
really seen in A Connecticut Yankee, yet that pops up as a diabolus
ex machina, a fit antagonist for the lone Protestant Great Man or
Whig Robinson Crusoe. As Twain’s superficial treatment of
Arthur’s court—culminating in the faceless knights of the final
battle—testifies, feudalism was not a believable antagonist in the
American 1880s. Nor was—at a time when Twain was defend­
ing the young trade unions and Howrells was agonizing over the
judicial murder of the Chicago anarchists—a robber-baron
bourgeoisie a believable protagonist. Indeed, as Howells noted,
“there are passages in which we see that the noble of Arthur’s
day, who battened on the blood and sweat of his bondsmen,
is one in essence with the capitalist of [1889] who grows rich
on the labor of his underpaid wagemen,’'23 and Hank’s prog­
ress through England ends in an almost Dickensian, or indeed
Blakean, horrible London. Thus the book collapses in a rather
perfunctory mixture of shadow-boxing and savage despair,
pronouncing “a curse on both parts.”24
This devolution of the story’s hopes for a plebeian progres­
sivism is also embodied in its language. In the comical, confident
beginning of the novel, Hank’s “machine-shop lingo collides with
the Malory-ese of the Age of Chivalry.” By the end, it has be­
come clear that the Yankee’s “language [is grounded] in cliches
and conventional syntax, [and] its character emerges by means of

23. W. D. Howells, review in Harper's (January 1890), p. 320, quoted in Henry Nash
Smith, Mark Twain: The Development of a Writer (Bibliography IV C), p. 146,
24. Kaplan, p. 19.
198 ANTICIPATING THE SUNBURST

exaggeration and calculated vulgarity,” that it is an artificial ver­


nacular of an artificial democrat, only “masquerading as burly,
rough talk.” This means that it attains neither consistently pre­
cise satire nor consistently wholehearted burlesque, but that in
final analysis it remains “a show, an act . . . not necessary to the
action, but simply decoration . . , [nothing] more than one of
Hank Morgan’s effects."2^
Twain’s liberal utopianism seems to have followed a course
quite parallel to Jules Verne's—a writer whose interest in map­
ping and clean communications he shares in this book with its
providential telephone wires and industries without a working
class. Twain's fabuiation relies, as does Verne, on the bedrock of
popular melodrama and farce, or on travelers’ tales (such as
balloon travel in Tom Sawyer Abroad, 1894). As it did with Verne,
the dimming of liberal horizons turned Twain toward a cyclical
theory of history (compare his Papers of the Adam Family [bp.
1962] and Verne’s Eternal Adam) and superdestructive weaponry,
which together account for the breakdown of a class society rent
by antagonisms in A Connecticut Yankee as in Verne’s Propeller
Island, On the other hand, Twain was quite alien to any idea of a
new social system going beyond the bourgeois-democratic one.
Though he privately much admired Bellamy’s sunburst of a
wholly different state of affairs, he could creatively envisage it
only as the glare of Armageddon, of the “fifty electric suns” by
whose light the final Battle of the Sand-Belt is fought, with its
ominous hint of our atomic bomb brighter than a thousand suns.
In Twain, as in the Mary Shelley-to-Verne tradition, the only
novelties are destructive ones; in opposition to the socialist tradi­
tion, culminating in Morris, the imagery of moon and sun,
eclipse and displacement, leads to catastrophe instead of millen­
nium, to violent nightmare instead of vision and dream of restful
dawn.
However, Twain is much more profoundly distressed than
Verne by the closure of progressive horizons. This distress issues
in a deep ambivalence toward not only the values but the reality
itself of the opposed epochs. The dreamy Arthurian Britain may
be the antebellum South of slavery, but it has likewise a pastoral,
childhood freshness. Obversely, the energetic Yankee represents

25. James M. Cox (Bibliography IV C). pp. 203. 215. 213, and 219.
ANTICIPATING THE SUNBURST 199

the capitalist North of the nineteenth century both as political


liberator and industrial destroyer. In the end, Twain’s hero
therefore sheds his representative, allegorical class traits and
takes a double refuge: first from the sixth century—where the
original “white Indians” have turned into sophisticated stock ex­
change scalpers—into the ultimate reality of private life and fam­
ily, and then from the nineteenth century into dreams of such a
privatized sixth century. Finally, in the “plague on both your
houses” situation, when annihilation has disposed of both sides,
Hank becomes adrift in time and withdraws into a dreaming
whollv outside the catastrophic history. Indeed, there is a hint
that he is not quite certain which is a dream, the sixth or the
nineteenth century. For all his genuine radical leanings, Mark
Twain is to be ranged on the other side from Bellamy and
Morris across the watershed of hope. This historical position led
even so powerful a writer to a formal dead end.
Thus, while there are richly burlesque or satirical as well as
generously indignant passages in A Connecticut Yankee, its final
horizon of havoc and sterility is what makes it both memorable
and indicative for a long-lasting historical moment. The activist­
hero battling against the age will recur in SF, from the pages of
Wells and London to innumerable later time travelers and politi­
cal plotters. But Twain was the first in SF to face directly, with­
out theological or biological myths, how “in bourgeois society . . .
the past dominates the present,"26 and the first to analyze rather
than merely present the obscure forces of history that rise from
the past not only to overwhelm the Promethean progressive hero
from the outside, but also to hollow him out from the inside.
Hank's career could be called “Prometheus Re-Bound” or “the
making of Victor Frankenstein,” for it shows how the bearer of
Promethean progress turns into the absurd causer of historical
catastrophe. Twain thus revived for the modern novel the cen­
tral ideological dilemmas of the age of anticipation, which would
be developed in twentieth-century SF because they ivould remain
the most sensitive historical problems. For example, the motif of
outside intervention or exporting of revolution into an “under­
developed” country will recur up to the Strugatskys and Le

26. Karl Marx and Friedrich Engels, “Manifesto of the Communist Parly." in Selected
H'orfc in One Volume (New York. 1968). p. 48.
200 ANTICIPATING THE SUNBURST

Guin, and the concomitant or obverse motif of the lone hero


becoming adrift in time will accompany the devolution of history
from Wells to Vonnegut and Sheckley.
A Connecticut Yankee does not handle such seminal motifs and
types in a fully satisfactory way. Its "juvenile fantasy” treatment
of the “collision between superstition and modern technology”21
and of politics reduced to a duel between the lone protagonist
and the world crops up, increasingly, as a cliché of space-opera
and other “degree zero” SF. Perhaps the causally and formally
central failing of A Connecticut Yankee, the one that made its
structure analogous to juvenile fantasy, was the absence of what
one might call a political economics of existence from Hank’s
social changes. It remains completely unclear whether and how
the laboring people that the Yankee meets on his peregrinations
have benefited or will benefit from his despotic industrialism or
State capitalism. The ads displayed on knights imply a market
economy, other passages imply unpaid distribution; but the
technical problems of political economy are not so important as
the fact that nobody in the novel even poses the question how
people would receive their sustenance in a post-Arthurian indus­
trialized system, or who would determine its amount and dis­
tribution channels. This is one of the sorriest blind spots in the
center of the imaginative picture that Twain has bequeathed to
practically the whole of subsequent English language SF, born of
the selfsame “New Deal” hope which took its name from a
phrase in this text. But finally, by the same token, this novel
shows the necessary collapse of Twain’s own and all such private
fantasies, and provides the means by which to identify them as a
flight from history. A Connecticut Yankee's function as yardstick
for the tradition it set in motion is also an indication of the
novel’s importance.
Many other SF fragments show Mark Twain’s concern with the
grimness of the coming times in which America’s uniqueness
would fall prey to the general fate of imperialist power—a suspi­
cion that grew into certainty after the 1898 Spanish-American
War for the “open door” to Latin America and East Asia. Twain
returned frequently in his fragmentary sketches to the image of
a future dictatorship—monarchist, technological, or theologi-

27. Smith, Mark Twain: Devriojment, p. 166.


ANTICIPATING THE SUNBURST 201

cal—in the United States, establishing in this way too a central


theme for SF. Had these fragments been completed and pub­
lished, he would have beyond a doubt stood instead of Wells as
the major turning point in the tradition leading to modern SF,
and instead of Stapled on as the inventor of fictional historiog­
raphy. But even without these fragments there is sufficient evi­
dence that strange places and states fascinated him, as witness
“The Curious Republic of Gondour,” “The Comedy of Those
Extraordinary Twins” (progenitors of the mutant twins that
recur from Heinlein to Dick), or “From the London Times of
1904” (where even extraordinary technology cannot prevail
against bourgeois law). The gloomy historical horizons were
transferred to the theme of other dimensions of life that might
perhaps be as real as ours in a number of SF stories culminating
in the fragmentary masterpieces of “The Great Dark" (bp. 1962)
and “Three Thousand Years Among the Microbes” (bp. 1966).
In the former, Edgar Allan Poe’s eerie exotic voyage combines
with Fitz-James O’Brien’s life in a drop under the microscope as
the setting for a multiple inversion of normal parameters such as
size, duration, and memory; the story would have constituted a
haunting parable of man's insecurity in the itinerary of life. In
the latter, the microbic microcosm becomes the scene of what
approaches a consistent satirical epic. More importantly, in this
late, testamentary text the satire often leaves behind the hesita­
tions caused both by Twain's earlier illusions and by his self­
censorship in order to respect the propriety his bourgeois read­
ership demanded, and rises to truly Swiftian bitterness and rele­
vance. It is the mark of Twain’s stature that he was the first SF
writer able to respond to the cruel times by exploring Swift’s
implicit query “What is Man?” in fully explicit, destructive terms.
Nonetheless, the completed Connecticut Yankee, which joins life’s
cruel and insecure itinerary to the movement of history, remains
Tw-ain’s major contribution to SF. The deep existential distress
and even epistemological puzzles of the individual in high
capitalism were already acknowledged in—were indeed the mo­
tive force of—Bellamy’s and Morris’s utopias, but they were there
superseded by the security or the beauty of human relations in
the new community. By disjoining the individual hero from the
bright communal novum while continuing to uphold the abso­
lute necessity and value of the utopian supersession, Twain
202 ANTICIPATING THE SUNBURST

situated his novel at the wellspring of rhe preoccupation that was


to dominate the next century of significant SF, from Wells and
Zamyatin to Lem and Dick.
2.2. Another way out of such an existential distress was to
transfer utopian yearning out of material history, into an incon­
gruous physico-religious heaven, as did Camille Flammarion. In
the last third of the nineteenth century he tirelessly ground out a
great number of nonfictional books ranging from astronomical
and geographic popularizations to arguments for a scientifically
proved psychic life after death. Some of his other works—Lumen,
Rêves étoilés, or Uranie—were an indigested mixture of reincarna­
tion on planets of our solar system or of distant galaxies and
supposed validation of the spiritual forces entailed by the mar­
vels of modern cosmology’ and electricity, in which regard they
were not unlike writings of the contemporary occultist wave in
Britain and the USA, from Bulwer to Mrs. Blavatsky and thence
to bestsellers like the novels of Marie Corelli. Flammarion’s
rather frenzied anti-materialism vented itself in pet peeves
against such indignities as material food-taking and in insuffer­
able passages detailing (literally) ethereal love encounters. Having
written the first modern survey of literature on imaginary
worlds, Les Mondes imaginaires et Us mondes réels (1865), he blithely
pillaged that whole tradition from Cyrano through Grainville
and Fourier to Defontenay for bizarre figures and incidents of
life on other planets which he inserted into his long lectures and
tirades. The life forms were either anthropomorphic or only
slightly changed—by being flying creatures, or androgynes, or by
having 17 heavenly senses (including, of course, electric, ul­
traviolet, and psychic ones)—or they could be quite different:
asexual, phosphorescent, telepathic, vegetable, or mineral. This
was heady brew for Victorians and had an immense influence
throughout the world by the 1880s. For late nineteenth-century
fantasy and SF it functioned as a repository of ideas and
topoi—much as Stapledon was to function for the SF after 1930,
but without his intelligence and tragic humanism. Flammarion’s
variations on Kant’s idea that the physical and moral perfection
of psychozoa is proportional to the planet’s distance from the
Sun contributed to a spate of spiritualist or utopian ideals being
reached by interplanetary voyages to Mars, Jupiter, or Saturn as
brainless as his were but more straightforwardly fictional, as in
ANTICIPATING THE SUNBURST 203
4

Robert Cromie’s A Plunge Into Space (1890) or John Jacob Astor's


A Journey in Other Worlds (1894). When an at least partially consis­
tent picture of new social institutions was accompanied by satire
of Earth customs, the result was slightly better, as in the
anonymous presentation of cooperative and progressive Politics
and Life in Mars (1883) or Robert D. Braine’s attempt at mixing
Flam mar ion with Plato in Messages From Mars (1892?). The
nearest Flammarion himself got to coherent books of fiction—
which was not very' near—was in his La Fin du Monde (1893,
translated as Omega), which included again ati anthologic review
of possible endings of the world and anticipations of a slightly
technocratic future ending after ten million years in an ice-death
of the human race, whose souls, however, flit on to better things
on Jupiter. The book is a good example of the strange feedback
between Flammarion and Verne, in which the former stressed
“far out" psychic phenomena and the latter a near time and
space but, by the last quarter of the century, both shared a simi­
lar pessimism toward the horizons of mundane history.
2.3. Some writers in the 1880s and 90s paralleled or echoed
Bellamy’s and Morris’s concern with a better political social fu­
ture; the intelligent anti-revolutionary' account of the collapse of
an oppressive State Communism in W. A. Watlock’s The Next
'Ninety Three (1886) or the pleas for women's equality in George
Noyes Miller’s The Strike of a Sex (1883?) and in the rather melo­
dramatic Gloriana of Lady Florence Dixie (1890) can be adduced
as examples with a certain interest. But in the dramatically ex­
panding production of SF—which, for example, doubled in Brit­
ain each decade, from approximately 45 book publications in the
1870s to some 90 in the 1880s and 170 in the 1890s—the great
majority of works up to Wells remained immature. To take an
example from the United States, the two poles that Mark Twain
attempted to fuse in Connecticut Yankee can be seen sundered and
impoverished in the shallow optimism of Stockton and the
monotonous horrors of Bierce. Frank Stockton wrote several
Vernean technological adventure tales. The Great War Syndicate
(1889) transferred the future war subgenre to the United States.
A capitalist syndicate, hired by the American government, wins a
war by means of superweapons, including a tremendous rocket
missile, whereupon the vanquished Britons join in an Anglo-
American domination of rhe world—a revealing story by a popu-
204 ANTICIPATING THE SUNBURST

lar writer. Ambrose Bierce is a weightier writer in the psychofan-


tastic tradition of Poe, on the borders of SF and fantasy. Some
of his stories, such as “The Damned Thing,” motivate invisible
beasts and similar horrors rationally, while “Moxon’s Master”
(both 1893) returns the Frankenstein theme to mechanical auto­
mata, paving the way for the countless bloodthirsty robots of
Gernsbackian SF. Stockton’s shallow' politics and jaunty plots as
well as Bierce’s deep ontological fears and haunting moods have
one common denominator with A Connecticut Yankee: the om­
nipresent violence. Only Wells was to render justice to this in­
creasingly menacing atmosphere, which belied the hopes for a
pure classless dawn.
4

Introduction to Newer SF History

I was thinking this globe enough till there sprang out so noiseless
around me myriads of other globes.
Wall Whitman
This survey stops at the threshold of contemporary SF, which can be
said to arise between the World Wars, after the October Revolution and
before the atomic bomb, with the modern “mass culture” of movies, radio,
and specialized magazines and paperback book-lines for commercial liter­
ary “genres“—one of the most prominent of which SF has become. The
period marked by E.R. Burroughs and Hugo Gemsback in the United
States (and some parallel developments in Germany, cut short by Nazism)
and by the influence this country has exerted, beginning in the 1930s, on
the rest of the world, was to be one not only of a huge quantitative
explosion of SF publication, distribution, and popularity—which alone
would be sufficient reason for a separate book to describe it—but also,
even more significantly, of qualitative complications in the status of
“paraliterature” which have so far not been adequately dealt with in
literary history and theory. What makes contemporary paraliterature, and
especially SF, so complicated is the sea-change it suffered in the last
couple of generations. In almost all the earlier epochs, as I have tried to
point out in my introduction to the first part of this historical overview,
there was a profound difference between the unofficial, popular or
plebeian (largely oral), culture and the official, dominant or upper-class
(usually written), culture. The cultures of these “two nations” -within each
lingtdstico-ethnic domain have, no doubt, always been connected in vari­
ous ways, from antagonistic suppression to partial permeation, but as a
rule they have been—except for such exceptional moments of cohesion as
a portion of the Elizabethan Age—sufficiently separate to preserve dis­
tinct identities. A renewal of given official, higher, or canonic Literature
and Culture came about by the ascent of earlier noncanonic forms to
canonic status together with the social class or group that was the “ideal
reader" of those forms or genres (for example, the psychological novel and
the bourgeoisie). But the complication with twentieth-century para-
Uterature—and especially SF—is, to put it baldly, briefly, and with­
out any mediation, that neither the facobin nor the Bolshevik revolutions
have accomplished their objectives, so that radically enlarged literacy and
205
206 INTRODUCTION TO NEWER SF HISTORY

economic welfare in the "North" of our planet (Europe, North America,


Japan) have coincided with and become enmeshed in the rise of im­
perialism and the welfare-warfare state. The ensuing very complicated
amalgam of suppression and permeation—occurring at different rates
and in somewhat different forms in diverse geopolitical areas—has just
begun to be identified but has not yet been properly studied.1 One might
hazard the hypothesis that in this period the domination or hegemony of
the bourgeois ideology and taste has been challenged but on the whole not
overthrown: new forms and genres rise into official culture largely at the
expense of their pristine plebeian horizons, from the ranks, not with the
ranks, of its social originators, and at the price of containment and
cooption. However this might be, it is clearly afieldfor different methodo­
logical approaches than the ones used in this book, though one of the
major aims of this book is to reach the threshold not only of contemporary
SF history but also of a methodology that would render it justice.
It could be argued that both the proliferation and the changed status of
SF begin with and at the time of H.G. Wells. I am not sure whether this
is in fact so, and at any rate new cultural epochs do not begin on a given
New Year’s Day. What is certain is that Wells’s opus could both intrinsi­
cally and in its interrelations with the popular media of his time be a
promising starting point for a simultaneously extensive and intensive
investigation. But this too would certainly demand a book unto itself. 1
have compromised with such an ideal by aiming in this section for a kind
of stereoscopic effect, which might, 1 hope, arise from the readers benevo­
lent superposition of a broad overview of Wells’s opus (attempted in
chapter 9) on a depth probe of his modelfor SF as found in the paradig­
matic ‘‘Time Machine” (attempted in chapter 10 in brief comparison to
the Morean model of earlier SF history). Since world cultures are not
synchronic, I have concluded my historical survey with two European
probes, one of a national tradition and one of an important single writer,
but both leading in their own context again to the historical and methodo­

1. The fundaments for such a study have been laid in the notes of Antonio Gramsci
available only since the war. for example. 11 materiaUsmo starieo r la filoSofia di Benedetto
Croce (Torino, 1948) or the useful compilation Letleratura e vita naiianalr (Torino, 1966);
in English partly available in The .Modern Prince and Other Wrihngs (New York, 1972) and
Selections From the Prison A’oleAouki (New York. 1975). An excellent first systematization of
theoretical achievements and problems so far can be found tn the relevant chapters of
Raymond Williams’s Marxism and Literature (London, 1977), Not only for SF, see also the
“Sociology of SF" issue of Science.Fiction Studies 4 (November 1977), with Marc Angenot's
annotated bibliography of the sociology of literature and some introductory comments of
mine enlarging on the problems touched upon in this introduction.
INTRODUCTION TO NEWER SF 207

logical thresholds for understanding contemporary SF. No doubt, even


within the period or periods chosen, it might have been possible to deal
with some more works by prominent writers—Jack London, Rosny Aîné,
or Paul Scheerbart—or with more national traditions—such as the
French one from 1900 to 1940.2 However, since this book opts for
representativeness suggesting a tradition rather than for the inclusiveness
of a formal literary history, I have remained consistent to this option.

2. I might perhaps be the more readily excused since I have contributed to some
spadework for a better understanding of Wells and London elsewhere: Darko Suvin, with
Robert M. Philmus, eds.. H. G. With and Modem Science Fiction (Bibliography V); Darko
Suvin and David Doughs, "Jack London and His Science Fiction: A Select Bibliography,”
Sdence-Fiction Studies 3 (July 1976): 181—87.
9

Wells as the Turning Point of the SF Tradition

H, G. Wells’s first and most significant SF cycle (roughly to


1904) is based on the vision of a horrible novum as the evolu­
tionary sociobiological prospect for mankind. His basic situation
is that of a destructive newness encroaching upon the tranquillity
of the Victorian environment. Often this is managed as a con­
trast between an outer framework and a story within the story.
The framework is set in surroundings as staid and familiarly
Dickensian as possible, such as the cozy study of The Time
Machine, the old antiquity shop of "The Crystal Egg,” or the
small towns and villages of southern England in The War of the
Worlds and The First Men in the Moon. With the exception of the
protagonist, who also participates in the inner story, the charac­
ters in the outer frame, representing the almost invincible inertia
and banality of prosperous bourgeois England, are reluctant to
credit the strange newness. By contrast, the inner story details
the observation of the gradual, hesitant coming to grips with an
alien superindividual force that menaces such life and its certain­
ties by behaving exactly as the bourgeois progress did in world
history—as a quite ruthless but technologically superior mode of
life. This Wellsian inversion exploits the uneasy conscience of an
imperial civilization that did not wipe out only the bison and the
dodo: “The Tasmanians, in spite of their human likeness, were
entirely swept out of existence in a war of extermination waged
by European immigrants. Are wre such apostles of mercy as to
complain if the Martians warred in the same spirit?” (The War of
the Worlds, book 1, chap. 1).
As Wells observed, the "fantastic element” or novum is “the
strange property or the strange world.”* The strange property
can be the invention that renders Griffin invisible, or, obversely,
a new way of seeing—literally, as in “The Crystal Egg," "The

I. “Preface” to Seven Famous Navels by H. G. Wells (Garden City, NY, 1934), p. vii.
208
WELLS -LS THE TURNING POINT OF THE SF TRADITION 209

Remarkable Case of Davidson's Eyes," and “ The New Ac­


celerator,” or indirectly, as the Time Machine or the Cavorite
sphere. It is always cloaked in a pseudo-scientific explanation,
the possibility of which turns out, upon closer inspection, to be
no more than a conjuring trick by the deft writer, with ‘‘precision
in the unessential and vagueness in the essential”2—the best
example being the Time Machine itself. The strange world is
elsewhen or elsewhere. It is reached by means of a strange inven­
tion or it irrupts directly into the Victorian world in the guise of
the invading Martians or the Invisible Man. But even when
Wells’s own bourgeois world is not so explicidy assaulted, the
strange novelty always reflects back on its illusions; an SF story
by Wells is intended to be “the valid realization of some disre­
garded possibility in such a way as to comment on the false se­
curities and fatuous self-satisfaction of everyday life.”3
The strange is menacing because it looms in the future of
man. Wells masterfully translates some of man’s oldest terrors—
the fear of darkness, monstrous beasts, giants and ogres, creepy
crawly insects, and Things outside the light of his campfire, out­
side tamed nature—into an evolutionary perspective that is sup­
posed to'be validated by Darwinian biology, evolutionary cos­
mology, and the fin-de-siecle sense of a historical epoch ending.
Wells, a student of T. H. Huxley, eagerly used alien and power­
ful biological species as a rod to chastize Victorian man, thus
setting up the model for all the Bug-Eyed Monsters of later
chauvinistic SF. But the most memorable of those aliens, the
octopuslike Martians and the antlike Selenites, are identical to
“The Man of the Year Million” in one of Wells’s early articles
(alluded to in The War of the Worlds); they are emotionless higher
products of evolution judging us as we would judge insects. In
the final analysis, since Lhe aliens are a scary, alternative human
future, Wellsian space travel is an optical illusion, a variation on
his seminal model of The Time Machine. The function of his in-
terplanetarv contacts is quite different from Verne’s liberal
interest in the mechanics of locomotion within a safely homo­
geneous space. Wells is interested exclusively in the opposition
2. Unsigned review (by Basil Williams) in Athenaeum, 26 June 1897, reprinted in
Patrick Parrinder, ed. (Bibliography V). p. 57.
3. H. G. Wells, “An Experiment in Illustration,” Strand Magazine, February 1920,
quoted in Geoffrey West (Bibliography V), p. 112.
¡10 WELLS AS THE TURNING POINT OF THE SF TRADITION

between the bourgeois reader’s expectations and the strange re­


lationships found at the other end: that is why his men do land
on the Moon and his Martians on Earth.
Science is the true, demonic master of all the sorcerer’s ap­
prentices in Wells, who have—like Frankenstein or certain
folktale characters—revealed and brought about destructive
powers and monsters. From the Time Traveller through Moreau
and Griffin to Cavor, the prime character of his SF is the
scientist-ad venturer as searcher for the New, disregarding com­
mon sense and received opinion. Though powerful, since it
brings about the future, science is a hard master. Like Moreau, it
is indifferent to human suffering; like the Martians, it explodes
the nineteen th-century optimistic pretentions, liberal or socialist,
of lording it over the universe:
Science is a match that man has just got alight. He thought
he was in a room—in moments of devotion, a temple—and
that his light would be reflected from and display walls in­
scribed with wonderful secrets and pillars carved with
philosophical systems wrought into harmony. It is a curious
sensation, now that the preliminary splutter is over and the
flame burns up clear, to see his hands and just a glimpse of
himself and the patch he stands on visible, and around him,
in place of all that human comfort and beauty he antici­
pated—darkness still.4
This science is no longer, as it was for Verne, the bright noonday
certain tv of Newtonian physics. Verne protested after The First
Men in the Moon: “I make use of physics. He invents ... he
constructs ... a metal which does away with the law of gravita­
tion . . . but show me this metal.” For Wells human evolution is
an open question with two possible answers, bright and dark;
and in his first cycle darkness is the basic tonality. The cognitive
“match” by whose small light he determines his stance is Dar­
winian evolution, a flame which fitfully illumines man, his hands
(by interaction of which with the brain and the eye he evolved
from ape), and the "patch he stands on.” Therefore Wells could
much later even the score by talking about "the anticipatory in­
4. Wells, “The Rediscovery of the Unique,” The Fortnightly Review, N.S. 50 (July
1891), reprinted in Robert Philmus and David Y. Hughes, eds. (Bibliography V), pp.
30-31.
WELLS AS THE TURNING POINT OF THE SF TRADITION 211

ventions of the great Frenchman” who “told that this and that
thing could be done, which was not at that time done”—in fact,
by defining Verne as a short-term technological popularizer.5
From the point of view of a votary of physics, Wells "invents” in
the sense of inventing objective untruths. From the point of view
of the evolutionist, who does not believe in objects but in
processes—which we have only begun to elucidate—Verne is the
one who “invents" in the sense of inventing banal gadgets. For
the evolutionist, Nemo's submarine is in itself of no importance;
what matters is whether intelligent life exists on the ocean floor
(as in “In the Abyss” and “The Sea Raiders"). Accordingly,
Wells's physical and technical motivations can and do remain
quite superficial where not faked. Reacting against a mechanical
view of the world, he is ready to approach again the imaginative,
analogic veracity of Lucian’s and Swift’s story-telling centered on
strange creatures, and to call his works “romances.” Cavorite or
the Invisible Man partake more of the flying carpet and the
magic invisibility hood than of metallurgy or optics. The various
aliens represent a vigorous refashioning of the talking and sym­
bolic animals of folktale, bestiary, and fable lore into Swiftian
grotesque mirrors to man, but with the crowning collocation
within an evolutionary prospect. Since this prospect is temporal
rather than spatial, it is also much more urgent and immediate
than Swift’s controlled disgust, and a note of fairly malicious
hysteria is not absent from the ever-present violence—fires, ex­
plosions, fights, killings, and large-scale devastations—in Wells’s
SF.
The Time Machine (1895), Wells’s programmatic and (but for
the mawkish character of Weena) most consistent work, shows
his way of proceeding and his ultimate horizon. The horizon of
sociobiological regression leading to cosmic extinction, simplified
from Darwinism into a series of vivid pictures in the Eloi, the
giant crabs, and the eclipse episodes, is established by the Time
Traveller’s narration as a stark contrast to the Victorian after-
dinner discussions in his comfortable residence. The Time
Machine itself is validated by an efficient forestalling of possible
objections, put into the mouth of schematic, none too bright, and

5. Jules Verne's interview is from T. P.'s Wirkly, 9 October 1903. reprinted in Parrin-
der, ed., pp. 101-02; Wells’s rejoinder is front the "Preface" cited in note 1 above, p. vii.
J? 12 WELLS AS THE TURNING POINT OF THE SF TRADITION

reluctantly persuaded listeners, rather than by the bogus th eon


of the fourth dimension or any explanation of the gleaming bars
glimpsed in the machine. Similarly, the sequence of narrated
episodes gains much of its impact from the careful foreshorten­
ing of ever larger perspectives in an ever more breathless
rhythm (discussed at length in the following chapter). Also, the
narrator-observer's gradually deepening involvement in the Eloi
episode is marked by cognitive hypotheses that run the whole
logical gamut of sociological SF. From a parodied Morrisite
model (“Communism," says the Time Traveller at first sight)
through the discovery of degeneration and of persistence of class
divisions, he arrives at the anti-utopian form most horrifying to
the Victorians—a run-down class society ruled by a grotesque
equivalent of the nineteenth-century industrial proletariat. Char­
acteristically, the sociological perspective then blends into biol­
ogy. The laboring and upper classes are envisioned as having
developed into different races or indeed species, with the
Morlocks raising the Eloi as cattle to be eaten. In spite of a
certain contempt for their effeteness, the l ime Traveller quickly
identifies with the butterfly-like upper-class Eloi and so far for­
sakes his position as neutral observer as to engage in bloody and
fiery carnage of the repugnant spider-monkey-like Morlocks, on
the model of the most sensationalist exotic adventure stories. His
commitment is never logically argued, and there is a strong
suggestion that it flows from the social consciousness of Wells
himself, who came from the lower middle class, which lives on
the edge of the “proletarian abyss” and thus “looks upon the
proletariat as being something disgusting and evil and danger­
ous.”8 Instead, the Time Traveller’s attitude is powerfully sup­
ported by the prevailing imagery—both by animal parallels, and
by the pervasive open-air green and bright colors of the almost
Edenic garden (associated with the Eloi) opposed to the subter­
ranean blackness and the dim reddish glow (associated with the
Morlocks and the struggle against them). Later in the story these
menacing, untamed colors lead to the reddish-black eclipse,
symbolizing the end of the Earth and of the solar system. The
bright pastoral of the Eloi is gradually submerged by the en­
croaching night of the Morlocks, and the Time Traveller’s

6. Christopher Cau dwell (Bibliography V), pp. 76 and 93.


WELLS AS THE TURNING POINT OF THE SF TRADITION 21S

matches sputter out in their oppressive abyss. At the end, the


unforgettable picture of the dead world is validated by the dis­
appearance of the Time Traveller in the opaque depths of time.
Many of these devices reappear in Wells’s other major works.
The technique of domesticating the improbable by previews on a
smaller scale, employed in the vivid vanishing of the model
machine, is repeated in the introduction to the Grand Lunar
through a series of other Selenites up to Phi-oo, or to Moreau’s
bestial people through the brutal struggles in the boat and
through the ship captain, or to the Cavorite sphere’s Hight
through the experimental explosion raising the roof. The loss of
the narrator's vehicle and the ensuing panic of being a castaway
under alien rule (in The War of the Worlds this is inverted as
hiding in a trap with dwindling supplies) recurs time and again
as an effective cliff-hanger. Above all, as we will see in the next
chapter, Wells’s whole first cycle is a reversal of the popular
concept by which the lower social and biological classes were
considered as “natural" prey in the struggle for survival. In their
turn they become the predators: as laborers turn into Morlocks,
so insects, arthropods, or colonial peoples turn into Martians,
Selenites, and the like. This exalting of the humble into horrible
masters supplies a subversive shock to the bourgeois believer in
Social Darwinism; at the same time, Wells vividly testifies that a
predatory state of affairs is the only even fantastically imaginable
alternative. The world upside-down—where strange animals
hunt Man, and the subterranean lower class devours the upper
class—recurs in Wells, as in Thomas More. But whereas More’s
sheep were rendered unnatural by political economics, Wells’s
Morlocks, Beast People, and so forth, are the result of a
“natural’’ evolution from the author’s present. Nature has be­
come not only malleable—it was already becoming such in More
and particularly in Swift—but also a practically value-free cate­
gory, as in bourgeois scientism. At the end, the bourgeois
framework is shaken, but neither destroyed nor replaced by any
livable alternative. What remains is a very ambiguous attack on
liberalism from the position of “the petty bourgeois which will
either turn towards socialism or towards fascism.”7
The human/animal inversion comes openly to the fore in The

7. V. S. Pritchett (Bibliography V), p. 128.


J14 WELLS AS THE TURNING POINT OF THE SF TRADITION

Island of Dr. Moreau (1896) with admirable Swiftian stringency.


Dr. Moreau’s fashioning of humans out of beasts is clearly
analogous to the pitiless procedures of Nature and its evolution­
ary creation. He is not only a latter-day Dr. Frankenstein but also
a demonically inverted God of Genesis, and his surgically
humanized Beast Folk are a counterpart of ourselves, semibestial
humans. Wells's calling their attempts to mimic the Decalogue in
the litanies of “The Saving of the Law” and their collapse back
into bestiality a “theological grotesque" indicates that this view of
mankind's future reversed Christian as well as socialist millen­
nial ism into the bleak vistas of an evolution liable to regression.
The Island of Dr. Moreau turns the imperial order of Kipling’s
Jungle Book into a degenerative slaughterhouse, where the law
loses out to bestiality.
Wells's next two famous SF novels, though full of vivid local
color, seem today less felicitous. Both have problems of focusing.
In The Invisible Man (1897) the delineation of Griffin hesitates be­
tween a man in advance of his time within an indifferent society
and the symbol of a humanity that does not know howr to use
science. This makes of him almost an old-fashioned “mad scien­
tist,” and vet he is too important and too sinned against to be
comic relief. The vigor of the narration, which unfolds in the
form of a hunt, and the strengths of an inverted fairy tale cannot
compensate for the failure of the supposedly omniscient author
to explain why Griffin had got into the position of being his own
Frankenstein and Monster at the same time. In this context, the
dubious scientific premises (an invisible eye cannot see, and so
forth) become distressing and tend to deprive the story of the
needed suspension of disbelief. The H’flr of the Worlds (1898),
which extrapolates into xenobiology the catastrophic stories of
the “future wars” subgenre discussed in chapter 7, descends in
places to a gleeful sensationalism difficult to stomach, especially
in its horror-fantasy portraiture of the Martians. The immediate
serialization in the US yellow press, which simply suppressed the
parts without action, made this portraiture the most influential
model for countless later Things from Outer Space, extendable
to any foreign group that the public was at that moment sup­
posed to hate, and a prototype of mass-media use of SF for
mindless scare-mongering (inaugurated by Orson Welles’s fa­
mous 1938 broadcast). The novel’s composition is marred by the
WELLS AS THE TURNING POINT OF THE SF TRADITION 215

clumsy system of two eyewitness narrators, improvised in order


to reconcile the sensational immediacy and the necessary over­
view. Of course, The War of the Worlds also contains striking and
indeed prophetic insights such as the picture of modern total
warfare, with its panics, refugees, quislings, underground hid­
ings, and an envisaged Resistance movement, as well as race­
theory justifications, poison gas, and a “spontaneous” bac­
teriological weapon. (In other tales, Wells—a lifelong lover of
war games—added air warfare, tanks, atom bombing of a major
city, and other bellicose devices.)
Except for the superb late parable “The Country of the Blind"
(bp. 1911), Wells’s sociobiological and cosmological SF cycle cul­
minated in The First Men in the Moon (1901). It has the merit of
summarizing and explicating openly his main motifs and devices.
The usual two narrators have grown into the contrasting charac­
ters of Bedford, the Social-Darwin ¡st speculator-adventurer, and
Cavor, the selfless scientist in whom W’ells manages for once to
fuse the cliche of absent-mindedness with open-mindedness and
a final suffering rendered irremediable by the cosmic vistas in­
volved. The sharply focused lens of spatial pinpointing and tem­
poral acceleration through which the travelers perceive the
miraculous growth of Lunar vegetation is the most striking ren­
dering of the precise yet wondering scientific regard often as­
sociated with the observatories and observation posts of Wells’s
stories. The Selenites not only possess the Aesopian fable back­
ground and an endearing grotesqueness worthy of Edward
Lear’s creatures, they are also a profound image of sociopolitical
functional overspecialization and of an absolute caste or race
State, readily translatable from insect biology back into some of
the most menacing tendencies of modern power concentration.
Most Swiftian among Wells's aliens, they sene a double-edged
satire, in the authentic tone of savage and cognitive indignation:

... 1 came upon a number of young Selenites, confined in


jars from which only the fore limbs protruded, who were
being compressed to become machine-minders of a special
sort. . . . these glimpses of the educational methods of these
beings have affected me disagreeably. I hope, however, that
may pass off and 1 may be able to see more of this aspect of
this wonderful social order. That wretched-looking hand
216 WELLS AS THE TURNING POINT OF THE SF TRADITION

sticking out of its jar seemed to appeal for lost possibilities; it


haunts me still, although, of course, it is really in the end a
far more humane proceeding than our earthly method of
leaving children to grow into human beings, and then mak­
ing machines of them, [chap. 23]
The usual final estrangement fuses biological and social disgust
into Bedford’s schizophrenic cosmic vision of himself “not only
as an ass, but as the son of many generations of asses” (chap. 19).
Parallel to that, Cavor formulates most clearly the uselessness of
cosmic as well as earthly imperialism, and articulates a refusal to
let science go on serving them (had this been heeded, we would
have been spared the Galactic Empire politics and swashbuckling
of later SF). Finally, Bedford’s narration in guise of a literary
manuscript with pretenses to scientific veracity, combined with
Cavor’s narration in guise of interplanetary telegraphic reports,
exhibit openly Wells’s ubiquitous mimicry of the journalistic style
from that heyday of early “mass communications"—the style of
“an Associated Press dispatch, describing a universal night­
mare."8
Yet such virtuosity cannot mask the fundamental ambiguity
that constitutes both the richness and the weakness of Wells. Is
he horrified or grimly elated by the high price of evolution (The
Island of Dr. Moreau)"? Does he condemn class divisions or simply
the existence of a menacing lower class (The Time Machine)"?
Does he condemn imperialism (The First Men in the Moon) or only
dislike being at the receiving end of it (The War of the Worlds)? In
brief, are his preoccupations with violence and alienation those
of a diagnostician or of a fan? Both of these stances coexist in his
works in a shifting and often unclear balance. For example,—to
translate such alternatives into an immediate determinant of
narration—Wells’s central morphological dilemma in the years of
his first and best SF cycle was: which is' the privileged way of
understanding the world, the scientifically systematic one or the
artistically vivid one? Faced with the tension between “scientific"
classification and “artistic” individuation, a tension that remained
constant (albeit with different outcomes) throughout his life,
Wells had already in 1891 satirized the deterministic rigidity in

8. Unsigned review in Critic, 23 April 1898, reprinted in Parrinder, ed.. p. 69.


WELLS AS THE TURNING POINT OF THE SF TRADITION ai7

his essay “The Universe Rigid” and gone on to find a first com­
promise in his “The Rediscovery of the Unique” and its succes­
sive avatars in “The Cyclic Delusion," "Scepticism of the Instru­
ment," and First and Last Things (1908). These articles attempt to
formulate the deep though unclear pulls which Wells at his best
reconciled by opting for representativeness, for fusing indi-
viduum and species into socially and biologically typical figures
like the Time Traveller, but which he often left unreconciled.
Wells’s SF makes thus an aesthetic form of hesitations, intima­
tions, and glimpses of an ambiguously disquieting strangeness.
The strange novum is gleefully wielded as a sensational scare
thrown into the bourgeois reader, but its values are finally held
at arm's length. In admitting and using their possibility he went
decisively beyond Verne, in identifying them as horrible he de­
cisively opposed Morris. Wells’s SF works are clearly "ideological
fables,”9 yet he is a virtuoso in having it ideologically both ways.
His satisfaction at the destruction of the false bourgeois idyll is
matched by his horror at the alien forces destroying it. He reso­
lutely' clung to his insight that such forces must be portrayed, but
he portrayed them within a sensationalism that neutralizes most
of the genuine newness. Except in his maturest moments, the
conflicts in his SF are therefore transferred—following the
Social-Darwinist model—from society to biology. This is a risky
proceeding which can lead to some striking analogies but—as
was discussed in chapter 6 a propos of Frankenstein—as a rule
indicates a return to quasi-religious eschatology and fatal abso­
lutes. Wells expressed this, no doubt, in sincerely Darwinist
terms, but his approach is in fact marked by a contamination of
echoes from a culturally sunken medieval bestiary and a Miltonic
or Bunyanesque color scheme (dark and red, for example, as
Satanic) with the new possibilities of scientific dooms (compare
the Ruskinian Angel of The Wonderful Visit [1895], presented as
an alien from a parallel world). The annihilation of this world is
the only future alternative to its present state; the present bour­
geois way of life is with scientific certainty leading the Individual­
ist hnmme moyen sensuel toward the hell of physical indignity and
psychic terror, yet this is still the only way of life he can return to
and rely on, the best of all the bad possible worlds. Thus Wells’s

9, Parrinder (Bibliography V), p. 18.


¿18 WELLS AS THE TURNING POINT OF THE SF TRADITION

central anxious question about the future in store for his


Even-man—who is, characteristically, a bright, aggressive, White,
middle-class male—cannot be resolved as posed. His early SF can
present the future only as a highly menacing yet finally inopera­
tive novum, the connection with its bearers (Time Traveller,
Moreau, Griffin, Martians, Selenites, or Cavor) being always bro­
ken off. Formally, this impasse explains his troubles with works
of novel length: his most successful form is either the short story
or the novelette, which lend themselves to ingenious balancings
on the razor’s edge between shock and cognitive development.
In them he set the pace for the commercial norms of most later
SF (which adds insult to injurv by calling such works novels).
Wells’s later SF abandoned such fragile but rich ambiguity in
favor of short-range extrapolations. His first attempt in that di­
rection, When the Sleeper Wakes (1899), was the most interesting.
Its picture of a futuristic megalopolis with mass social struggles
led by demagogic leaders was “a nightmare of Capitalism trium­
phant” and an explicit polemic against Bellamy’s complacent op­
timism about taming the organizing urge and the jungle of the
cities, in Wells’s complex corporate capitalism “everything was
bigger, quicker and more crowded; there was more and more
(lying and the wildest financial speculation.”10 Since Wells’s
sketch of the future was full of brilliant and detailed insights (as,
for example, those about competing police forces and stultifying
mass media) that turned out to be close to actual developments
in the twentieth century, this novel became the model for anti-
utopian anticipation from Zamyatin and von Harbou to Heinlein
and Pohl. But Wells’s imaginative energy- flagged here at the cru­
cial narrative level: the observer-hero waking after tw-o centuries
behaves alternatively like a savior (suffering his final passion on
an airplane instead of a cross) and vacillating liberal intellectual.
The jerky plot concerns itself primarily with the adventure of a
beautiful soul in the future, and is thus coresponsible for a spate
of similar inferior SF with more rugged heroes who are given
wonderful powers and who experience sentimental entangle­
ments. “A Story of Days To Come” (1899) and “A Dream of
Armageddon” (1903), told wholly from inside the same future.

!0. Wells, first quotation from his "Author's Preface" to The Sleeper Auakes (London,
1921), second quotation from his Experiment in Autobiography (New York, 1934), p. 551.
W£LLS AS THE TERMING POINT OF THE SF TRADITION 219

are not much more than an exploitation of that interesting lo­


cale for sentimental tales seen from the bottom, respectively the
top, of society. Wells’s later SF novels—though even at their
worst never lacking flashes of genuine insight or redeeming
provocation—do not attain the imaginative consistency of his
first cycle. In The Food of the Gods (1904) the fundamental equa­
tion of material and moral greatness is never worked out. His
series of programmatic utopias, from A Modern Utopia (1905) to
The Holy Terror (1939), has interesting moments, especially when
he is describing a new psycholog}- of power and responsibility
such as that of the “Samurai" or the “holy terror” dictator. How­
ever, its central search for a caste of technocratic managers as
“competent receivers”11 for a bankrupt capitalist society oscil­
lates wildly from enlightened monarchs or dictators, through
Fabian-like artists and engineers, to airmen and Keynesians (in
The Shape of Things to Come, 1933): millennium has always been
the most colorless part of Christian apocalypse. What is worst,
Wells’s fascinated sensitivity to the uncertain horizons of human­
ity gives only too often way to impatient discursive scolding,
often correct but rarely memorable. A visit to young Soviet Rus­
sia (where his meeting with Lenin provided an almost textbook
example of contrasts between abstract and concrete utopianism)
resulted in the perhaps most interesting work in that series, Men
Like Gods (1923), where Wells gave a transient and somewhat
etiolated glimpse of a Morris-like brightness. But his work after
die first World War vacillated, not illogically for an apocalyptic
writer, between equally superficial optimism and despair. His
position in the middle, wishing a plague on both the upper and
the working classes, proved singularly fruitless in creative
terms—though extremely influential and bearing strange fruit in
subsequent SF, the writers and readers of which mostly come
from precisely those “new middle classes" that Wells advanced as
the hope of the future.
With all his strengths and weaknesses Wells remains the cen­
tral writer in the tradition of SF. His ideological impasses are
fought out as memorable and rich contradictions tied to an in­
exorably developing future. He collected, as it were, all the main
influences of earlier writers—from Lucian and Swift to Kepler,

11. Wells, in Autobiography, p. 206.


220 WELLS AS THE TURNING POINT OF THE SF TRADITION

Verne, and Flaminarion, from Plato and Morris to Mary Shelley,


Poe, Bukver, and the subliterature of planetary and subterra­
nean vovages, future wars, and the like—and transformed them
in his own image, whence they entered the treasury of sub­
sequent SF. He invented a new thing under the sun in the time­
travel story made plausible or verisimilar by physics. He codified,
for better or worse, the notions of invasion from space and cos­
mic catastrophe (as in his story “The Star," 1899), of social and
biological degeneration, of fourth dimension, of future meg­
alopolis, of biological plasticity. Together with Verne’s roman sci-
entifique, Wells’s “scientific romances” and short stories became
the privileged form in which SF was admitted into an official
culture that rejected socialist utopianism. True, of his twenty­
odd books that can be considered SF, only perhaps eight or nine
are still of living interest, but those contain unforgettable visions
(all in the five “romances" and the short stories of the early
sociobiological-cum-cosmic cycle): the solar eclipse at the end of
time, the faded flowers from the future, the invincible obtuse­
ness of southern England and the Country of the Blind con­
fronted w ith the New, the Saying of the Law on Moreau’s island,
the wildfire spread of the red Martian weed and invasion panic
toward London, the last Martian’s lugubrious ululations in
Regent’s Park, the frozen world of “The New Accelerator,” the
springing to life of the Moon vegetation, the lunar society. These
summits of Wells's are a demonstration of what is possible in SF,
of the cognitive shudder peculiar to it. Their poetry is based on a
shocking transmutation of scientific into aesthetic cognition, and
poets from Eliot to Borges have paid tribute to it. More harrow­
ing than in the socialist Utopians, more sustained than in Twain,
embracing a whole dimension of radical doubt and questioning
that makes Verne look bland, it is a grim caricature of bestial
bondage and an explosive liberation achieved by means of
knowledge. Wells was the first significant writer wrho started to
write SF from within the world of science, and not merely facing
it. Though his catastrophes are a retraction of Bellamy’s and
Morris’s utopian optimism, even in the spatial disguises of a
parallel present on Moreau’s island or in southern England it is
ahvays a possible future evolving from the neglected horrors of
today that is analyzed in its (as a rule) maleficent consequences,
and his hero has “an epic and public . . . mission” intimately
jyELLS AS THE TURNING POINT OF THE SF TRADITION 221

bound up with “the major cognitive challenge of the Darwinist


age.”12 For all his vacillations, Wells’s basic historical lesson is that
the stifling bourgeois society is but a short moment in an impre-
dictable, menacing, but at least theoretically open-ended human
evolution under the stars. He endowed later SF with a basically
materialist look back at human life and a rebelliousness against
its entropic closure. For such reasons, all subsequent significant
SF can be said to have sprung from Wells’s Time Machine, which
will be examined next.

12. Parrinder (Bibliography IV C). p, 273.


10

The Time Machine versus


Utopia as Structural Models for SF

In this chapter I shall tn- to show that Wells’s The Time Machine
is (to put it prudently in the absence of further evidence) at least
one, and that More’s Utopia was another, among the basic histor­
ical models for the structuring of subsequent SF. One does not
need to be a structuralist in the sectarian sense of opposing syn­
chronic analysis to cultural genetics or taking myth as synony­
mous with literature to use some of the methods which struc­
turalism shares with a whole exegetic tradition extending from,
say, medieval discussions to some of Lukacs’s analyses or Kuhn’s
Structure of Scientific Revolutions. A student of Wells is embold­
ened in such an approach by the fact that comparative mor­
phology was in Wells’s student days one of the first great modern
breakthroughs of the structural method. As he himself noted,
biologs was in T. H. Huxley’s days establishing the phylogenetic
tree, or “family tree of life”: “Our chief discipline was a rigorous
analysis of vertebrate structure, vertebrate embryology', and
the succession of vertebrate forms in time. We felt our particular task
was the determination of the relationships of groups by the acutest pos­
sible criticism of structure."'1 Wells left no doubt of the indelible
vistas the “sweepingly magnificent series” of zoological exercises
imprinted on his eager imagination, leaving him with an ur­
gency for "coherence and consistency”: "It was a grammar of
form and a criticism offact. That year I spent in Huxley's class was,
beyond all question, the most educational year of my life.”12

1. H. G. Wells, Experiment in Autobiography (New York, 1934). p. 160, italics added.


Students of Wells will recognize my large debts toward the critics and scholars, from
Brooks and West to Bergonzi and Hi liegas, Parrinder and Philmus, listed in Bibliography
V, which will as a rule be acknowledged only in cases of direct mention or quote.
2. Welts, pp. 160-61, italics added. Compare also Wells’s explicit preoccupation with
biological “degradation” inherent in evolution under capitalism in the articles "Zoological
Retrogression," The Gentleman's Magazine, 7 September 1891; “On Extinction,” Chambers’s

222
THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 223

It should not, thus, be too surprising to find in The Time Ma­


chine—which has much to say about succession of zoological
forms in time—an attempt at coherence and consistency, “a
grammar of form and criticism of fact.” Of course, this does not
prejudice the particular grammar or criticism, the type of coher­
ence and consistency that might be found in it.
I am proceeding on the hypothesis that the basic device at The
Time Machine is an opposition of the Time Traveller’s visions of
the future to the ideal reader’s norm of a complacent bourgeois
class consciousness with its belief in linear progress, Spencerian
“Social Darwinism,” and the like. The Victorian norm is set up in
the framework of The Time Machine and supplemented by the
Time Traveller’s reactions. His visions are shaped by means of
two basic and interlocking symbolic systems: that of biological
regression, and that of a color imager)' polarized between light
and darkness—both systems being allied with violence, pain, and
the basic confrontation between Xian and Death.
1. The Converging Biological Series •
The one thing earlier drafts of The Time Machine have in com­
mon with the vastly different and superior final version is an
opposition between the present and a different future.*34How­
ever, the narrative organization of the final version manifestly
rook its cue from Darwinism as expounded by Wells’s teacher
Huxley from I860 on, and applied to “Evolution and Ethics” by
Huxley’s homonymous Romanes Lecture and the subsequent
“Prolegomena” in the year preceding Wells's writing of the final
version (1893-94).
In the “Prolegomena,” Huxley tried to face the implications of
evolution applying not only to “progressive development" but
also to “retrogressive modification,” not only to “gradual change

Journal. 30 September 1893; '"The Wan of the Year Million.” Pall Mall Gaietle, 9
November 1893; and “The Extinction of Man," Pali Mall Gazette, 23 September 1894—
the first wo reprinted in Philmus and Hughes, ed. (Bibliography V) and the last two in
H. G. Wells. Certain Personal Matters (London, 1898 [1897]); also Wells's comment in '42 to
‘44 (London, 1944), p. 9.
3. See Bernard Bergonzi, "The Publication of The Time Machine, 1894-1895,” Review
of English Studies, N.S. II (I960): 42-51; also Bergonzi. The Early H. G. Wells (Bibliog­
raphy V). and Philmus and Hughes, ed.
4. T. H. Huxley, “Evolution and Ethics—Prolegomena,” in his Evolution and Ethics.
Collected Essays, 9 (London, 1903). note 1 on p. 4. and p. 6.
234 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

from a condition of relative uniformity to one of relative com­


plexity" but also to “die phenomena of retrogressive metamor­
phosis, that is, of progress from a condition of relative complex­
ity to one of relative uniformity.”4 Evidently the connotations of
progress in the bourgeois liberal sense were being challenged by
connotations that made it synonymous with any evolutionary
change, for better or for worse; progress was being expanded to
encompass the antonymic possibility of "retrogressive metamor­
phosis.” Another variation of this ambiguity is the possibility of
envisaging evolution in terms of devolution. Again in the "Pro­
legomena,” setting up his basic exemplum or parable of English
vegetation that might evolve from a primitive slate of nature into
a garden under purposeful (that is, ethical) human intervention,
Huxley mused that “if every link in the ancestry of these humble
indigenous plants had been preserved and were accessible to us,
the whole would present a converging series of forms of gradually
diminishing complexity, until, at some period in the history of the
earth . . . they would merge in those low groups among which
the boundaries between animal and vegetable life become ef­
faced.
Huxley was, of course, only indulging in the common evo­
lutionist device of exalting even the humblest “indigenous
plants” as wonderful products of an evolutionary chain.56 This
chain is here traversed backward into the past and functioning as

5. Huxley, p. 5. italics added.


6. See. as outstanding examples, the final paragraph of chapter 6 and also chapter 21,
in Darwin's The Descent qf Stan (London, 1874)—not to mention the famous parable of
the tangled bank that concludes The Origin of Species. T. H. Huxley uses the same device,
for example, at the end of chap. 2 ("On the Relations of Man to the Lower Animals") of
Man's Place in Nature (London, 1863). The collocations of such passages al the climaxes of
books or book sections testifies both to the rhetorical effectiveness of the parable and to
the Darwinist sense of its importance as ethico-aest he deal justification of evolution. The
upward, excelsior course of the arrow of time provided a new type of dynamic sublimity
analogical to the per aspera ad astra (per evolutionem ad hominemf) rise of the Victorian
“self-made man." For an approach to Darwin’s aesthetics and rhetoric see Stanley Edgar
Hyman. The Tangled Bank (New York, 1962). and Walter F, Cannon, “Darwin's Vision in
On the Origin of Species," in George Levine and William Madden, eds.. The Art of Pictarian
Prose (New York, 1968), pp. 154-76. who situate them convincingly in ideological time
and place. On Darwin's concept of the sublime see Donald Fleming’s "Charles Darwin,
the Anaesthetic Man," in Philip Appleman, ed., Darwin (New York, 1970). pp. 573-89;
Fleming speaks of Darwin’s hidden “Carlylean self,” p. 583, but for differences be­
tween them see Theodore Baird, "Darwin and the Tangled Bank," American Scholar 15
(Autumn 1946).
-THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 225

a kind of double negation—since the reader is tacitly invited to


reascend the evolutionary ladder from the Protistae to the
“humble” (now not so humble) indigenous plant in the direction
of a diverging series of increasing complexity. But what if one
were to take this formal exercise of Huxley’s literally, and his
sense of uneasiness about evolution and progress versus devolu­
tion and regress seriously—that is, refusing a rhetoric that de­
scends into the deeps of the problem and of time in order to end
with an upward flourish? .All that would be needed is to suppress
Huxley’s second negation by inverting his vision from past to
future, and to imagine a canonic sociobiological “converging se­
ries of forms of gradually diminishing complexity” unfolding as
advolution that retraverses the path of evolution backward to a
fin du globe. That is what Huxley’s heretical student did in The
Time Machine: a slip under the time telescope, to use a Wellsian
phrase.7
“Canonic series,” of course, begs the question “according to

7. This inversion of the Darwinian time-arrow seems to have been one of Wells's basic
intellectual, morphological, and visionary discoveries. His first work that transcends ado­
lescent doodling (a student debating address in 1885) was entitled The Past and Future of
the Human Race, the title of a key book in 1902 is The Discovery of the Future, and in 1936 it
is (characteristically) the archaeologist in his The Croquet Player who speaks of the abyss of
the future bringing the ancestral savage beast back (chapter 3). tn The Future in America
(London. 1906), p. 10, Wells explicitly connects monstrous science-fictional projections
with such Darwinian anticipations.
Strictly speaking. Darwin's theory is neutral as far as prospects or the present state of
mankind are cotKerned, He himself, although quite aware of biological retrogression,
extinction, and such, assumed that biological groups "which are now' large and trium­
phant . . . will for a long period continue to increase," and stressed the ennobling aspects
of evolution—see note 6, and On the Origin of Species (Cambridge, MA, 1964), pp. 126 and
488-90, For philosophical implications of the Darwinian time-arrow see. e.g., Loren
Eiseley, Darwin's Century (New York. 1958), pp. 330-31; for literary ones, the stimulating
essay by A. Dwight Culler, "The Darwinian Revolution and Literary Form," in Levine
and Madden, eds., pp. 224—46. It was the opponents of Darwin's theory who first seized
upon its malevolent aspect—see chapters 12-14 of Leo J. Henkin, Darwinism in the English
Novel 1960-1910 (New- York, 1940). On the other extreme, Spencer’s contention that
evolution through struggle for life “can end only in the establishment of the greatest
perfection and the most complete happiness"—First Principles (New York, 1900), p. 530
—is the real villain of this ideological drama or piece a these. The naive capitalist Spcn-
cerians or Social Darwinists (in the United States, e.g., Rockefeller or Carnegie) whole­
heartedly embraced trampling the "less fit" multitude; John D. Rockefeller's parable
of the American Beauty rose, "produced , . , only by sacrificing the early buds," de­
serves to be as famous as Darwin’s tangled bank or Menenius Agrippa's fable of the
belly and the members; see Richard Hofstadter, Social Darwinism in American Thought
(Bibliography IV C). especially chapter 2.
225 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

what type of canon?” An answer is to be found by comparing the


orthodox Darwinist and Huxley an canon with the one actually
used in The Time Machine. The orthodox seriation would, in sim­
plified outline, look as follows:8

FIGURE 1
Nonhuman
Primates
J
Placentals Marsupials
& Monotremes
I

Mammals Birds

Warmblooded Coldblooded

Amniotic Nonamniotic
(fully ter­ (Amphibia, part-
restrial) ly terrestrial) Arthropods
Molluscs
Land Animals Sea Animals

Vertebrate Invertebrate
(Chordate) (Nonchordate)

COELOMATE NONCOELOMATE

Three-Layered Two-Layered

Multicellular Unicellular
I________________ I
Animals Plants

Protistae
(Organic Inorganic
Entities) Entities

Existence Nonexistence
I_______________I
8. Adapted from J. B. S. Haldane and Julian Huxley, Animal Biology (Oxford, 1927),
fig. 81 on pp. 258-59.
~THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 227

The final two levels are an extrapolation going beyond biology


but present in Lyell, Darwin, and T. H. Huxley.
As differing from this converging series, Wells not only used
the symmetrically inverse time direction, changing the sign from
plus to minus, but also considerably foreshortened and re­
grouped the series (see figure 2). Comparing the two series, a
number of significant divergencies obtrude:
(1) Even the simplified Darwinian serialion (omitting the level
of chordates versus nonchordates, the distinctions between bony
fish, cartilaginous fish, and cyclostomes, and so on) contains 12
levels beginning with placentals versus marsupials, whereas the
corresponding Wellsian seriation contains six levels beginning
with placentals versus marsupials, or five without this level,
which is omitted in Wells's final version. 1 he levels Wells re­
tained are those that can be vividly represented by striking im­
ages. fhe differences between existence and eclipse, plant and
animal, sea and land animal, “amphibian" crab and mammal are
readily understood without the Darwinian schematism. They are
based on a “self-evident" or commonsensical topical bestiary an­
tedating Linnaeus, indeed, harking back to the dawn of human
imagination. Therefore, they were eminently usable for produc­
ing in the average non scientific but science-believing reader ef­
fects of stark opposition, such as the revulsion felt by the Time
Traveller when faced with the giant crabs or with the unnamed
archetypal “thing" from the sea. Though Wells’s task as far as
lower rungs of the phylogenetic series are concerned was facili­
tated by the unsettled state of invertebrate phylogeny at the
time,910there is little reason to suppose scientific scruples carried
significant weight with him. Indeed, when Wells had to choose
between the ABC of Darwinian classification and a kind of folk
biology, he unhesitatingly chose the latter, inventing—like the
hero of his parable "The Triumphs of a Taxidermist”—new
taxonomic positions: the triumph of a quasi-taxonomist, in­
deed.19 This is evident in the third episode of The Time Machine
where the crabs are—against all biological taxonomy—situated

9. Set Wells, Experiment, p. 160.


10, See Wells» Experiment. for his references to **de-jndi virtualizing" and perceiving
individuals in relation io a story and a thesis, on p. 173, p. 520» and particularly the
“sedation al” account of hts creative imagination tn the dispute with Conrad on p. 528.
THE TÍME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS
FIGURE 2
THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 229

between the opposition placental versus marsupial (in the omit­


ted “kangaroo” episode) and land animal versus sea animal, that
is, in the false, “folk taxonomic” position of amphibian creature
because of their location or ecological niche. That ecological
niche—the line where sea meets land—is easily represented and
possesses rich literary overtones; the representation of a true,
taxonomic amphibian (say Capek’s giant salamanders) would lack
the element of menace present in the insectoid antennae and
claws, the alienness of eyes wriggling on stalks, and so on. Wells
was to use this biologically collateral branch of arthropods for
similar “creepy crawly” effects in “The Empire of the Ants” and
"The Valley of Spiders,” just as he was to do with the taxono­
mically isotopic molluscs in “The Sea Raiders.” Incidentally,
through the octopoid Martians from The War of the Worlds, the
insectoid Selenites from The First Men in the Moon, and perhaps
the vaguely reptilian bipeds of “In the Abyss," Wells set the xeno-
biological paradigm for SF’s Bug-Eyed Monsters and Menaces
from Outer Space right down to the supremely unscientific ap­
pellative “the thing.”
(2) The beginning or top of Wells’s devolutionary series stems
from a curious hybrid of deterministic or Malthusian pseudo­
Darwinism and bourgeois, or indeed imperialist, social theory
(and practice). That hybrid was represented not only by the “So­
cial Darwinism” of Spencer but also, later, by Carnegie, Rock­
efeller, Nietzsche, and fascists of various stripes, all of whom
translated differences in socioeconomic position into a biological
terminology and stressed the “survival of the fittest." T.H. Hux­
ley’s concern with the relationships of evolution and ethics is due
precisely to an uneasiness about such uses of Darwinism. How­
ever, the usual strategy of the Social Darwinists was to use the
more convenient and mystified vocabulary of racism, preaching
social peace in the imperial metropolis and among “Whites” at
the expense of colonial peoples or “lower races.”11 Wells’s per­
sonal class experiences and conversance with Plato, Blake, Shel­

11. See. e.g.. Wells’s explicit comment on such an attitude, which fused scientific
progress and a sense of imperial mission in fin-de-siecle Britain, in Joan and Pttrr
(London. 1918). book 1, chap. 3. In his admiration of Malthus, however, Welts himself
was not always immune from it—say in Anticipations (New York and London. 1902), pp.
313-14 and the last three chapters generally. Such instances could be multiplied.
¿30 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

ley, Morris, and Marx precluded at the moment of writing The


Time Machine such a mystification and induced him to put the
problem in its basic terms of social class rather than in
pseudobiologicai ones. He was to be followed on that level by
some of the best social criticism in SF, from Jack London’s fusion
of Wells and utopianism in The Iron Heel to the anti-utopia of
Zamyatin, the political “new maps of hell” of American SF in the
1940s and 1050s, and the satirical SF from the Warsaw Pact
countries (Lem, Dneprov, the Strugatskys).12 Indeed, the se­
quence of the Time Traveller's hypotheses about the “Eloi
episode”—(a) Communist classless society; (b) degenerated class­
less society; (c) degenerated class society; (d) degenerated in­
verted class society—comprises the whole logical gamut of socio­
political SF, or of utopian and antiutopian fiction as the ideal
poles of sociological SF, from More and Plato to the present day.
Finally, with the Time Traveller’s realization that the capita­
lists and workers have not only degenerated and inverted their
power roles but have also differentiated into separate biolog­
ical species, one of which is (he “cattle" of the other,13 Huxley’s
evolution that encompasses devolution comes true with a ven­
geance, even as the ideological basis of such speculations in real
class fears and hopes is uncovered. The resulting “race” level of
oppositions was to be used by Wells, again inverted, in The War of
the Worlds or stories such as “Lord of the Dynamos.” Reverting to
crude xenophobia and losing Wells’s ambiguities, this became the
model for a whole group of subsequent and inferior SF narra­
tives exporting social and national conflicts into outer space.
(3) Lt also becomes clear why Wells felt he had to delete the
“kangaroo and centipede” episode. Not only did it completely
break up the narrative rhythm by introducing at a still leisurely
stage of narration two new phylogenetic levels in a single epi­
sode—marsupial and arthropod, of which, furthermore, the ar­
thropod level is used again in the following “crab episode”—it
added little to the basic opposition of the Time Traveller as
mammal, land animal, and so forth, to non-mammals, sea ani­
mals, and so forth. On the contrary, it logically raised the disso­

12. See some of their stories in (and also the preface co) D. Suvin, ed., Other Worlds,
Other Seas (X'ew York, 1972).
13. See the identical and no doubt seminal metaphor tn Huxley, Evolution, p, 17.
fHE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 231

nant question of using a full Darwino-Huxleyan converging


series, beginning with the opposition of man (Time Traveller) to
primates. With commendable tact, Wells was unwilling to ven­
ture onto grounds later annexed by Tarzan of the Apes, al­
though he subsequently compromised under Kipling’s influence
to the point of exploring the opposition Man versus (mammali­
an) Beast in The Island of Dr. Moreau (where Kipling’s serpent is
needs omitted).
(4) Finally, if one looks at the distribution of the seriation
levels in the episodes of The Time Machine, it is possible to gain
further insight about its basic narrative rhythm, also charac­
terized by growing pace and compression as the reader is swept
into the story, the motivations and justifications gradually dis­
pensed with, and the levels cumulated in an. exponential pro­
gression. The Time Machine consists of a framework and the three
phantasmagoric evolutionary futures 1 have called the “Eloi," the
“crab”, and the “eclipse” episodes. In the first British edition,
their quantitative relationships are as follows:
Framework established, chapters 1-3, pp. 1-26
Eloi episode, chapters 4-13, pp. 27-133
Crab episode, first half of chapter 14, pp. 134-39
Eclipse episode, second half of chapter 14, pp. 139-41
Framework reestablished, chapters 15-16 and Epilogue, pp. 142-52
Or, taking into account only the inner narration of the Time
Traveller's experiences:
Eloi episode, year 802,701—one seriation level (or two), 107 pages
(Omitted “kangaroo and centipede" episode, year?—two seriation
levels, ca. 4 pages)
Crab episode, several million years hence—one seriation level, 5 pages
Eclipse episode, 30 million years hence—four seriation levels, 3 pages
Taking the Eloi episode as two levels (first a class and then a
“race” or species oncH), there is in the above four future epi­
sodes:

14. Some evidence that Wells associated class and race as isomorphic antagonistic
oppositions in conflicts between oppressors and oppressed, such as those that presumably
led to the development of the Eloi and Morlocks, can be found in statements like the
following: “the driving discontent has often appeared as a conflict between oppressors
and oppressed, either as a class or as a race conflict . . (Exfmrmenl, p. 626).
232 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

1. one level per 54 pages


2. (one level per ca. 2 pages—omitted)
3. one level per 5 pages
4. one level per 0.7 pages
One could venture further into a discussion of such an expo­
nentially regressing rhythm—which is certainly analogous on its
own structural level to the whole regressive structure of The Time
Machine—but I shall confine myself to one general observation.
The rhythm starts as lento, with two sociobiological levels en­
visioned for 107 pages. It continues as presto, with one biological
level (mammal versus “amphibian”) for five pages, and ends in
an abrupt prestissimo with four existential levels (land versus sea
animal; animal versus plant—lowest plant forms at that; organic
existence versus sand, snow, rocks, and sea; and existence of
Earth versus eclipse) ail present pell-mell, outside of their proper
taxonomic order, within about three pages. I'his telescoping and
foreshortening powerfully contributes to and indeed shapes the
effect of the logical or biological series. Also, this asymptotic
series makes it imperative that the Traveller finally vanish: its
final and validating member can only be zero or nonexistence,
extinction.1®
Thus, Wells's Time Machine has in the organization of its cogni­
tive thematic material hit upon the law—inherited, as much else
was, from Gulliver’s Travels, and apparently unshaken in sub­
sequent significant SF—that the cognitive nucleus of narration,
or theme, can become a principle of narrative organization only
by fitting into the storytelling parameters of pace, sequence, sym­
bolic systematization, and so on. Wells knew of Haeckel’s law that
ontogenesis (development of any species’ embryo) is a fore­
shortened recapitulation of phylogenesis (that species' evolution);
in other words, that the new environment of individual embry­
onic gestation inflects and modifies—though it does not change
the general outline of—the evolutionary sequence. Consciously
or not, he applied the same principle to the new narrative and
aesthetic environment—an environment to which the cognitive
evolutionary sequence of Darwinian seriation had to adapt by
evolving or indeed mutating. The principle of a Wellsian structure

15. See Robert M. Philmus, "The Logic of ‘Prophecy’ in The Time Machine“ in Ber­
gonzi, ed. (Bibliography V’), for effects of this structural device.
,THE TIME MACHINE VERSUS LT0PI.1 AS STRUCTURAL MODELS 233

of science fiction is mutation of scientific into aesthetic cognition. A


Wellsian narration is oriented toward cognitive horizons that it
shares with any good handbook of sociology, biology , or philos­
ophy of science. But the orientation is achieved in its own way,
following the autonomy of a narrative, fictional aesthetic mode.
2. The Synoptic Paradigm
Obviously, in order to account for Wellsian narrative strategies
paradigmatic of later SF, one would have to analyze the symbolic
system that is intertwined with this regressive biological seriation
inflected toward “folk taxonomy.” As a number of critics have
noted, this svmbolic system is based on violent oppositions of
color, polarized between the Doomsday connotations of “eclipse
black” and “fiery red” on one hand, and the green and bright
colors of the utopian garden and sunlit landscaped vegetation on
the other. It is not too difficult to see in these poles a coloristic
translation of the opposition between tamed and untamed, safe
and menacing, evolutive and devolutive nature. Huxley’s parable
of the cultivated garden or evolutionary Eden is thus supplied
with the missing black hues.16 Moreover, the analysis of this as­
pect would entail a full and lengthy exploration of Wells’s par­
ticular anthropology and cosmology. Instead, I would like to
further examine the temporal orientation of The Time Machine
and the basic oppositions implicit in the annunciation of the bad,
devolutionary, or black future in store for the bourgeois reader
as a sociobiological entity, or for any reader as a cosmological
entity, and compare this to the utopian alternative.
The Time Traveller’s futures are a geometrically progressive
series of devolutions, which can, by explicating the implicit op­
position between (1) Social-Darwinist Britain and (2) the particu­
lar future vision, be tabulated using a Levi-Straussian schema­
tism. This would play upon the Social-Darwinist preconceptions
of a “natural” order of power and of a safe evolutionary pro­
gress keeping each “lower” evolutionary rung in its place as prey
of the “higher” predator; Wells takes these preconceptions over
wholesale and simply inverts them. What results is an inverse
and symmetrical structure, which can be finally reduced to a
general abstract scheme or paradigm:
16. But see Evolution, pp. 17 ft., for Huxley's Malthusian "serpent within the garden.”
234 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

FIGURE 3
(1) Social-Darwinist B retain (2) The Future
a. Eloi episode
Predator Prey Predator Prey
Upper Class
(Capitalist) + Eloi
+
Lower Class
(Laborer) + Mor locks
+
i. Crabs episode

Man + Time Traveller +


Semi-Terrestrial
Animal + “Amphibian” Crab +
This cumulates the logical progression of man/mammal and land
animal/sea animal—compare the “kangaroo” episode left out.17
(Clearly, the missing link in and paradigm behind The Time
Machine, that is, the opposition between man and land animal,
is to be found—as are many other aspects—in Gulliver's Travels.}

FIGURE 4
c. Eclipse episode

Land Animal 4" Time Traveller +


Sea Animal + “Thing”
+
17. Wells’s 1894 essay "The Extinction of Man” (see note 2) discusses this inversion of
power roles with man succumbing—among other possibilities, all later used in his SF—to
huge land-crabs or unknown sea monsters: "In the case of every other predominant
animal the world has ever seen, 1 repeat, the hour of its complete ascendency has been
the eve of its complete overthrow,” he concluded in the hyperbolic night-and-day im­
agery he was to use in his SF too.
TQE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 235

In the last episode, the tentacled “thing” does not attack the
Time Traveller because he flees in time, but it is clearly the
master of that situation. It is reinforced by the additional pres­
ence of liverwort and lichen, the only land survivors; of the deso­
late inorganic landscape; and of the blood-red Sun in eclipse,
which suggests the nearing end of Earth and the whole solar
system. The episode, as has been explained, telescopes the tax­
onomic progression of land animal/sea animal, animals/plants,
organ ic/inorganic, existence of the Earth and the solar system/de-
struction of same—the last being left to the by-now-conditioned
extrapolative mechanism of the reader.
The progression is a “black” progression, or regression, also
insofar as both parties of any preceding paradigm are subsumed
as prey in the succeeding one. All classes of mammals are (sym­
bolically, by way of the Time Traveller as Everyman) the prey of
crabs in (b); all land animals, even the “amphibious” crabs, have
by (c) succumbed to more vital and primitive sea animals, mosses,
and lichens; and in the suggested extrapolation of a destruction
of Earth and/or the solar system, all life would succumb or “be­
come prey” to inorganic being, cosmic processes, or just entropy.
The general scheme of Wellsian SF, true of his whole early
period of SF novels and stories, is thus:

FIGURE 5
(1) Social-Darwin 1st Britain (2) The Future

Predator Prey Predator Prey


Higher
Evolutionary
Class + Higher
Evolutionary
Class +
Lower
Evolutionary
Class
+ Lower
Evolutionary
Class
+
It is logically, genetically, and genologically relevant to com­
pare this scheme to the basic oppositon in More’s Utopia and
Morris’s News From Nowhere—18 it should be remembered the
) 8. In News From Nowhrrt the formally identical opposition is no more spatial (as with
More) but temporal, an opposition between present and future but a hopeful rather than
a black one (as with Wells). Morris’s norm of a utopian future is historically, in fact, the
^36 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

first reaction of the Time Traveller was to suppose lie had found
a pastoral communism. The relevant oppositions here are: Eng­
land is empirically present, but axiologically empty or bad; uto­
pia is empirically absent {ou-topos, nowhere) but its values are
axiologically affirmed ieu-topos) or present. The oppositions Lo-
cus/Value and Present/Absent give rise to the following scheme,
and it should be no surprise that it shows Utopia or Nowhere as
(inverted) mirror images of England in Wells’s here-and-now:
FIGURE 6
(1) Britain (2) Utopia
Present Absent Present Absent

Locus + Locus +
Value + Value +
Utopia as a literary genre is defined by a radically different
location and a radically more perfect community, by an alterna­
tive formal framework functioning by reference to the author’s
empirical environment. As it is argued in chapter 3, No-place is
defined by both not being and yet being like Place, by being the
opposite and more perfect version of Place. It stands on its head
an already topsy-turvy or alienated world, which thus becomes
dealienated or truly normal when measured not by ephemeral
historical norms of a particular civilization but by “species-spe­
cific” human norms. Utopia is thus—like The Time Machine but
also quite unlike it—always predicated on a certain theory of
human nature. It takes up and refunctions the ancient topos of
mundus inversus: utopia is a formal inversion of significant and
salient aspects of the author’s world, an inversion which has as
its ultimate purpose the recognition that the author (and reader)
truly live in an axiologically inverted world.

hinge between the norms of Renaissance utopia (spatial and optimistic) and Wellsian SF.
Together with the liberal optimism of, say, a Verne, Morris's norm was thus what Wells
was reacting against by returning to Swift's horizon. See more about this at the end of this
chapter as well as in the preceding historical chapters. However, it should already be
clear that the two main influences on modern SF are sociological optimism and an­
thropological pessimism. Wells’s career oscillates between these two.
THE TIME .MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 237

In More’s Utopia the analysis of space without value in book 1


(England; "here”) is opposed to an axiologically “full” space pre­
sented in book 2 as Hythloday’s revelation (Utopia; “out there,”
isomorphic with folktales of a just and abundant place beyond
seven seas). This induces the logical and magical possibility of
arriving at that utopia in space or time, of finding or construct­
ing a sociopolitical Earthly Paradise. On the contrary, the general
predatory paradigm of Wellsian SF is finally ambiguous. In its
inversion of Social Darwinism, it supplies a subversive shock to
the bourgeois reader; in its use of the parameters of Social Dar­
winism, however inverted—that is, of the anthropological vision
dividing all life, including man, into predator and prey—it sup­
plies a subversive shock to the humanist and socialist reader, it
is anti-utopian or black. Wells’s seriation converges upon absence
of value and existence, a lay hell:

FIGURE 7

+
+
as opposed to the utopian terrestrial paradise of More and
Morris:
FIGURE 8

+
+
3. The Proportions of Power, and
A Return Into History
One final aspect of The Time Machine's structure, which is again
a methodological key for much SF, is the use of proportion or rule
of three (A is to B as B is to C). Wells himself admitted he used
this extrapolative method—arriving at C given the known in­
crease from A to B—to depict the megalopolis in When the Sleeper
298 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

Wakes, but lamented its arithmetic linearity and obviousness as


“thoroughly wrong.”19 As often, he was right in the particular
and wrong in the general, for proportion does not have to be
used as a simple arithmetic rule of three or linear extrapolation:
that is merely its more primitive variant (see chapter 4, section
2.5.). Indeed, proportion as a specific method for formalizing
classification and seriation seems to be one of the basic ap­
proaches and inescapable epistemological tools of SF, if this liter­
ary genre is based upon the cognitive estrangement induced by a
significantly different novum (figures and/or loci) of narration.
For “significantly different” means also “belonging to another
classifying category” (sociological, biological, anthropological,
cosmological, and so on).
In both the Darwino-Huxleyan and the Wellsian biological se­
ries from section 1 of this chapter, each subsequent term cumu­
lates, as was noted in section 2, the two preceding terms, thus
setting up a new opposition on a more primitive evolutionary
level. Thus in the Darwinian series placentals embraces both man
and nonhuman primates in opposition to marsupials, warm­
blooded subsumes mammals and birds in opposition to cold­
blooded, and so forth. The peculiarity of this seriation, Wells
saw, is that its two extremes, Man and Nonexistence, are
privileged members. In both series, Man is the highest evolu­
tionary category, and he is contained in all the left-hand mem­
bers of both Huxley's and Wells’s series. In the Darwino-
Huxleyan series, Man is placental, mammal, warm-blooded,
amniotic, land animal, vertebrate, and so on down to organic
being. In the Wellsian series, Man is mammal, land animal, ani­
mal, organic being, and existing entity as against the respective
negations. Any negation can come into play only when opposed
to Man, that is, when the Time Traveller in the last two episodes
beholds for himself, and for the human readers, the crabs, the
“thing” from the sea, the lichen and liverwort, rhe desolate rocks
and sea, and the eclipse. At those points, the Time Traveller is a
generic representative of Homo sapiens, an Everyman defined
in terms of biological rather than theological classification, as a
species-creature and not a temporarily embodied soul. The
medieval Everyman was an immortal soul in a mortal body; the

19. WeJis. Tht Futurt in Amtñca, pp. 11-12.


THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 239

principle of individuation was saved. Symmetrically but inversely


the Darwinist Everyman is a quasi-immortal species (germ-plasm)
in a mortal body; the principle of individuation is lost. But
quasi-immortal is not immortal:20 even the generic principle
could get lost, and biology is ftdl of cautionary tales about domi­
nant species or whole orders (such as the giant reptiles, sig­
nificantly absent in The Time Machine though a staple of much SF
since Jules Verne’s Journey to the Centre of the Earth) that disap­
peared in the depths of geological time. The mighty are hum­
bled indeed in such perspectives, which Wells turned against
Victorian complacency with zest and relish.
In the Eloi-Morlocks episode, however, the Time Traveller
cannot be simply a representative Man, since he is faced with
creatures that are maybe no longer Homo sapiens, but are cer­
tainly other races or species of the genus Homo (say Homo eloii
and Homo morlockius).21 What he can be, as against the unifunc­
tional Hominidae of year 802,701, is a complex Victorian
gentleman-inventor who displays various fin-de-siècle attitudes
when faced with shifting situations and interpretations of the
Eloi and the Eloi-Morlocks relationship. Christopher Caudwell’s
crilique implying that the Time Traveller occupies an inter­
mediate position between the two new species, a position isotopic

20. See, e.g., Wells’s anonymous article "Death" in 189S, reprinted in Philmus and
Hughes, eds.. which makes exactly this point.
21. Tangentially be it remarked that T. H. Huxley explained at length in his
works—and no doubt tn his lectures—the difference between physiological and
morphological species, which is intricate and mainly resolvable by experimental
crossbreeding. In Weena. the mawkish avatar of Dickens's Little Nell and similar
Victorian girlish heroines, Wells supplied the Time Traveller with a somewhat imperfect
subject for such experimentation. Wells's private and later literary efforts at sexual libera­
tion prove, 1 think, that he passed up the clearly present sexual considerations only out of
deference to very strong social taboos. One has to regret this, though one can blame him
the less, considering that the taboo has been prudently respected in SF until the end of
the 1960s in the most ludicrous ways (sec for the example of the Tarzan-Jane relation­
ship Richard D, Mullen, "E. R. Burroughs and the Fate Worse than Death," Riverside
Quarterly 4 [June 1970]: 186-91); and in the last decade this sex taboo has often been as
ludicrously and immaturely infringed. It almost goes without saying that French SF was
different, from L'Ève future by Villiers de t’Isle Adam to Vercors’s Murder of the Missing
Like (Les Animaux dénaturés), which has exactly this experiment in miscegenation—
performed by a Daniel Ellsberg among zoologists to force a test-tri al for antigenocidal
purposes—for its theme. Still, it is a loss that Wells never really fused his sexual liberation
novels and his scientific romances into xenoerotics of the Rosny Aîné, Doré mieux, or
Farmer type; but here, too, he provided at least an "empty" model.
240 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

with the position of the petty bourgeois Weils disdainful of a


decadent upper class but horrified and repelled by a crude lower
class, seems to me, for all the nuances and elaborations it needs,
to remain a key for interpreting the topographic and the color
symbolism of that episode.22 In the later episodes the Time
Traveller is placed in an ever widening temporal perspective that
corresponds to the descent down the phylogenetic series.
Returning to the proportions inherent in The Time Machine,
it becomes clear that section 2 of this chapter is a synchronic
presentation of the devolutionary diachrony discussed in sec­
tion 1. On each devolutionary level in Wells’s narration there is
a symmetrically inverted situation usually mediated by the Time
Traveller, as:
(a) Victorian upper class dominates over the Victorian work­
ing class, which is the inverse of the Eloi’s position in respect to
Morlocks; or VUC : VWC = [E : MJ“1.
(b) In the present geological epoch, Mammals are more pow­
erful than “Amphibians,” which is the inverse of what the Crabs
are to the Time Traveller; or M : A = [TT : C]_1.
(c) In the present geological epoch, Land Animals prevail in
the mastery of the globe, which is the inverse of the panic flight
the “thing” from the sea puts the Time Traveller to; or LA : SA
= [TT : Th]-J. The “normal” hierarchical relation Animals :
Plants is inverted in the survival of hardy lichen and liverwort
after animals have died out (though there is a sleight-of-hand
here since the sea “thing” has not died out, thence its first inani­
mate appearance), thus A : P — [TT : L&L]-1. The same holds
true for the relations Organic : Inorganic Existence, and finally
for Existence : Entropy.

22. See Caudwell (Bibliography V). For Wells’s view of any historical “social edifice"
as divided into a basic "labouring class" and a "superior class"—a view clearly echoing The
Communist Manifesto, probably by way of Morris—see his Anticipations, pp. 75-83; that
passage leads into a discussion on pp. 83-91 of the parasitic decadence of both the
fin-de-siede upper and lower classes, the ’‘shareholders" and the "abyss” (the latter a
metaphor that Jack London was to pick up in The People of the Abyss and The Iron Heel).
That such a "future decadence” is the source of the Eloi-Morlocks episode is explicitly
brought out in the conversation with Theodore Roosevelt at the end of The Future in
America. However, the Anticipations passage continues with a long discussion, pp. 92 ff., of
the rise of a new middle class of educated engineers and scientific managers that holds
the only hope for rhe future. Wells's subsequent ideological career is, as Caudwell righdy
remarked, a search for this third social force.
THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS 241

The unity of this series of proportions and of The Time Ma­


chine’s composition hinges on the full—though somewhat unmoti­
vated because ideologically mystified—sympathy the Time Trav­
eller must, for all his cavils, feel for the Eloi. They occupy the
same position, which is simultaneously characterized by
functionally appertaining to the upper class and yet being power­
less in relation to the “lower” Social-Darwinist class of Morlock,
Crab, “Thing," and so further into cosmological entropy. The
series of proportions can be recapitulated as:
VUG : VWC = [E : Mp*
M : A = [TT : C]"1
LA ; SA = [TT : Th]’1
A : P = (TT : L&Lp1
(and so on)
One notices the isotopism of the Eloi and the Time Traveller. It
springs from the fact that the horror of this inversion of basic
power norms is predicated upon the Time Traveller’s being also a
mammal, a land animal, an animal, an organic being, and finally
an existing entity; this is a continuation and validation of the Eloi
and Morlocks being descended from, apparently corresponding
to, and certainly being magically associated with the Victorian
upper and lower class. The different sociological, biological, and
cosmological oppositions in the Wellsian series have the final
common denominator of dominant existence or power. Power is
the arbiter, fate, or nemesis in SF—and in utopias as social-science
fiction.
As was suggested in section 2, the paradigm for these propor­
tions is undoubtedly supplied by the relationships in Gulliver's
Travels, in particular the black Swiftian variant of book 4: Horse :
Houynhnhnm = Man : X. “X" is the animal rationale Swift be­
lieved might perhaps be found in some individuals but not in
mankind as a whole. Though there might be a few noble men
rationis capax, such as Captain Mendez or Lord Munodi, there
was no rational Noble Man corresponding sub specie humanitatis to
the Noble Horse, and the only fully consistent categorical alter­
native was that mankind is Yahookind, animal implume bipes22 Be-

23. For a first approach to the complicated categories and proportions in book 4 of
Gulliver’s Travels, see Elliott (Bibliography II). Brady, ed.. Greenberg, ed., and Tuveson,
cd. (all in Bibliography HI E). especially the essays by R. S. Crane in the Brady volume
242 THE TIME MACHINE VERSUS UTOPIA AS STRUCTURAL MODELS

yond that, section 2 tried to show how Wells’s paradigm leads


back to More’s Utopia. Between More’s inversion from axiologi­
cally bad to good, and Wells’s inversion of biologically dominant
to dominated, Swift’s absence of the Noble Man (synthesizing
axiology and biology) provides the middle term. This section on
proportion might fittingly conclude with noting the secular
proportion:
Utopia : Gulliver’s Travels = Gulliver’s Travels : The Time Machine.
In terms of ideological vision—though perhaps not in terms
of formal accomplishment—this is also a devolutionary series.
Beside Man, Nonexistence or Death is the second privileged
member of Wells’s scale. Finally, as in the magnificent eclipse de­
scription, the two allegorical protagonists Everyman and Death
meet again: and it is not Everyman who wins. Having adopted
such horizons, modern Anglophone SF from Stapledon to
Heinlein or Orwell, Pohl or Aldiss, Vonnegut or Ballard had to
concentrate on filling in Wells’s paradigm and varying its sur­
face. The only other course would have been to return, on a
higher bend of the spiral, to the original Morean paradigm: a
turn that it has so far been either unwilling or unable to take.
For better or for worse (as Russian literature is said to have
sprung from Gogol’s Overcoat), all of this SF has sprung from The
Time Machine. Furthermore, even the temporary or final aban­
doning of S F in favor of futurology, popular punditry, and more
prestigious literary genres is another case of W'ells’s paradigmatic
position in relation to the sociology of SF (from Asimov and Pohl
to Vonnegut and Ballard). And in Wells’s concomitant utopian
speculations (so much more muddled and less cognitive than his
early SF) there can be found the reasons for and social roots of
such unwillingness, inability, and abandonings.

and Joseph Horrell in the Tuveson volume. See for these categories, as well as the
evident connections between Swift and More, also my chapter 5.
Aristotle points out in Poetics 1457b that any metaphor is either (I) a relation of species
to or within genus, or (2) an analogy that always presupposes the A : B = C : D proportion.
In the theoretical chapters of this book I attempted to demonstrate that significant SF
operates on the analogical model. Et is thus understandable that the clearest paradigm of
SF is to be found in Gulltver's Travels, and that Wells closely followed it. He revived the
paradigm by substituting Darwinist evolution for Swiftian or Christi an-humanist ethics,
and the classification methods of arbor huxleiana for those of arbor porphyriana—Huxley
being to Darwin what Porphyrius was to Aristotle (on the Porphyrian tree see Waiter J.
Ong, S. J., Ramus, Method, and ihr Decay of Dialogue [Cambridge, MA, 1958), pp. 78-79
and passim).
11

Russian SF and Its Utopian Tradition

1. The tradition of SF is a time-honored one in Russia. Its


strength is based on blending the rationalist Western European
strain of utopianism and satire with the native folk longings for
abundance and justice. These were embodied, first, in the
ubiquitous dream of a land of Cockavne-like abundance, often
the goal of extraordinary voyages (in the case of landlocked Rus­
sia, overland ones) toward India, Persia, or China: Marco Polo’s
Cathay was not more fabulous than the luxurious Kitezh-gorod
of folk imagination. Justice to man regardless of the traps and
trappings of his social station is the central interest of a second
large segment of oral and folk literature. Perhaps it might be
enough to mention here only the strong theme of the humble
person who is finally exalted. From the wishful and magical
folktales about Ivanushka the Fool, the youngest or third son
who is poorer and apparently more stupid than his brothers but
ends up more successful than the norms of class society would
allow for, through a fusion with plebeian (especially heretical
and sectarian) Christianity, such a theme flowed into the main­
stream of modern Russian literature. From Pushkin’s and Mus­
sorgsky’s mad folk prophet in Boris Godunov, through the mem­
orable humble arrogants in Dostoevsky, (say, Prince Myshkin in
The Idiot), all the way to many Tolstoian and Chekhovian charac­
ters, these figures bear utopian values into a world not yet ready
for them.
In the middle of the nineteenth century this tradition fused
with earlier echoes of imaginary sociopolitical set-ups. The first
traces of an exemplary political tale used as vehicle for a fictional
blueprint go as far back as Ivan Peresvetov’s sixteenth-century
Legend of Sultan Mahomet, a plea to Tsar Ivan the Terrible for
strong state centralization, which was to be remembered approv­
ingly and appropriately by Stalin. But a real trend in that direc­
tion developed in Russian literature following the continent-wide
243
. 244 RUSSIAN SF AND ITS UTOPIAN TRADITION

spread of the Rationalist Staatsroman in the eighteenth century.


From the middle of that century on, utopian writings such as
Lvov’s and Levshin’s extolled ideally harmonious countries of
economic abundance and enlightened absolutism. Tredyakovsky
adapted Fénelon’s Télémaque', writers like Kheraskov, Dmitriev-
Mamonov, and Emin supplied in their works parallels and para­
bles on such an “enlightened absolutist prince" model, usually in
a spurious classical setting. In the second half of the century,
prince Mikhail Shcherbatov wrote an incomplete Voyage to the
Empire of Ophir expressing the longings of the higher aristocracy;
and More’s Utopia was translated into Russian in the year of the
French Revolution. That event, as well as the ensuing wars,
prompted a speedy suppression of the whole genre for a gen­
eration. Empress Catherine, who had affected friendship with
Voltaire and Diderot while quenching the peasant rebellions,
brutally supressed any “natural right" pleas against serfdom—
such as Aleksandr Radishchev’s famous Journey from Petersburg to
Moscow, which contained passages of a democratic blueprint for
the future.
2. It was only in the 1820s that the official propagandist
Thaddeus Bulgarin could publish in his magazine three mildly
satirical marvelous voyage tales. The most substantial of them,
his Untrue Un-Events, or a Voyage to the Center of the Earth (1825), is
supposedly the manuscript of an anonymous narrator’s voyage
through three underworld countries—an idea taken from Hol-
berg’s Niels Klim (also translated into Russian in the eighteenth
century, but having in turn for antecedents ubiquitous myths
and folktales which were to be revived for SF 40 years later by
Jules Verne). Bulgarin’s narrator falls down a cave in an arctic
island into the country of Ignorance, where grotesque inhabi­
tants live in perpetual darkness, caring only for food, drink,
gossip, and gambling, A whirlpool takes him into the dawnlight
country of Beastliness, where apelike inhabitants are half-blind
and have some superficial and mistaken acquaintance with arts
and sciences. Finally, the narrator finds an underground passage
to the nethermost country of Enlighted-ness perpetually lit by the
fires of Earth's center, where humans live in the capital city of
Utopia and outlying villages. The Enlightened are brought up to
be self-disciplined, obedient to lawful authority and to a strict
code of life which regulates working, writing, traveling, and even
RUSSIAN SF AND ITS UTOPIAN TRADITION 845

women’s fashions. Obviously, though he differentiates his coun­


tries by spiritual “enlightenment," Bulgarin's country of Igno­
rance is a satire on the lower orders, mainly on the Russian peas­
ant living under what Tolstoy will—using the same metaphor—
aptly call “the power of darkness.” The country of Beastliness, in
which people are in a way worse than the Ignorantians because
they pretend to knowledge they do not possess, is in many ways a
put-down of the middle class and the intellectuals just emerging
in tsarist Russia. Finally, the smug patriarchal country of En-
lightedness is an autocratic emasculation of More, silent on his
basic insights about property and economics, propagating an
idealized stance popular at the tsarist court from the times of
Peter the Great. With all its obvious limitations, Bulgarin’s work
nonetheless testifies to a certain interest in the field and to a
rather good knowledge of its stock devices, and its obscurantist
depicting of social classes as separate races and nations prefig­
ures Wells’s triumphant use of this device in The Time Machine.
I'he Russian SF tradition came up to worldwide nineteenth­
century standards with Prince Vladimir Odoevsky’s unfinished
Year 4338, circulated in manuscript in the 1840s and published
only in Soviet times. In this epistolary tale the Enlightenment
tradition—visible in the reliance on new knowledge as well as in
the form, taken from .Montesquieu's Persian Letters with a dash of
Mercier’s Year 2440—is supplemented with Romantic extrapola­
tion by Odoevsky, a disciple of Hoffmann and Pushkin as well
as of Schelling. Furthermore, although SF had—in a generally
more activistic Europe affected by the French Revolution and
stubborn utopian hopes—even in lagging Russia already shifted
from space to time, as, for example, in Bulgarin’s True Un-Events,
or Voyages in the World of the 29th Century (1829), yet Odoevsky's
tale was the first significant Russian anticipation. Not that this
rich and erudite eccentric, amateur scientist, philosopher, and
litterateur went any further than to envisage an enlightened aris­
tocratic empire rewarding both birth and talent. Human nature
was for him rather static, and in social relations he envisaged
only more of the same: a conurbation extends from Petersburg
to Moscow; having sold everything else under the sun, Ameri­
cans are notv auctioning off their cities and trying to loot China,
this being the only armed conflict left in the world. But Russian
officers and peasants returning from the Napoleonic wars had
246 RUSSIAN SF AND ITS UTOPIAN TRADITION

brought a spark of novelty home, and at least once in the inter­


vening years, in the Decembrist revolt of 1825, this spark had
threatened to set fire to the antiquated tsarist structure. For
Odoevsky, the new is to be expected from effects of scientific
development, about which he is quite sanguine. In his notes to
the fragmentary Year 4338 often as interesting as the torso itself)
he criticizes Mary Shelley’s gloomy The Last Man for not looking
far enough into the future. He envisages aerostatic communica­
tion radically changing the commerce, politics, and morals of the
future; expeditions for mineral exploitation of the Moon, taking
their own atmosphere along and solving the problems of global
overpopulation; renewed nomadism; machine authorship of
novels and patriotic plays, with other printing and fiction giving
way to pure information, “electrical" talking, and public lectur­
ing; and scholars who do not publish but lecture daily in a dozen
places flying from each to each.
Fhe novel itself is supposed to have been written with the help
of a mesmeric visionary, who had tuned into the mind of a
young Chinese writing back to a friend in Peking from the
Petersburg of 4338 or 4339, on the eve of a comet’s crash onto
Earth, Odoevsky obviously never dreamed of a social upheaval
of such proportions as to change the hallowed tsarist name of the
city. But again, he interestingly discusses such matters as the use
of "elastic glass” or “elastic crystal" for clothing, odoriferous
gases instead of wines, photocopied newsletters by the important
families, teleprinters, air and underground electric traffic, elec­
tric lights, heating of northern countries by giant hot air tubes
from the Equator, and the breeding of horses down to the size
and nature of lapdogs. The main high society game is a “mag­
netic bath” inducing somnambulism, in which state many people
declare their secret actions and sympathies, making hypocrisy
impossible. Clearly, the society envisaged is hierarchical and
bureaucratic. In one of his notes Odoevsky—who is quite sarcas­
tic about middle-class equality—describes a Platonic scientific
oligarchy, headed by poet-philosophers. They are flanked by his­
toriographers, linguists, physicists, and other scientists of rhe
second highest rank. Each historiographer heads in his turn a
group of chroniclers, philologists, geographers, archeologists,
and so forth; each physicist a group of chemists, mineralogists,
and other third-rank scientists. A mineralogist has assigned to
RUSSIAN SF AND ITS UTOPIAN TRADITION 247

him fourth-rank metallurgists, and so on down to the lowest rank


of copyists and laboratory assistants, Odoevsky's Year 4338 can
thus be thought of as a liberal-aristocratic ansiver to Bulgarin,
pioneering in scientific extrapolation into the future, but failing
to realize that a radically new productivity demands radically
new social relations. Since even much of present-day SF is still
stricken with the same blindness, Odoevsky remains one of the
more interesting SF writers of the pre-Wellsian age in Europe,
particularly if one also takes into account his anti-utopian and
anti-bourgeois stories “The Nameless City” (bp, 1844) and “The
Final Suicide” (1844), which are the obverse of his rationalist
optimism.
3. The fact that even such politically timid anticipation could
not be printed under tsarism testifies to the unfavorable climate
of obscurantism SF had to contend with in Russia. Thus the
significant fusion of political utopianism and anthropological an­
ticipation, which came about in Nikolai Chernyshevsky’s What Is
To Be Done? (1862), saw the light of the day from the dungeon in
which it was written only because of a bureaucratic snarl-up in
tsarist censorship; even so, it did not get into book form until the
1905 revolution. Nevertheless, its impact was immense, compar­
able only to Bellamy's in the United States (and in Russia too);
every high-school girl read handwritten copies under her bench,
and for half a century the populists and later the socialists used it
in underground education. Much of the veneration was initially
attributable to the legendary figure of the author, the leader of
Russian radical intelligentsia, who was to remain in prison or
strict Siberian internation for the rest of his life. A brilliant critic
and pre-Marxian materialist who uncompromisingly rejected
both tsarism and the capitalism toward which the regime was
groping, Chernyshevsky cautiously smuggled into the novel his
ideal of a cooperative and communal libertarian socialism, in
which the self-enlightened interest of each would be guaranteed
by the free development of all, based on the liberation of labor.
But the main reason for the novel’s lasting success in stagnant
eastern Europe (both Lenin in Russia and Dimitrov in Bulgaria
glowingly remembered its decisive influence on their personal
commitment to the revolution) was Chernyshevsky’s refusal to
separate the public and private lives of his heroes. "Freedom
now” could have been the slogan of these “new people,” and
248 RUSSIAN SF AND ITS UTOPIAN TRADITION

while devoting themselves to the radical cause, they prefigured


in their exemplary, free, and sincere personal relationships Rus­
sia’s utopian future. Despite some old-fashioned literary conven­
tions used by Chernyshevsky, an admirer of George Sand and
Thackeray, his novel remains of considerable interest. The liber­
ation of mankind symbolized by Woman is its main theme, and
in the loving happiness Vera Pavlovna pursues there is no break,
although there are interesting tensions, between the erotic and
the political aspects, just as there is none between the conscious
and the subconscious. In fact, the utopian anticipation is for­
mally present in a sequence of four dreams Vera Pavlovna
dreams within the novel, a brilliant innovation soon to be picked
up and turned against its originators by Dostoevsky’s rebuttal of
radicalism in Raskolnikov.
Most clearly utopian in the strict formal sense is the famous
Fourth Dream. A vision in two parts, divided by a self-censored
space which stands for the account of the liberating revolution,
this dream shows, first, the liberation of female personality
through a series of allegorical glimpses leading to the new
woman born of revolutionary equality, and second, a future
socialist utopia incorporating both Fourier’s idea of cooperative
producing collectives (phalanstères) which would annul the con­
tradictions between town and rural life, and Owen’s idea of a
new moral world based on equality and made possible by
machine productivity. These ideas were widespread in
nineteen th -century Europe and America, but Chernyshevsky
was, after Cabet, one of the first to use them in fiction. More
importantly, he was the first largely to avoid didactic dryness by
making them the supreme emotional interest of his characters.
The political dream is also the deepest personal dream of a
warm heroine, in a manner which would have been entirely
understandable to Dante, Langland, Cervantes, or Marvell, but
which had since disappeared from the European cultural
mainstream. Chernyshevsky’s Fourth Dream fused romantic
pathos and rational belief in social change. The huge crystal
palace in the midst of fertile fields where a happy association of
free producers lives and w’orks, each at variegated tasks changing
every few- hours, symbolizes even architecturally—by uniting
under the same roof a variety of apartments, workplaces,
studios, theaters, museums—the harmony of personal and public
RUSSIAN SF AND ITS UTOPIAN TRADITION 249

life. The dream of Vera Pavlovna is thus not unworthy of the


great liberating current of “warm” utopias in western Europe,
from More and Rabelais to Chernyshevsky's teachers—Rousseau
and the utopian socialists, who set up the principle that the
emancipation of women is the measure of social liberation. It
certainly set the tradition of a whole aspect of Russian SF.
A countercurrent to radicalism, equally messianic and anti­
bourgeois but with a diametrically opposed point of view, found
a most powerful voice in Fyodor Dostoevsky. In his youth a
member of an illegal circle which propagated utopian-socialist
ideas and even tried to put them into practice, Dostoevsky’s shat­
tering prison and Siberian experience channelled his utopian
concerns into a mystical deification of the tsarist system. His
deepest hatreds were from that time evenly divided between the
degradation of man under the impact of capitalist economics and
the radical proposals for rational rehumanization; curiously, he
had come to think of these opposites as two faces of the same
prideful coin. The famous Crystal Palace of the mid-century
London World Exhibition became for him the symbol of indus­
trialized inhumanity, dividing brothers into the domineering rich
and the gin-sodden, sectarian poor. In his Noles from the Under­
ground, written immediately after the publication of What Is To
Be Done?, he sarcastically adduced the irrationality of the indi­
vidual’s free will, the stupidity evidenced by unceasing mass
bloodshed, and the general senselessness of world history against
the rationalist builders of new economic relations and of the
Crystal Palace. Although unnamed in these repressive circum­
stances (where even Dostoevsky was censored), there is little
doubt that Chernyshevsky’s novel was meant. In Raskolnikov,
the hero of Dostoevsky’s next novel, all radicals recognized a
distorted portrait of themselves.
Though he was energetically battering away at the idea of a
future Crystal Palace, the ambiguous Dostoevsky remained ob­
sessed with its themes of innocence, brotherly love, and tran­
scending social antagonisms. True to his religious bent, he
placed them in a static, arcadian Golden Age, a motif which was
to run through his whole w’ork, breaking out explicitly in places
like Versilov’s speech in A Raw Youth, the suppressed chapter of
The Possessed, and finally formalized into a separate work as The
Dream of a Ridiculous Man (1877). That story is a far cry from
250 RUSSIAN SF AND ITS UTOPIAN TRADITION

the venomous polemics of the 1860s. Although Dostoevsky re­


mained skeptical about the feasibility of man’s salvation through
history—witness the Grand Inquisitor passages in The Brothers
Karamazov—The Dream of a Ridiculous Man is a heartfelt cry for
the necessity of such a salvation. It depicts a distraught narrator,
who has lost all touch ivith humanity and who decides to commit
suicide, dreaming that he has died and been transported across
space to a perfect twin of our Earth. There, on a Greek island,
he finds an unfallen and loving people in a pastoral Golden Age;
yet after a while he corrupts the happy Utopians through lies,
cruelty, and individualism. A complete civil society soon evolves,
with crimes, science, codes of honor and law, warfare, and a full
history of slaves and saints, suffering and formal religion. The
horrified narrator asks to be crucified but is laughed at, and
awakes. He abandons his suicidal intention and devotes the rest
of his life to preaching the possibility of happiness and beauty on
our planet. Dostoevsky’s fulminations against the utopian aboli­
tion of suffering notwithstanding, for once the spell of an earthly
happiness has managed to assert itself in this neglected gem of
his: a tardy, isolated, and wistful but significant concession to the
dream of Vera Pavlovna.
4. Never absent from Russian literature, the anticipatory al­
ternative became especially relevant at times of revolutionary re­
surgence, which also chipped awaj' at censorship. Such a period
were the 1900s, when the utopian flame flared up in many
non-SF works, for which the future-oriented daydreams of some
Chekhovian heroes, such as Vershinin in The Three Sisters, may be
taken as representative. In circumstances that called for active
intervention into the here and now, the traditional far-away uto­
pian country, such as the one expounded by the Tolstoian folk­
preacher Luka in Gorky’s Lower Depths, was already felt as evad­
ing the issue. The major SF witness to the urgency of intervening
was the prominent symbolist poet Valery Bryusov. In his play
Earth (1904) a revolt of youth shatters the glass dome that bars
people in a decadent giant city from sunshine and open space;
the revolt is seen both as seeking liberation and exposing the city
to the risk of annihilation. At the same time as London’s more
precise The Iron Heel, Bryusov thus saw the distant possibilities of
a new culture and the immediate prospect of destroying the
present one. After the defeat of the 1905 revolution, Bryusov’s
RUSSIAN SF AND ITS UTOPIAN TRADITION 251

horizons contracted; his Poesque story abouL a future city, The


Republic of the Southern Cross (1907), is frankly dystopian: a huge
capitalist metropolis on the very South Pole, capital of the
greatest industrial power of the world and again enclosed by an
impenetrable dome, falls prey to a mortal epidemic of mania
contradicens which makes the afflicted do the opposite from what
they wish to do. Science and society are powerless to prevent its
spread, and a resolute bourgeois minority fighting for order is
finally overwhelmed by the brutalized inhabitants. The story is a
curious préfiguration of Camus’s The Plague, possibly because
both descend from Raskolnikov’s dream at the end of Crime and
Punishment, of a worldwide epidemic, but more significantly be­
cause they both allegorize the great social convulsions of our
century. But Bryusov was readier to pay any price for the de­
struction of the “ugly and shameful” capitalist system. Though
his apocalyptic verse, stimulated not only by other poets but al­
so by Flammarion, Verne. Wells, and Renard, envisaged wolves
baying on the banks of the Seine and happy children burning
books in the London Parliament, he apostrophized "the coming
Huns” (see also his story The Last Martyrs); “But you, who will
destroy me, I I meet with an anthem of welcome!” Together with
his great fellow-symbolist Alexander Blok, who had also envis­
aged the coming of new "Scythians,” Bryusov became thus one
of the few prominent non-Marxist writers who took an active
part in post-revolutionary cultural life in Soviet Russia.
All these examples testify to a continued interest in and a
significant tradition of Russian SF. One could add to them a few
more scattered works: on the one hand, the theocratic
apocalypse of Vladimir Solovyev’s Short History of the Antichrist
(1900), the confused occult interplanetary voyages of Mrs.
Kryzhanovskaia, and Sharapov’s Slavophile. anti-Bellamy utopia
Fifty Years Later (1902); and on the other end of the ideological
spread, Alexander Kuprin’s popular stories “A Toast” (1906)
and “The Liquid Sun” (1913) and some technological anticipa­
tions written with popularizing aims by engineers and scien­
tists—like V. Chikolev’s Electrical Story (Neither True Nor Made
Up, 1895), A. Rodnykh’s Rolling Road (1902), V. Bakhme­
tyev’s The Billionaire’s Legacy (1904), and N. Komarov's Coldtown
(1917)—and culminating in the opus of Tsiolkovsky. Komarov’s
anticipation of a refrigerating technique on a scale to counter­
252 RUSSIAN SF AND ITS UTOPIAN TRADITION

act a rise of atmospheric temperature already led into a tech­


nocratic sketch of the next two centuries’ history. Concentrat­
ing on the sociopolitical aspects of such a history, Alexander
Bogdanov-Malinovsky, a prominent though ideologically unor­
thodox Bolshevik leader, became, alongside Bryusov and Tsiol-
kovsky. the most interesting SF writer of the prerevolutionary
years. His novels about an exemplary Martian society, The Red
Star (1908) and Engineer Menni (1913)—especially the first one—
renewed the tradition for left-wing Russian SF, which had been
in abeyance since Chernyshevsky’s times. They fused earthly
political struggles with the interplanetary tale (Mars is a “red
star” both because of its Wellsian reddish vegetation and because
of the proletarian movement) either directly or by juxtaposition
within the same work as préfigurations, extrapolations, or alter­
native paths. Bogdanov’s superior Martian technology (including
a foretaste of nuclear energy, automation, anti-matter, and in­
terplanetary vessels with ionic propulsion) and social organiza­
tion successfully updated the European “Martian story” of
Lasswitz and Wells in a manner unsurpassed between them and
the American 1940s, or even 1960s. The free, science-oriented
social system will be taken up 50 years later by Yefremov, the
revolutionary protagonist's illumination and love affair on
another planet by Tolstoy, and both these elements by much
Soviet SF in the 1920s and 1960s. Even Bogdanov’s activistic
characters and, in places, lyrical style were to prove trend-setting.
Yet for all its vitality, the prerevolutionary SF tradition was
very tenuous, written at great cost by exceptional, heroic, and
isolated figures. The weight of industrial and scientific back­
wardness combined with the obtuse oppression of tsarism was
loo great to allow a flowering of the genre. In the twenty years
before 1917, only about 25 original Russian SF books were pub­
lished. Nonetheless, together with copious translations and imita­
tions of Jules Verne (by Volokhov-Pervukhin, Semenov, and
Uminsky, for example), as well as Wells, Flammarion, or Renard,
they prepared the ground for the flowering. This came about
with a vengeance in the 1920s, in the first flush of a revolution­
ary regime committed to industrialization and modern science as
a means for achieving utopian mastery over man's destiny.
5. In Russia this was one of those epochs when new Heavens
touch the old Earth, when the future actively overpowers the
RUSSIAN SF AND ITS UTOPIAN TRADITION 253

present, and the sluggish and disjointed flow of time is suddenly


channelled into a wild waterfall, generating a rainbow on the
near horizon and capable of dispensing light and warmth from
scores of dynamos. Wells visited the Soviet stale in the midst of
the Volga famine and found Lenin confidently tracing plans for
a fully electrified self-governing Russia. Quite rightly, he recog­
nized in the author of State and Revolution a utopian dreamer:
but Wells the utopographer had forgotten that certain utopias
are realizable.
5.7. In the literature of the 1920s this atmosphere evoked a
flurry' of anticipations and planetars novels. The tireless Bryu-
sov, long an enthusiast for interplanetary travel, dreamed with
Pasternak and the Futurists of a scientific poetry, and wrote of
billions of worlds ready to hear the call of an Earth spiralling
in its “planetary revolution” (Distances, 1922). Two of his unpub­
lished plays discussed the possibilities and pitfalls of interplane­
tary relations. The tragedy Dictator, probably modeled on Wells’s
When the Sleeper Wakes, portrays a militarist who wants Earth to
conquer the universe, but is overthrown by the partisans of
peaceful labor. The World of 7 Generations (1923) takes place on a
Flammarion-inspired comet, symbolizing the revolved historical
cycle, the inhabitants of which are called upon to sacrifice them­
selves in order to save Earth. Taking off from such a widespread
identification of the social revolution with man’s leap into the
universe, a whole school of Russian versifiers called themselves
the “Cosmists” and extolled a somewhat vague “planetary aware­
ness.” Much too little is known about such sociologically repre­
sentative Chernyshevskian or Bogdanov-type visions of a perfect
classless future as Vivian I tin’s warm Land of Gonguri (1922),
Yakov Okunev's Tomorrow (1924), and several other works of
that kind up to V. Nikolsky’s In a Thousand Years (1927, featuring
intelligent plant life and the first nuclear explosion in 1945), E.
Zelikovich’s Coming World (1930), and the remarkable Yan Larri’s
Land of the Happy (1931), a novel which even contains a satirical
portrait of Stalin. A number of noted prose writers wrote at the
outset of their careers at least one largely or marginally SF work,
often with a setting in the near-future and with strong elements
of a political adventure or crime story, such as some stories of
Nikolai Aseyev, Ilya Ehrenburg’s novel Trust D.E. (1923), Valen­
tin Katayev's Ehrendorf Island (1924), Marietta Shaginyan’s “Jim
254 RUSSIAN SF AND ITS UTOPIAN TRADITION

Dollar" trilogy beginning with Mess-Mend (1924-26), Boris Lav-


renyev’s The Fall of the Ytl Republic (1925), and Vsevolod Ivanov
and Victor Shklovsky’s Yperite (1926).
Another and similar type was ‘‘the catastrophe novel,” which
dealt with the social consequences of a new scientific invention
(often a kind of ‘‘death ray,” and brought about by a solitary
scientist of the Frankenstein-Nemo type) ending invariably with
a catastrophic downfall of the capitalist system and the victory of
world revolution. One can mention, for example, the self-
explanatory titles of Katayev’s novel The Lord of Iron (1925) or of
two works from 1927—A. Paley’s Gulf Stream, in which a socialist
Old World is contrasted with a Taylorite and alienated United
States, and A. Shishko’s The Microbes' Appetite, operating with
robots and chemical warfare. This type of story started from the
Vernean roman scientifique and broadened it into sociopolitical
anticipations of a near future. Its global dimensions latched on to
Wells’s middle-period romances The War in the Air, The World Set
Free, and especially When the Sleeper Wakes, but their primary
.impulse came from the general Soviet anticipation of a world
revolutionary upheaval in the first ten years after the October
Revolution. The undoubted culmination of this type was Alexei
Tolstoy's Hyperboloid of Engineer Garin (translated as The Garin
Death Rar), published in its first version in 1925-26. Tolstoy’s
revisions of this political anticipation over the next dozen years
would almost suffice to provide a microhistory of shifting at­
titudes in and around Soviet SF in that period. Finally, the tor­
mented and ironic aspects of a Zamvatin-like skepticism were
continued by Sergei Bobrov’s Revolt of the Misanthropes (1922),
Lev Lunts’s expressionistic drama The City of Truth (1923), some
stories of Veniamin Kaverin, and especially by Mikhail Bulgakov
(The Fatal Eggs, 1924; The Heart of a Dog, 1925) who soon crossed
into allegorical or satirical fantasy.
Perhaps most representative of this 1920s mainstream were
certain works—even though, or perhaps because, only partly or
marginally SF—of Vladimir Mayakovsky, its most popular poet.
In poems such as “About It," “150,000,000” and “The Fifth In­
ternational,” in short propagandist pieces such as Before and
Now, in film scenarios, and most clearly in his three post­
revolutionary plays, the mainspring of Mayakovsky’s creation
was the tension between anticipatory utopianism and recalcitrant
RUSSIAN SF AND ITS UTOPIAN TRADITION 255

reality. A Futurist and admirer of Wells and London, Mayakov­


sky wrote his witty masterpiece Mystery Buffo to celebrate the first
anniversary of the October revolution, envisaging it as a sec­
ond cleansing Flood in which the working classes, inspired by a
poetic vision from the future, get successively rid of their mas­
ters, devils, heaven, and (in the 1921 version) of economic chaos,
and finally achieve a Terrestrial Paradise of reconciliation with
Things around them. The revolution is thus both political and
cosmic, it is an irreversible and eschatological, irreverent and
mysterious, earthy and tender return to direct sensuous relation­
ships of men with a no longer alien universe. No wonder that
Mayakovsky's two later plays became satirical protests against the
threatening separation of the classless heavens from the Earth.
The future heavens of the sun-lit Commune remain the constant
horizon of Mayakovsky’s imaginative experiments, and it is by its
values that the grotesque tendencies of petty-bourgeois restora­
tion in The Bedbug (1928) or of bureaucratic degeneration in The
Bath (1929) are savaged. Indeed, in the second part of both
* plays, the future—though too vaguely imagined for scenic pur­
poses—irrupts into the play. In The Bedbug it absorbs and quar­
antines the petty “bedbugus normalis” in its bestiary. In The
Bath rhe newly proclaimed Soviet Five-Year-Plan slogan of “Time
forward!’’ materializes into the invention of a time machine that
communicates with and leaps into the future, sweeping along the
productive and the downtrodden characters but spewing out the
bureaucrats. The victory over time was for Mayakovsky a matter
of central political, cosmic, and personal importance: intrigued
with Einstein’s theory of relativity, he firmly expected it to make
immortality possible for men. His suicide in 1930 cut him off in
the middle of a fierce fight against the bureaucrats whom he
envisaged as holding time back and who engineered the failure
of Meverhold’s first-rate production of The Bath.
The seemingly divergent concerns of Evgeny Zamyatin, when
looked at more closely, turn out also to deal with the relation­
ships of the new, future heavens and the old, present earth. The
difference is that Zamyatin did not believe in any eschatological
end of history. An ex-Bolshevik and rebel against tsarism, a
scientist-specialist in ship-building who introduced into his novel
We (1920-21; bp. 1927) the atmosphere of the shipyards and of
the illegal movement, Zamyatin too despised Western capitalism
25Ê RUSSIAN SF AND ITS UTOPIAN TRADITION

as life-crushing. Certain of the features of a novel satirizing


bourgeois respectability and clerical philistinism, which he wrote
in England during World War I, such as coupons for sex and a
Taylorite “table of compulsory salvation” through minutely regu­
lated daily occupations, recur in We. Even after he left the USSR,
his major project in the 1930s was a historical novel about Attila
and the fall of decadent Rome, a situation which he—with
Bryusov and Blok—considered analogous to the East-West con­
flict of our times. Most significantly, for Zamyatin too the rev­
olution is the undoubted sun I ike principle of life and move­
ment, while he bestowed the name of entropy on the principle of
dogmatic evil and death. An anti-entropic science, society, and,
of course, literature is needed, he affirmed, “as a means for
struggling against hardening of arteries, rigidity, moss and peace
. . . a utopian literature, absurd as Babeuf in 1797; it will be
proved right after 150 years.” It is evident that Zamyatin thought
of himself as a utopian, paradoxically more revolutionary than
the latter-day Bolsheviks, since “the truths of today are the er­
rors of tomorrow: there is no final number” (“On Literature, Rev­
olution and Entropy”). It is thus disingenuous to present him as a
primarily anti-Soviet author—even though the increasingly
dogmatic and bureaucratic high priests of Soviet letters thought
of him so, never allowed his novel to be printed in the USSR,
and induced him to leave his country in 1981. Extrapolating the
repressive potentials of every strong state and technocratic set­
up, including the socialist ones, Zamyatin describes a United or
Unique State 12 centuries hence having for its leader “the Ben­
efactor” (a prototype for Orwell’s Big Brother and the situation
in 1984}, where arc has become a public utilitarian service, and
science a faultless guide for linear, undeviating happiness.
Zamyatin’s sarcasm against abstract utopian prescriptions (like
those of the feebler Wells) takes on Dostoevskian overtones: the
threat of the Crystal Palace echoes in the totally rationalized city.
The only irrational element left is people, like the split (Marx
would have said alienated) narrator, the mathematician and rock­
etship builder D-503, and the temptress- from the underground
movement who for a moment makes of him a deviant. But man
has, as Dostoevsky’s Grand Inquisitor explained to Christ, a
built-in instinct for slavery, the rebellion fails, and all the citizen
“Numbers” are subjected to brain surgery removing the possibil-
RUSSIAN SF AND ITS UTOPIAN TRADITION 257

ity of harmful imagination.


However, Zamyatin’s novel is not consistent. Of a bold gen­
eral concept, it hesitates midway between Chernyshevsky and
Dostoevsky—undecided as to what it thinks of science and rea­
son. After the physician and philosopher Bogdanov and the
mathematician Tsiolkovsky, Zamyatin was the first practicing sci­
entist and engineer among significant Russian SF writers. The
scientific method provided the paradigm for his thinking, and he
could not seriously blame it for the deformations of life. In that
case, how is it that a certain type of rationalism, claiming to be
scientific, can be harmful in certain social usages? This question,
which came up in SF with Butler’s Erewhom and Wells’s best
works, Zamyatin was unable to answer except in mythical, Dos-
toevskian terms: there is “only one conceivable victory—to be
crucified. . . . Christ, when factually the victor, becomes the
Grand Inquisitor” (in his essay “Scythians?”). The achievement
of any lofty ideal inevitably causes it to founder in philistinism.
To the extent that We equates Leninist Communism with in­
stitutionalized Christianity and models its fable on an inevitable
Fall from Eden ending in an ironical crucifixion, it has a strong
anti-utopian streak. Zamyatin’s evocative style shifted the focus
to systematic image-building at the expense of the plot, making
thus a virtue of his inability to explicate the chosen situation and
to reconcile its poles of rationalism and irrationalism, science and
art (including the art of love). However, this obscures the prob­
lem of whether any utopia—even a dynamic one that refuses
More’s Platonic and Christian model—must of inherent necessity
become repressive and dehumanizing. Zamyatin’s social ideology
conflicts with his own favorite experimental approach: a mean­
ingful exploration of this theme and situation would have to be
conducted in terms of the least alienating utopia imaginable—
one in which there is no misuse of natural sciences by a dogmatic
science of man.
Yet when all this is said, the basic values of We imply a stub­
born revolutionary vision of a classless new moral world free
from all social alienations, a vision common to Anarchism and
libertarian Marxism. Zamyatin confronts an anti-utopian, ab-
solutistic, military-type control—extrapolated both from the
bourgeois and early socialist state practices—with a utopian-
socialist norm. As he wrote in the essay “Tomorrow”: “We do not
258 RUSSIAN SF AND ITS UTOPIAN TRADITION

turn to those who reject the present in the name of a return to


the past, nor to those hopelessly stupefied by the present, but to
those who can see the far-off tomorrow—and in the name of
tomorrow, in the name of man, we judge the present.” This
point of view differs from Mayakovsky's in The Bath principally
by its ascetic concentration on the deformities, without the ex­
plicit counterbalance of a vague future. Indeed, it is significant
how for Mayakovsky, too, the utopian anticipation draws further
off in time: in poetry, from the twenty-first century of “The Fifth
International” (1922) to the 30th century of “The Flying Pro­
letarian" (1925); in dramaturgy, from the 25 years of the
scenario Forget the Hearth (1927) to the 100 years of The Bath
(1929). Simultaneously with the poetry of Mayakovsky (whom he
called “a magnificent beacon"), Zamyatin brought to Russian SF
the realization that the new utopian world cannot be a static
changeless paradise of a new religion, albeit a religion of steel,
mathematics, and interplanetary Hights. Refusing all canoniza­
tion, the materialist utopia must subject itself to a constant
scrutiny by the light of its own principles; its values are for Zamya­
tin centered in an ever-developing human personality and ex­
pressed in an irreducible, life-giving, and subversive erotic pas­
sion. For all its resolute one-sidedness, the uses of Zamyatin’s
bitter and paradoxical warning in a dialectical utopianism seem
to be obvious.
The language of the novel is an interesting Expressionist
medium vouching for at least some cognitive veracity. It is ma­
nipulated for speed and economy, which Zamyatin himself de­
fined as “a high voltage of every word":
In one second there has to be condensed what before fitted
into a whole minute; the syntax becomes elliptical and airy,
the complex pyramids of the paragraph are scattered into
stone blocks of independent sentences. . . . The picture is
sharply focussed, synthetic, it has one basic trait only, which
could be noticed from a moving car. Provincialiams, neolo­
gisms, science, mathematics, engineering have invaded a vo­
cabulary canonized by usage. [“On Literature, Revolution,
and Entropy"]
Zamyatin is a heretic; in places vague and possibly confused, he
probably fails to attain a fully consistent structure because of the
RUSSIAN SF AND ITS UTOPIAN TRADITION 259

one-sided assumptions which underlie his writing—but he is cer­


tainly not counterrevolutionary. On the contrary, in his own way
he tried to work for a future different from that of the en­
visioned United State. His protagonist is defeated, but the novel
as a whole remains concerned with the integrity of man’s knowl­
edge (science) and practice (love and art). Even the symbol of
V — 1 (which has an equivalent in the retribalized, “hippie”
Me phis is the Green World) in an antithesis to and appeal
against a limited Rationalism (the United State) that does not
simply reject the thesis—as Dostoevsky’s Underground Man
did—but includes it in a higher dialectical synthesis prehgured in
D-503's oscillation between love and the Integral. Yet, as they are
not in Wells, the guilt and possible solution are here placed
squarely on man and not on mvthicai outsiders. Like the formal
model of D-503’s personalized notes, the laboratory conspectus,
the structure of H> remains open to new cognitions, restless,
anti-entropic, and never finally complete. By systematically and
sensitively subjecting the deformities it describes to the experi­
mental examination and hyperbolic magnification of SF,
Zamyatin’s method makes it possible to identify and cope with
them. In his own vocabulary, the protagonist’s defeat is of the
day but not necessarilv of the epoch. The defeat in the novel We
is not the defeat of the novel itself, but an exasperated shocking
of the reader into thought and action. It is a document of an
acute clash between the “cold” and the “warm” utopia: a judg­
ment on Campanella or Bacon as given by Rabelais or Shelley.
5.2. The 1920s saw, parallel to predominatingly social-science­
fiction, the first wave of Russian SF that organically blended
sociological with natural-science fiction primarily oriented to­
ward interplanetary adventures. The great pioneer of both Rus­
sian astronautics and Soviet SF was Konstantin Tsiolkovsky, a
mathematics teacher who by the 1890s had begun his specula­
tions on mankind’s cosmic destiny in the depths of a tsarist pro­
vincial town, writing for propagandist purposes two SF booklets
illustrating a possible happy weightless life on asteroids, the
Moon, and on rocket colonies in space (On the Moon, 1887; Day­
dreams of Earth and Heavens, 1894), The Soviet regime enthusias­
tically look up his unheeded ideas; Tsiolkovsky wrote his best SF
story, Outside Earth, in 1918 and proceeded to develop his scien­
tific plans with public means, becoming the venerated teacher of
260 RUSSIAN SF AND ITS UTOPIAN TRADITION

the future Sputnik and Vostok constructors as well as of SF fans


and writers such as Belyaev. I referred earlier to the general
enthusiasm for a revolutionary “storming of the heavens” in the
1920s, as expressed by Bryusov, Mayakovsky, and the “Cos-
mists,” and skeptically reflected even in Zamyatin’s We as well as
in his story7 “The Most Important Thing” (1924). Astronautic
study circles, public lectures and expositions got under way at
that time; in the midst of the civil war the tireless Lenin, having
read Percival Lowell’s book about Mars and debated the Martian
novels with Bogdanov, went in 1920 to listen to a public lecture
on the project of a cosmic ship, and. talking with Wells, con­
cluded that if mankind ever comes to other planets all our
philosophical, moral and social assumptions will have to be re­
examined. In 1925 Moscow University7 organized a debate on
“The Flight to Other Worlds,” and in 1925-26 the first special­
ized adventure periodicals carrying largely SF appeared, such
as War of the Worlds, Universal Detective, Knowledge is Strength,
and World of Adventure. In the XEP heyday of private publishing
the Soviet book market was flooded by large quantities of trans­
lated European and US subliterary SF, such as E.R. Burroughs’s
Martian cycle; not counting the perennially popular Verne and
Wells, more than 100 novels and stories were translated between
1923 and 1930 (as compared with 155 original Soviet Russian
novels and stories published between 1920 and 1927). In such an
atmosphere the genre was given the accolade of literary quality
and respectability by the well-known writer Alexei N. Tolstoy,
who in his novels Aelita and The Garin Death-Ray blended the
adventure of interplanetary7 flight and conflicts—namely, the
global struggle for a new scientific invention—with a utopian
pathos arising from revolutionary social perspectives in a way
calculated to please almost all segments of the reading public.
This blend was to remain the basic Soviet SF tradition till the end
of the 1960s.
In Aelita (1922), written while Tolstoy was still an emigre pre­
paring to return to the Soviet Union, this blend is endearingly
enriched with a lyrical component, the love of Los, the inventor
of the rocketship, for the Martian princess Aelita. Los, the cre­
ative intellectual, with his vacillations and individualist concerns,
is contrasted to but also allied with Gusev, a shrewd man of the
people and fearless fighter who leads the revolt of Martian
RUSSIAN SF AND ITS UTOPIAN TRADITION 261

workers (the Martians are descendants of the Atlantans) against


the decadent dictatorship of the Engineers’ Council. If the
standard adventure and romance were taken over from Wells
and the contemporary subliterary SF (Benoit’s Atlantis and Bur­
roughs’s A Princess of the Moon probably—both were translated in
the USSR), the politics are diametrically opposed to Lasswitz’s
and Bogdanov’s idea of a Martian benevolent technocracy. Yet if
the workers’ uprising led by a Red Army man was a clear parable
for the times, such as could have been shared by all Soviet SF
from Mayakovsky to Zamyatin, the dejected and somewhat hasty
return which has Los listening at the end to the desperate wire­
less calls of his beloved is as clearly of a Wellsian gloom (The First
Men on the Moon). But this ambiguity, which sometimes strains
the plot mechanics, makes also for a counterpointing richness, an
encompassing of differing attitudes and levels that envisages the
price as well as the necessity of an activist happiness. Since the
novel's plastic characterization, rich and differentiated language,
and consistent verisimilitude (once the reader accepts the under­
lying premise) lifted it to the level of the literary mainstream,
much as Wells had done for the SF novel in Britain, it became
the first universally accepted masterpiece of Soviet SF. In par­
ticular, Aelita raised extraterrestrial utopianism heyond pulp im­
itations of Verne, Wells, and Burroughs (by Mukhanov, Arelsky,
Grave, Yaz.vitsky or Goncharov), to a height not to be reached
again until Yefremov.
Tolstoy’s second SF novel, Engineer Garin’s Death-Ray (in four
versions from 1926 to 1937), is a retreat to the “catastrophe”
novel: Vernean adventures and Chestertonian detections and
conspiracies center around an amoral scientist who beats the
capitalist industry kings at their own game but comes to grief
when faced with popular revolt. It moves fast if jerkily; as
Tolstoy was a trained engineer, its science is believable (atomic
distintegration of a transuranium element is posited as well as
something resembling lasers), and it remains a prototype of the
anti-imperialist and anti-fascist concern in Soviet SF. This con­
cern has always been a vigorous strand in its skein, whether as
direct political satire (as in Ehrenburg or Lavrenyev, or in later
interwar works such as Turov’s Island of Gorilloids, Zuyev-
Ord y nets’s Panurge’s Herd, or Shishko's The End of Common Sense),
as the more dubious Soviet-invention-cum-foreign-spy thrillers
262 RUSSIAN SF AND ITS UTOPIAN TRADITION

cropping up frequently from that time on (for example, atomic


energy in the play Gold and Brains by A. Glebov, 1929), or as the
pamphleteering story popular among postwar writers like Lagin,
Savchenko, Toman, Dneprov, and Varshavsky. In the 1920s
numerous novels adopted the Garin Death-Ray combination of
scientific thriller and politically virtuous anticipation (beside
those already mentioned: A. Yaroslavski’s The Argonauts of Space,
V. Orlovsky’s The Horror Machine and The Revolt of Atoms, N.
Karpov’s Death Rays, S. Grigoryev’s The Fall of Britain, F. Bog­
danov's Twice Born, I. Keller and V. Hirschhorn's Universal Rays,
and others).
Soviet SF of the 1920s had thus established a tradition ranging
sociologically from facile subliterature to some of the most in­
teresting works of “highbrow” fiction produced in that golden
age. It had embraced further and nearer anticipation, global and
interplanetary tales; adventure, politics, utopianism, and tech­
nology; ethics and romance; novels, stories, poems, plays, essays,
and movies (eight silent SF movies were shot in the years before
1926, beginning with Jack London's friz« Heel in 1919, for which
the scenario was written by no less a dignitary than the comissar
for education Lunacharsky; continuing with a rather bad Aelita
in 1924, and ending with The Sale of an Appetite based on Paul
Lafargue's story). Il had sketched in most of the themes and
topoi of modern SF—galactic warfare being conspicuously rare
—including anti-utopianism, automation, and the social conse­
quences of "value-free” natural science. Its common denomina­
tor was a sometimes naive but genuinely enthusiastic, thorough­
going, and humanist critique of old Europe and America, that is
to say, of the short-sighted and alienating capitalist lifestyle.
6. In about 1927 Soviet concentration on national industrial
buildup at the expense of global revolutionary romanticism, with
future sociological horizons jelling into a planned quantitative
growth and thus decreasing in imaginative novelty, began
strongly to favor linear technological and natural-science ex­
trapolation. Within a general European movement toward a
“realistic objectivity” (Neue Sachlichkeil), Soviet literary horizons
grew less cosmic and grandiose: the SF estrangement shifted into
the domain of amazing adventures and/or inventions. World
revolution and far-ranging anticipations were replaced with de­
tailed but flat descriptions of the technologically changed near
RUSSIAN S F AND ITS UTOPIAN TRADITION 263

future or of SF adventures in an abstract, often foreign-


sounding (a convention still strong today) or planetary setting.
This constituted a return to the nineteenth-century traditions of
Verne and Wells. Prerevolutionary Vernean writings had con­
tinued to exert influence, as in the geologist and geographer V.
Obruchev’s two novels about prehistoric environments, Plutonia
(1924, a hollow-Earth setting) and Sannikov-La nd (1926, North
Pole oasis). As for the never-ceasing Russian fascination with
Wells, it could simply turn to his ostensibly shorter-range tales
like The Island of Dr. Moreau or The Invisible Man. In this tra­
dition, the most prominent and still widely read author is Alex­
ander Belyaev, the first Russian writing exclusively SF. In nu­
merous stories and about 20 novels he fused breathtaking
Vernean adventures of a romantically alienated hero with new
and bold scientific themes: biological adaptation and transplanta­
tion, often brain surgery, as in his first novel Professor Dowell’s
Head (1925)—much superior to its French predecessor, Maurice
Renard's Le Docteur Lerne; the impact of various, often humorous
scientific inventions in the cycle of “Professor Wagner,” 1926-36;
and several novels in the 1930s domesticating Tsiolkovski's no­
tions. His most interesting works focus on the resistances to and
anxiety of the novum’s bearer—a scientist with humanistic ideals,
a biologically modified man like the flying Ariel in the novel of
die same title (1941) or Ichthyander in his most popular novel
The Amphibian Man (1928), or quite openly an artist like Presto in
the two variant novels The Man Who Lost His Face (1929) and The
Man Who Found His Face (1940). Such works, imbued with an
aching lyricism and a vibrant humanistic vehemence, remain of
significance today, as do some articles by this penetrating SF
critic. But often Belyaev’s hero triumphs simply thanks to an
essentially fairytale metamorphosis that allows him to vanquish
physical gravity and social injustice. The black-and-white opposi­
tion of his threatened hero to a grotesque capitalist environment
becomes then a form of escapism into a wicked Ruritania, to
which Belyaev added in the 1930s—partly because of political
constraints—detective and spy-thriller elements.
In fact, the promise of the revolutionary years—when it ap­
peared probable that the Russian school (or indeed schools)
would dominate our times in SF as well as in movies, painting, or
theater—was not fulfiled: (he historical sense, the dimension of
264 RUSSIAN SF AND ITS UTOPIAN TRADITION

imaginative experimentation (both utopian and anti-utopian) was


forcibly expunged from it. The sectarian RAPP critics opened in
1929-30 a strong campaign against SF as a harmful genre, and
succeeded in ending almost totally its publication. From an av­
erage of about 25 new books per year in the mid-1920s (47 in
the peak year 1927), the publication plummeted to 4 in 1931,
and one each in 1933 and 1934. Only in 1935 was SF, prop­
erly sterilized, partly rehabilitated as a marginal,—juvenile and
popularizing—genre. Anticipating possible developments be­
came a suicidal pursuit at a time when Stalin was the only one
supposed to “foresee” the future, when cosmonautics were men­
tioned in the same breath with bourgeois cosmopolitism, and a
number of scientists and writers—including the utopographer
Nikolsky and most of the I^eningrad Section for Scientific Fic­
tion, comprising the well-known SF writers and critics Perelman,
Rynin, Uspensky, as well as others—wrere jailed or indeed exe­
cuted. As a result, in the 25 years beginning with Mayakovsky’s
death and Zamyatin’s departure and up to Yefremov's An­
dromeda, of the approximately 300 new stories or novels in Rus­
sian SF, extremely few were significant anticipations—though the
appearance of anticipations by oblique incorporation into the
work of major fiction and screen writers such as Yury Olesha,
Aleksandr Dovzhenko, or Leonid Leonov (one layer of his
three-level novel The Journey Toward the Ocean, 1935, with its
picture of future world wars and interplanetary flight, verges on
a self-contained vision worthy of Wells and The Iron Heel} testifies
to its latent vitality. But the constitution of SF moods and ele­
ments into a significant genre in its own right was cut short by
the imposed Stalinist attitude toward SF known as the “theory of
limits" or, with more sophistry, the "theory of nearer aims," Its
acolytes propounded that (as the Soviet critic Riurikov sums it
up) literary anticipation had to solve only technological problems
of the nearest future, and that it should not attempt to go be­
yond such limits, for only thus will it remain based on socialist
realism. Thus SF was reduced to extolling technology, and its
ethics turned pragmatic. Of all such novels about bigger and
better oil-drills, radars, or solar energy uses, the least boring
were those depicting Arctic exploration and transformation (Lis-
sovoy, Grebnev, Adamov, Kazantsev), since this was luckily both
in the Vernean semi-utopian tradition and the region of many
RUSSIAN SF AND ITS UTOPIAN TRADITION 265

intrepid exploits of the Soviet 1930s. But just as the few remain­
ing anticipation novels were exclusively juvenile, so all of these
mixtures of technological adventure and patriotic—sometimes
even militarv—pride (for example, the closing sections of Pav­
lenko’s In the East or Adamov’s Secret of Two Oceans) fade more or
less into subliterature. Yuriy Dolgushin’s novel The Generator of
Miracles (1940, bp. 1959), approaching some bionic ideas, was
probably the most interesting among them.
Such stagnation within utilitarian horizons and stereotyped
situations and characters (the heroic expedition or project­
leader, the corrupted intellectual doubter, the foreign spy or
saboteur) earned Soviet SF from the mid-1930s to the mid-1950s
the reputation of a second-rate cross-breed, neither really artistic
nor scientific. A first reaction to this was an increase, in the
1950s, of detective and adventure elements (Nemtsov, Kazantsev).
This was perhaps helpful in momentarily dispelling the reigning
monotony, but a meteor alarm instead of a sabotage does not much
raise the cognitive or imaginative level of SF. (The technological-
adventure SF—analogous to rhe SF in the United States in the
1930s—based on a Morality triangle of a starry-eyed beginner
set between a wise elder leader and a careerist, egotistic anta­
gonist has lived on tenaciously in the foothills of Soviet SF.)
7. The second great age of Soviet SF accordingly came about
with a specific regeneration of its utopian imaginativeness in the
decade or so following on 1956. The reasons which made it
possible are obvious. The twentieth congress of the Soviet Com­
munist party in 1956 destroyed the indisputability of Stalinist
myths about society and literature. They were further shaken by
the sensational achievements of Soviet natural sciences, ex­
emplified by the first Sputnik. The new SF wave, rich in tradi­
tion and individual talent, eager to deal with an increasing range
of subjects, from sociological to cosmological and anthropologi­
cal, from astronautic through cybernetic to anticipatory-utopian,
found a wide audience among the young and the intelligentsia.
We have no sure statistics on this reading public, but it was
probably as large if not at times larger than its American coun­
terpart. It was perhaps unsophisticated, but impatient of the old
clichés and thirsting after knowledge and imagination. Its tastes
carried the day in the great “Yefremov debate.”
Indeed, in the history of Russian and Soviet SF, only
266 RUSSIAN SF AND ITS UTOPIAN TRADITION

Chernyshevsky’s—and probably Mayakovsky’s—work had so


taken by storm young people, especially the younger scientists,
and earned the genre such general esteem as did Yefremov’s
novel Andromeda, Subsequent developments in Soviet SF can be
understood only as growing out of its having, against violent
ideological opposition, consummated in 1957-58 the victory of
the new wave—which was really the victory of the pristine Soviet
Russian tradition, in abeyance since the Leninist 1920s. The
writers and critics of the “cold stream" rebuked the novel’s
heroes as being “too far from our times" and thus unintelligible
to the reader, especially the juvenile reader(!). In short, they
were saying that Yefremov's scope was too daring. Such pressure
had for 15 years hindered the publication and development of
Yefremov’s SF (his first SF story, “The Hellenic Secret” from
1942, dealing with “gene memory,” was thought of as mystical
and published only in 1966). However, the opinion of “warm
stream” critics, and of the thousands of readers who wrote to the
author, newspapers, and periodicals, that this was a liberating
turning-point in Soviet SF finally prevailed. The novel has since
been reprinted about 30 times, not counting its probably equally
numerous translations.
Yefremov's work achieved such historical significance because,
in its own way, it creatively revived the classical utopian and so­
cialist vision, the resilience of which had so flabbergasted Wells in
his meeting with Lenin. This vision (Marx’s, Chernyshevsky’s,
Morris’s, or even the mellower Wells’s) looks forward to a uni­
fied, affluent, humanist, classless, and stateless world. Andromeda
is situated in the four hundred and eighth year of the Era of the
Great Ring, when mankind has established informational contact
with inhabitants of distant constellations who pass on such infor­
mation to each other through a “ring” of inhabited systems. The
Earth itself is administered—by analogy' writh the associative centers
of the human brain—by an astronautic council and an econom­
ic council which tallies all plans with existing possibilities; their spe­
cialized research academies correspond to man’s sensory centers.
Within this framework of the body politic, Yefremov is primarily
interested in the development of a disalienated man and new eth­
ical relationships. For all the theatrical loftiness of his characters,
whose emotions are rarely less sublime than full satisfaction
and confidence (only an occasional melodramatic villain feels
RUSSIAN SF AND ITS UTOPIAN TRADITION 367

fear or hate), they can learn through painful mistakes and fail­
ures, as distinct from the desperado and superman cliches of
“socialist realism” or much American SF after Gernsback.
The reexhumation of socialist utopianism brought back into
Soviet fiction whole reaches of the SF tradition: the philosophical
story and romantic etude, classic sociological and modern cos­
mological utopianism. Yet in Yefremov’s novel the strong narra­
tive sweep full of adventurous actions, from a fistfight to an
encounter with electrical predators and a robot-spaceship from
the Andromeda nebula, is imbued with the joy and romance of
cognition. This certainly embraces an understanding of, and in­
tervening into, the outside world of modern cosmology and
evolutionist biology. But Yefremov’s strong anthropocentric bent
places the highest value on creativity, a simultaneous adventure
of deed, thought, and feeling resulting in physical and ethical
(body and mind being indissolubly connected in this materialist
writer) beauty. His utopian anthropology is evident even in the
symbolic title: the Andromeda nebula recalls the chained Greek
beauty rescued from a monster (class egotism and violence, per­
sonified in the novel as a bull and often bearing hallmarks of
Stalinism) by a flying hero aided by superior science. Astronau­
tics thus do not evolve into a new uncritical cult but are claimed
as a humanist discipline, in one of the most signifiant cross-
connections among physical sciences, social sciences, ethics, and
art that Yefremov establishes as the norm for his new people.
Even the novel structure, oscillating between cosmic and terres­
trial chapters, emphasizes this connection. Furthermore, this fu­
ture is not an arrested, pseudo-perfect end of history—the bane
of optimistic utopianism from Plato and More to Bellamy. Freed
from economic and power worries, people must still redeem
time, which is unequal on Earth and in space, through a
humanist dialectics of personal creativity and societal teamwork
mediated—in a clear harking back to the ideals of the 1920s—by
the artistic and scientific beauty of functionality (Dar listening to
the cosmic symphony, or the Tibetan experiment). Creativity is
always countered by entropy, and self-realization paid for in ef­
fort and even suffering. In fact, several very interesting ap­
proaches to a Marxist “optimistic tragedy” can be found in the
book (for example, in the Mven Mass "happy Fall” motif). Fi­
nally, the accent on beauty and responsible freedom places
268 RUSSIAN SF AND ITS UTOPIAN TRADITION

Chernyshevskian female heroines in the center of the novel, in­


teracting with the heroes and contributing to die emotional
motivation of new utopian ethics, in complete contrast to con-
tempo rar)- American SF (with which Yefremov was obviously in
a well-informed polemical dialogue).
How difficult it is for an SF writer to portray basically differ­
ent, even if philosophically already sketched human relations,
can be seen from some places where the novel’s dialogue,
motivation, and tone flag, so that it falls back on pathos and
preaching, which slow down its rhythm. Yefremov’s characters
tend to be statuesque and monolithic in a kind of neoclassic way,
and his incidents often exploit the quantitatively grandiose:
Mven blows up a satellite and half a mountain, Veda loses the
greatest anthropological find ever, and to think of the manly Erg
blushing or the pure Nisa stepping into, say, offensive jellyfish
offal on the iron-star planet is practically blasphemous. Most of
this can be explained by Andromeda's having had to achieve sev­
eral aims at once: it was the first work to burst open the flood­
gates closed for 25 years, and it overflowed into clogged chan­
nels. One feels in it the presence of a reader unused to fast
orientation in new perspectives and, as Yefremov himself wrote,
"still attracted to the externals, decorations, and theatrical ef­
fects of the genre.” But it cannot be denied that some aspects
of Yefremov's ethics and aesthetics, such as the erotic and gen­
erally intimate interpersonal relations—though understandable
enough in the context of the social taboos obvious to a Soviet
Russian scientist of an older generation—are curiously old-
fashioned for a sweeping SF glance. His limitations are more
clearly manifested in the later long story Cor Serpentis, where
Terrans meeting a fluorine-based mankind put an end to its
loneliness by promising to transmute fluorine into oxygen.
This story—an explicit counterblast to the US story “First Con­
tact” (by Murray Leinster), with its bellicose and acquisitive
presuppositions—might be a legitimate pacifist-socialist allegory
for changing American capitalist meritocrats into Russian
socialist ones, yet such an ethnocentric view precludes a full de­
velopment of imaginative SF vistas, the point of which is unity in
variety. But again any further discussion of such vistas in Soviet
SF was made possible by Yefremov’s pioneering effort. Androme­
da’s polyphonic scope, with its large number of protagonists, is
RUSSIAN SF AND ITS UTOPIAN TRADITION 269

Tolstoian rather than Flaubcrtian. Not being limited to the con­


sciousness of the central hero, it is perhaps, together with Lem’s
Magellan Nebula in Poland, the first utopia in world literature
which successfully shows new characters creating and being
created by a new society, that is, the personal working out of a
collective utopia (analogous to what Scott did for the historical
novel). Yefremov’s basic device of unfolding the narration as if
the anticipated future were already a normative present unites
the classic “looking backward" of utopian anticipations with the
modern Einsteinian conception of different coordinate systems
with autonomous norms; twentieth-century science and the age-
old dreams of a just and happy society' meet in his novel. This
meeting made it the nodal point of the Russian and socialist
tradition examined in this chapter. And this is also why the novel
was able to usher in the new era of Soviet SF—an era which
seems to have closed with the 1960s.
Subsequent developments and differentiations within Soviet
SF can be fully appreciated only from this perspective. In the
dozen years of its second flowering, mastering new themes and
approaches, it has produced a considerable body of significant
works. It has assembled about 50 habitual (though not full-time)
writers, an insatiable reading public, an expectation of literary
clarity and careful craftsmanship, and it has avoided being con­
fused with fantasy, gangster novels, horror thrillers, and cloak-
and-dagger adventures, though not with the more deeply rooted
national tradition of the folktale. Perhaps most important for
understanding it is the fact of its leading writers having opted
for a hope that grows out of the central position of functionally
clear human figures. They have, in other words, developed the
basic philosophical and literary tradition of socialism—utopian­
ism, equated here with the open horizons of anticipated human­
izing potentialities.
The “Yefremov era’’ in Soviet SF is now fast receding into
history. But the failure of the original utopian confidence from
the mid-1960s on, to which Yefremov’s own later works also
testify both in overt ideological profiles and in their reduced
significance, is in the practice of significant subsequent Soviet
SF—in the cases of the Strugatsky brothers, of Varshavsky or
Shefner—itself measured and judged by the values of that uto­
pian and dealienating horizon that is its permanent hallmark.
12
Karel Capek,
or the Aliens Amongst Us

l. There is both irony and poetic justice in the fact that Karel
Capek is today, at least outside the Slavic countries, remembered
mainly as the creator of the word robot. A first irony is that this
neologism—from the archaic Czech robota, meaning “drudgery”
with strong feudal connotations of the serf’s compulsory work
on the master’s property—was coined by Karel’s brother Josef, a
prominent painter and writer with whom he collaborated on a
number of early works, including the plays From, the Life of the
Insects and Adam the Creator, and who himself wrote a symbolistic
SF play (Land of Many Names, 1923). A second irony is that Karel
Capek’s eight plays are, despite the world popularity of some
among them (including Rossum’s Universal Robots, or R.U.R.), the
weakest part of his opus, which comprises about fifty books of
stories, essays, travelogues, novels, and articles. The poetic jus­
tice, however, stems from the fact that a quite central preoccupa­
tion of his was with the potentials and actualizations of inhuman­
ity in twentieth-century people, and that this preoccupation was
throughout his whole opus translated into the image of the Nat­
ural Man versus the Unnatural Pseudo-Man. This manlike, rea­
sonable but unfeeling being is in Capek’s work represented by a
number of approximations, one of the first among which were
the robots of R.U.R.
Josef Capek’s happy coinage, with its blend of psychophysio­
logical and political meanings, hits thus the bull's-eye of what
his brother was trying to get at. For all of Karel’s menacing,
inhuman beings are associated with and rendered possible by
capitalist industrial technology with its accompanying social
extremes of the upper-class tycoon and the working-class mul­
titude. The criterion of naturalness is for Capek drawn from
the middle class, and his heroes range from smalt employees and
270
KAREL CAPEK, OR THE ALIENS AMONGST L'S S7I

craftsmen to doctors and engineers. Born in 1890, he himself


grew up as the son of a physician in a rural district, in a home
which he characterized “like thousands of other homes in the
bourgeois world of that time,” and in a semi-colonial state which
made of the Czechs "on the whole, a nation of petty bourgeois.”
Capek’s grandparents were peasants, and some of his most stub­
born values and prejudices can be traced back to the traditional
peasant confidence in the immediately available, secure, every­
day things and relationships of the little people, as opposed to
the hubris of the hustling and bustling modern industry and the
swift changes it brings about. In the plays, this attitude is openly
expressed by his small people who act as ideological arbiters, such
as Nana in R.U.R. or Kristina in The Makropoulos Secret-, in the
novels, it is implied by strategic collocations of key actions, such
as the return to normality at the endings of Krakatit and The
Factory for the Absolute; but as a rule, it pervades all of Capek’s
works. On the other hand, he was “enchanted and terrified” by
the world of factories and technology—the workers’ districts of
Prague and the miners of his childhood days—as well as the
“pride, power, wealth” of the capitalists such as the emblematic
Mr. Bondy from both the symbolically named The Factory for the
Absolute and War With the Newts. Capek’s SF was written to deal
with “great social interests and collective spiritual problems” aris­
ing out of “the leading ideas of science, guesses about the future,
feats of technology”—that is, to deal with the destructive men­
aces which the irruption of modern mass production brings to
the little man.1
Therefore, in Capek’s first SF phase, which extends through
the decade following World War I, there is a basic tension be­
tween the “natural,” average little people, representative of the
audience he was writing for, and the catastrophic forces of in­
human violence amid which thev have to live, suffer, and die.
For Capek, inhuman, large-scale technology’ and industry lead in
politics logically to international and civil warfare. A fatal am­
biguity between the menace to Man as such and the menace to
the middle-class man vitiates this whole phase of his work, its

1. Ail these quotations are from Karel Capek as cited in Alexander Matushka (Bib­
liography VII}—the first from p. 78, the second from p, 94, the third and fourth from p.
230, and the last two from p. 93.
272 KAREL ¿APEX, OR THE ALIENS AMONGST US

qualities lying in what he managed to do in spite of and on the


margins of such an ideological muddle. For the menace to Man’s
existence arises from aliens created and abetted by large industry
and its capitalist masters and engineering managers; but the
menace to middle-class life arises from the workers, who had in
Russia just seized power in the Bolshevik revolution. The
"robots” of R.U.R. (1920), synthetic androids outwardly indistin­
guishable from men, are mass-produced for the express purpose
of being “workers with the minimum amount of requirements"
(act 1), improved producers with a machinelike precision but
without the nonexploitable qualities of emotion, sex, art, and the
like. In their first story7 “System” (1908) the Capek brothers had
in fact dealt with a workers’ revolt in such circumstances. The
machine-men of R.U.R, each of whom replaces two and a half
workers, are a technolgical variant of the workers from “System”
or of the Morlocks, Wells’s workers of the future who also be­
came the new lords of the creation. But in Capek robots are not
only stand-ins for workers but also—in an ideological mystifica­
tion which brought him instant fame because it corresponded to
deep needs for self-delusion in his audience—inhuman aliens
“without history” (R.U.R., act 1). Thus their creation does not
lead to Dornin’s engineering utopia of total abundance and lei­
sure which would breed Nietzschean supermen, but to a genoci­
dal revolt of the submen against humans. The only surviving
man is the Tolstoian pacifist Alquist, an architect who had dealt
in construction rather than in the blasphemous progress of the
robots' producers. But at the end of the play, the robots again
grow more like a new human order than like inhuman aliens,
more like workers than machines; reacquiring pain, feelings, and
love, they usher in a new cycle of creation or civilization. For all
the interest inherent in the basic concept and the theater tricks
of Capek’s, this fundamental oscillation between mutually in­
compatible ways of envisaging the robots—which also means an
oscillation between old-fashioned psychological and modern “col­
lective” drama—has by now dated this play.
In From the Life of the Insects (1921, also translated as The Insect
Play or Comedy; The World We Live In; and And So Ad Infinitum)
human moral concepts and social behaviors are personified as
insects: the flighty erotics of the butterflies are a takeoff on the
upper-class “golden youth” of postwar Middle Europe, the
KAREL dAPEK, OR THE ALIENS AMONGST US 273

dung-beetle and cricket families on the acquisitive and sentimen­


tal petty bourgeoisie, and the ants on robotlike militarism and
chauvinist nationalism of the masses and their leaders. The ani­
mal fable is used for a "surprising and cruel analogy” and bitter
satire of the "insect selfishness” of human individuals and collec­
tives,2 for a Morality play which is the Capeks’ best stage product.
Especially memorable is the ants’ act, with its portrayal of the
war-machine collective, which progresses from Taylorism trans­
ferred wholesale from industrial to military behavior to a death­
lust racism. On the contrary, both the biblical Expressionism of
Adam the Creator (1927) and the psychological detection of The
Makropoulos Secret (1922, about the elixir of longevity and very
similar to Godwin’s St, Leon) are very shakily allied to a defeatist
humanism, and make for ¿apek’s feeblest SF plays.
Capek's real strength as an SF writer lies in his novels. The
Factory for the Absolute (1922, also translated as The Manufacture of
the Absolute and The Absolute al Large) is a revue-novel (Capek
called it a “column-novel”) based on another destructive inven­
tion which was supposed to usher in a utopia. It is the Absolute
or God, mass-produced as a by-product of atomic fission in “kar-
burators” that rapidly take over as the cheap suppliers of the
world’s energy. The concept is a fortunate one, combining wildly
hilarious with weighty philosophical possibilities. And the novel
is chock-full of irony and satire on the uses and abuses of the
Absolute by church and state, corporations and individuals,
academics and journalists. It bears out the cynical Catholic
bishop who declares that mankind cannot cope with “a real and
active God”: both the economic system, when faced with bound­
lessly cheap overproduction but no adequate distribution, and
the personal relationships, when faced with the Absolute work­
ing within variously oriented individual consciences, simply col­
lapse. Sectarian and national fanaticism multiplies when the
Absolute gets control of the two greatest powers in the world,
“industry and the masses.” This leads to a final “Greatest War,”
recapitulating the experiences from the Thirty Years’ War
through Napoleon to World War I, which peters out only after
all “ato motors" have been destroyed and most people killed.

2. Èapek's formulation, cited in Matushka, p. 298.


274 KAREL CaPEK, OR THE ALIENS AMONGST US

However, it is not clear why the “mystical Communism" of the


Absolute must bend itself to the capitalist forms of economic
organization and to individualist competitive psychology, which
are ¡(logically assumed to be stronger than even an absolute
power. Further, the disparate workings of the Absolute in things
(overproduction of industrial goods only, not of farm produce or
communications) and in people (destructiveness to match the
overproduction) do not follow anv consistent pattern either.
Thus, instead of being a true Absolute, the power which is un­
leashed by the atomotors is for Capek simply a chaotic magnifica­
tion of the antagonisms inherent in acquisitive economics and
psychology. This makes for brilliant if spotty social satire, but
hardly for consistent SF conceptualization. That is why by the
middle of the novel Capek has to substitute for logical expla­
nations a dazzling display of chronicle overviews alternating with
hyperbolic and grotesque snapshots. The only people who retain
a tightfisted "normality” are the farmers, and the final chapter
culminates in a typically Czech beer-and-sausage feast of the little
people, which is consecrated as the highest achievable form of
tolerance and everyday happiness. As already seen in R.U.R.,
any high minded idea leads necessarily to huge disasters, and one
should stick to the pragmatic immediacy of people believing in
other people. Not even the Absolute can improve men, religions,
or the bourgeois institutions: the paradoxes of this creed explain
but do not cancel out the inconsistencies of the novel.
The parallel of men and matter is carried forward in Capck’s
most unjustly neglected work of SF, the novel Krakatit (1924, also
transalted as An Atomic Phantasy). Different from his earlier
works, it has a romantic hero—Prokop, the naive inventor-genius
who pierces the secret of explosiveness (atomic fission). But such
“destructive chemistry” is also taking place in humans, and this
bulky novel is as full of Dostoevskian fevered dreams and night­
mares, dissociations of memory, and explosive human encoun­
ters as of physical explosions, fights, and escapes. The plot traces
Prokop's development from a solitary “value-free" scientist car­
ing only about matter to a wiser human being who has painfully
learned about the primacy of human relationships: “His hair
stood on end in horror at the nature of the forces among which
we live.” Capek has in Krakatit made a conscious and significant
attempt to integrate popular paraliterature—in particular the de­
KAREL CAPEK, OR THE ALIENS AMONGST US 275

receive story and the epic adventure from Homer through die
folktale to pulp thrillers—into sophisticated, poetic SF dealing
with central questions of modern life. He wanted to fuse die
“love and heroism” of a sensational newspaper story or a “novel
for serving-maids”3 with a psychological narrative, and the over­
tones of countercreation inherent in a mad scientist story with a
sympathetic, suffering and relatively complex hero whose educa­
tion advances through a series of erotic-cum-political tempta­
tions. For once, Capek’s hero rejects not only the militarist temp­
tation of established power and the nihilist temptation of new,
personal power (since the explosive force of power necessarily
leads to violence), but also the small-town idyll or island of re­
pose. He is left with a resolve to achieve useful warmth instead of
destructive explosions: “He whose thoughts are full of the high­
est turns his eyes away from the people. Instead you will serve
them. . . Prokop is told by the mysterious Grandfather he
meets at the end. From die fog of suffering and yearning, the
hero has finally emerged into the clarity of moderation; his
earth-shaking invention will be refunctioned. Though the novel
has not quite succeeded in fusing realism and allegory, because it
has not quite solved how to fuse ethical moderation with the
certainties of the folktale, it is a largely successful first try at
transcending the sterile opposition between scientific progress
and human happiness which has haunted science fiction from
Swift and Voltaire (Capek’s teachers) to the present day. For the
first and last time in ¿apek’s SF, a believable hero fights success­
fully back at the destructive forces within himself and society.
2. The second phase of Capek’s SF was the result of the rise of
Nazism, which threatened directly both his native land and his
basic values. It comprises a few minor stories on the margins of
SF, satire, and fantasy, as well as the novel PFar With the Newts
(1936), and the The White Sickness (1937, also translated as The
White Plague or Scourge, and The Power and the Glory). By this time
Capek had shed many of his illusions, in particular his prejudice
in favor of the little man’s instinctual rightness and of everybody
having his own truth. In R.V.R., for example, all figures—from
the Nietzschean utopist Dornin through his various director coi-

3. Eva Strohsova (Bibliography VII), p. 131,


2T6 KAREL CAPER, OR THE ALIENS AMONGST US

leagues to the robots themselves—were right in their own way;


the same is true in The Factory for the Absolute. But already in
Krakatit this relativism was waning, and after the rise of Hitler,
Capek reconsidered the role of the intellect which he satirized
earlier (for example, in the name of the robots’ inventor, which
translates as Mr, Reasson). Now he wrote sharply against an in*
teliect that is giving up its rights "in favor of irrationalism and
da ¡monism, be it the cult of will, of the land, of the subconscious,
of the mass instincts, or of the violence of the powerful—that is a
decadent intellect because it tends toward its own downfall.”4 A
limit was found beyond which the pseudo-human became clearly
evil; that limit is reached when the new crearures in War With the
Newts grow into an analogy to the Nazi aggressors. That is why in
his final SF novel Capek’s satire is most clearly focused, the de­
velopment of the novum most consistent, and there is no con­
ciliatory happy ending. “This is not a speculation about some­
thing in the future,” Capek said about it, “but a mirroring of
what exists and amid what we live."5 Let us briefly analyze the
development of this remarkable novel.
War With the Newts (properly War With the Salamanders) is di­
vided into three “books,” the first of which, “Andrias
Scheuchzeri,” shows, beneath its ironically neutral, scientific
name, the Salamanders entering the life of mankind on the
wrong foot, as it were—that is, under a cloud of delusions and
misperceptions. The satire is aimed at the fictions of mankind
which prevent it from seeing the rise of a catastrophic menace.
The first five chapters are an increasingly ironic counterfeit of
the exotic adventure tale associated with the names of Sindbad,
Kipling, Conrad, London, and others. It shows how the Salaman­
ders are led to an independence (confirmed by the ironic dates
of July 3 and 13 instead of the proper US and French liberation
dates of Julv 4 and 14) from their biological and geographical
barriers. This is possible because first Captain van Toch and then
the business tycoon Bondy see in them a juvenile romantic ad­
venture, in which the misanthropic captain with the baby-blue
eyes plays the role of a Salamander culture-hero, and the busi­
nessman that of a merchant-adventurer’s financier. In the dou-

4. ¿apck's article in the newspaper Ptitemnoit, No. 29 (1984).


5, See Capek's book Pozndmiy o (Prague, 1959), p, 110.
KAREL CAPEK, OR THE ALIENS AMONGST US 277

ble chapter 6-7 the Salamanders are recognized by the world


through its two most virulent illusion industries—capitalist
movies and sensational tabloid newspapers oriented toward erot­
ic and exotic clichés. The Hollywood way of apprehending the
world in terms of platinum blondes, exotic monsters, and native
treasures is mercilessly satirized because it screens from mankind
the reality of human, social, and biological events both by sup­
pression of episodes that do not fit the clichés and by an inability
to think outside such clichés (Abe and Li). Thus the rise of the
Salamanders is interwoven with a satire of human obfuscations,
in particular of the means of communication. This satire is most
intimately woven into the very composition and texture of
Óapek’s novel, since its various segments debunk the various
genres and consentions of informational transmission. Alter the
exotic, basically juvenile adventure story, the tabloid sensations,
and the South Sea movie (to which will later be added the
pseudo-historical movie of the Henry VIII type), in chapters 8-11
the pseudo-scientific way of data collecting and dissemination
from observations in zoos, interviews, and newspaper populari­
zations is taken to task. It is made clear that it is as charlatanic as
the interpolated country-fair show of a false “Captain van Toch
and his trained lizards.” (Il is worth noticing that the exotic
names of both the captain and the movie star Lily Valley mask
the prosaic Czech names of Vantoch and Nowak.) Professional
scientists and academics are shown to be as myopic, timid, and
ideologically limited—especially in their nationalistic subservi­
ence to the powers that be—as the public at large; in the Hiller
decade Capek noted with a sinking heart but firm glance that the
treason of the intellectuals to their humanist calling was proceed­
ing apace. But the real political villains were the capitalists who
financed the menace to humanity (Nazis as well as Newts); and
the fulcrum of the novel’s first part is chapter 12, the formation
of the Salamander Syndicate, which moves from the fairy tale
and juvenile-epic mode of dealing with the Salamanders to an
industrial utopianism, a “hymn of construction” satirized in
Óapek's minutes of a corporation meeting. At this point, the
Salamanders—who began as exotic monstrosities or pets, and in
chapters 9 and 10 demonstrated that they were psychologically
and physically no more monstrous than the average human (the
newspaper-reading Englishman or the Czech fat lady)—begin to
278 KAREL ¿APEK. OR THE ALIENS AMONGST US

be treated as an ‘‘extremely cheap labor force,” “toilers of the


sea" halfway between expendable under men and a new trained
subproletariat. Capek thus gives us a shortened genesis of
capitalism. Beginning as a competitive colonial merchant­
buccaneering, it is in this chapter qualitatively transformed to a
global, monopolistic, and exploitative corporation aiming at a
new Atlantis, at remodeling the Earth; eventually, it will end as
.Atlantis did, unable to control the globally destructive forces it
has unleashed for exploitation but never really understood. "Do
I know what a Newt is like? What good is that to me?” says the
wistful tycoon Bondy, echoing Brecht's famous “Song of the
Commodity”:
What is in fact rice?
Do 1 know what rice is?
Do I know who knows that?!
I don’t know xvhat rice is.
I know only its price.
[The Measures Taken]
From this point on, the global scale and technologicoeconomical
acceleration of Salam a ndr ism will necessitate a larger and brisker
overview. There is just time, in the first part, to insert an appen­
dix on the Salamanders’ sexual life, as a clue that the Salamander
Syndicate’s utopia is the greatest illusion of them all. The
Salamanders’ ominous, Nazi-like "Collective Male” horde, capa­
ble of politics and technics but not of real sociability, testifies that
the Syndicate has misunderstood the basic nature of the demon
it has let out of the bottle.
The second book of War With the Newts is titled “Along the
Steps of Civilization”—which, bitterly, means along the steps to
war. The Salamanders “progress” from animals first to slaves,
then to an increasingly powerful alternative society with class
differentiation, industry, economic and military power. At the
same time they remain barbarians with a perverted intellect and
with languages “rationalized into [their] simplest and most
rudimentary form.” Equal in power to humans, they are in fact a
robotic or absolutistic caricature of mankind, a mankind de­
prived of human qualities. On the other hand, in a brilliant use
of satire, it is mankind that not only abets them but also teaches
them the inhuman combination of slavery and stock-market,
KAREL ¿APEK, OR THE ALIENS AMONGST US 279

warlike aggression and ideological propaganda—in brief, the


corruption of reason for violent purposes.
In the third book, the Salamanders begin their warfare upon a
very corrupt mankind. Capek’s satire is here most bitter and
topical, though no less precise. Briefly but impressively, the
nodal points of defeatism in face of rising Salamandrism (read
fascism) are passed in review, from the Spenglerian philosopher
Meynert, through “Salamandrist" fashions in art and entertain*
ment, to the secret deals of the bourgeois states with it and the
public hypocrisy of international conferences. The military suc­
cesses of the Salamanders run parallel to the failures of reason
and humanism among men. In a startling maturation for Capek,
even his beloved small Czech people are found guilty of neu­
tralist complacency in the person of Povondra, the only character
to be found in all parts of the novel.
In a way, the satire of literary, journalistic, and essayistic forms
in War With the Newts is also a critique of past and (unintention­
ally) of subsequent SF. The history of the Salamanders starts
with an echo of H. G. Wells' Island of Dr. Moreau, as well as of
Arthur Conan Doyle’s Lost World and of the animal fable from
Aesop to Pierre MacOr Ian’s story “The Conquering Beast.” Il
proceeds with a global overview which latches onto Wells’s later
SF, from The World Set Free to The Shape of Things To Come, as well
as onto Anatole France’s ironic Penguin Island. It ends with
havoc-wreaking aliens out of The War of the Worlds or “In the
Abyss.” But instead of Wells’s alternatively pessimistic and op­
timistic outcomes, Capek’s end is, much more realistically, an
open question. The final chapter, where the author's two inner
voices debate the possible outcome, is to my mind much more
mature than the facile extremes of, on the one hand, the a priori
optimism of Bellamy and the feebler Wells (later picked up by
writers such as Robert Heinlein, Isaac Asimov, and others) or, on
tlte other, the despair of Aldous Huxley’s Brave New World or the
Spenglerian pessimism criticized in Wolf Meynert (and later
picked up by James Blish, A. E. Van Vogt, and others). Capek is
much superior to ordinary SF mystifications: “No cosmic catas­
trophe, nothing but state, official, economic, and other causes.
. . .” He is therefore much more activistic: the menace could be
(or could have been) stopped—all depends on men organizing
to fight it. But instead:
280 KAREL ¿APEK, OR THE ALIENS AMONGST US

“Do you know who it is that is feverishly working day and


night in the laboratories to invent still more efficient
machines and materials for annihilating the world? Do you
know who lends [the Newts] money, do you know who
finances this End of the World, all this New Flood?”
“Yes, I do. .AJI the factories. All the banks. All the different
states.”
Such writing makes of Capek the pioneer of all anti-fascist and
anti-militarist SF in the world, from the later Wells and Sinclair
Lewis to the postwar American and Slavic writers. And he is still
one of the best among them.
The enduring power of War With the Newts (and to a smaller
degree of The Factory for the Absolute) is due to Capek's mastery of
stylistic strategies for building up his vision. In the first part of
this mosaic the events are introduced and refracted in comically
conflicting, individual though typical, points of view. As the
focus of the action widens, impersonal points of view from arti­
cles and documents become more and more important, and the
third part is mainly a kind of histórico philosophical essay emo­
tionally strengthened by what one could call a tragic lampoon.
The traditional individualist psycholog}' is used in a masterly—
though mainly comical and abbreviated—fashion where neces­
sary, but it has been found insufficient for events of a global
character. As in all modern literature, from James Joyce and
John Dos Passos on, traditionally “nonfictional” forms arc re­
functioned for fictional use, resulting in a far richer and wfder
view than any combination of merely individual points of view
could have effected. In this way, Capek is—together with
Evgeniy Zamyatin—the most significant world SF writer between
the World Wars. Though three or four English-language writers
of that epoch have had a much greater influence on the de­
velopment of postwar SF, more complex preoccupations have
again resulted in a return to Capek’s cognitive polyphony in
writers such as Stanislaw Lem (clearly influenced by him), John
Brunner, or Ursula Le Guin.
This is not to say that even War With the Newts is perfect. Capek
did not quite manage to overcome his permanent ambiguity of
the inhumans being on the one hand a wronged inferior race or
class (at the beginning) and on the other a menacing embodi­
KAREL CAPEK, OR THE ALIENS AMONGST US 281

ment of the worst in modern humanity—both Nazis and


robotized masses. It is indicative that in a novel published in
1936 and supposed to take place in the immediate future, Russia
is represented as a tsarist state. Capek wrote for twenty years a
column in an anti-communist paper and was a friend of Presi­
dent Masaryk, and yet he had a considerable ambiguity toward
communism. On the one hand it represented the wronged, on
the other an abstract idea allied to modern masses, a large state,
and large industry. Therefore, he defined himself as a kind of
ethical socialist: “I believe in the socialization of means of pro­
duction, in the limitation of private property, in an organization
of production and consumption, in the end of capitalism, in the
right of each to life, work, sufficiency and freedom of mind, I
believe in peace, solidarity and equality of the peoples, I believe
in humaneness and democracy, and in man, Amen.”9 That is
why he felt once compelled to write an article "Why I am Not a
Communist,” and why he felt unable to deal not only with
socialism but also with any positive radical novelty in his major
works. Thus he dealt in catastrophes; and at the time of the rise
of Nazism, when Czechs looked to the USSR for a counterpoise,
he concentrated his fire almost entirely (except for a silly proc­
lamation by the Third International to the Salamanders) on
bourgeois society. Conversely, this strange love-hate relationship
led the communist bureaucracy in Czechoslovakia to neglect him
in the early 1950s, although he has since been rehabilitated.
Capek’s evolution is by no means simple. The work that is to
my mind as significant as War With the Newts, Krakatit, was written
early in his career. And his final SF play, The White Sickness, fits
well into an ideational development toward an active anti­
fascism, a critique of militarism, chauvinism, and subservient
medical science as well as of the “neutral” petty bourgeoisie, yet
it is a second-rate work reposing upon misleading allegory.
Perhaps it w’ould not be fair to analyze it too thoroughly: as he
said after the Munich agreement, after his beloved England had
left the Czechs to Hitler, “the world I believed in has fallen
apart.”67 He died at the end of 1938 having lost the will to live—if

6. Capek, O victth obetnych Cili Zoon politikim (Prague, 1932), pp. 136-37,
7. Capek s formulation, cited in Halina Janaszek-Ivaniikova (Bibliography VII), p.
244.
282 KAREL ¿APEK, OR THE ALIENS AMONGST US

you will, of a broken heart. His writings were forbidden by the


Nazis, and his brother Josef died returning from Bergen-Belsen.
But in a few works, such as War With the Newts he has left us a
precious legacy. It may not be as significant as that of his coun­
trymen Franz Kafka or Jaroslav Hasek; but these are the high­
est standards applicable. And for SF at least, he—rather than
Edgar Rice Burroughs or Hugo Gernsback—is the missing link
between H. G. Wells and a literature which will be both enter­
taining (which means popular) and cognitively (which means also
formally) avantgardist. He took the adventure novel and the
melodramatic thriller, the legacy of French and British SF as well
as of German fantasy from the Romantics to the Expressionists
(Hoffmann, Meyrink, Kaiser), and infused all this with the pros­
pects of modern poetry, painting, and movies, with an eager and
constant interest in societal relationships, in natural and physical
sciences, and above all in the richly humorous and idiomatic
language of the street and the little people. In that way, he is the
most "American" of the often elitist European SF writers; and
yet he is also not only intensely Czech, but a “European local
patriot” for whom Europe meant culture and humanism. When
they were betrayed, the Salamanders had arrived. In order to
defend them, in order to understand what was happening, he
wrote literature:
Discovery—what a great and insatiable passion. I am con­
vinced that I write for that reason, to discover. ... I am
interested in everything there is. That is why I cannot be­
come anything more than just a writer.8
That is also why, in particular, he wrote SF. As he said about his
SF plays:
... it would be easier to say [much about human lives] more
pointedly, with fewer words, preparations, demonstrations,
and proofs, in short more effectively than in a play in which
humans only would be portrayed.
. . . [the dramatist] must include in his play those great
forces that determine human life, forces that are revealed to
us step by step by biology, sociology, economics. . . . the

8. Cited in Matushka, pp. 66-67.


KAREL ¿APEK, OR THE ALIENS AMONGST US 283

poet of our day must master facts as yet unknown, complex


and exciting problems, glowing and terrible dreams. He
can make use of collective psychology' as well as individual,
and he can study the heroism of ideas as well as inner simplicity ,9

The part I italicize seems to me an excellent definition of sig­


nificant science fiction—and of Capek.

9. Both quotations cited in Macushka* the first on pp. 73—74, the second on p. 372.
Bibliography

Students of various matters discussed in this book will recognize my


very large debts to other scholars and critics, far too numerous to
acknowledge properly in the notes. 1 hope the following checklist may
suffice to indicate them, though it had to be confined to a strongly
curtailed selection from a much longer list which I hope at some future
time to publish separately as an annotated bibliographic guide, and to
which the Zantovskâ-Suvin item from Section I is an introduction. In
the initial work on such a bibliography, without which this list could
not have been compiled, I gratefully acknowledge a Québec Ministry
of Education FCAC grant given to a team headed by Professor Marc
Angenot and myself at McGill University.
However, the following lists are also intended to be fairly complete
surveys of the indispensable secondary literature from the major Euro­
pean languages (excluding dissertations) a graduate student or scholar
would need for further independent research in all but the minutest
details of the fields indicated by the section titles (and even such details,
could, 1 believe, be pursued with help of the bibliographies and notes
in the items listed). Striving for finiteness and practicality, I have as a
rule also excluded handbooks, biographies, general bibliographies, and
general surveys of periods or authors, however useful they often might
be. In order not to let this selection grow into a Frankensteinian crea­
tion, I have after much reluctance decided to exclude the introductory
category of “methodological landmarks,” for which I refer the curious
reader to “Bibliographie <f>” in my Pour une poétique de la sàence-fictian
(Montréal: Presses de l’Université du Québec, 1977), as well as to Marc
Ange not's annotated bibliography of the sociology of literature in
Sctenee-Fiction Studies, No. 13 (1977). As a rule, I have listed books and
articles by their short titles and the most accessible edition which 1 took
to be an English-language one wherever it exists; for texts in languages
other than English and French i have also used the French version
whenever possible. Articles from periodicals have been listed only if I
found them of fundamental significance or if my argument was—by
agreement or disagreement—more or less directly indebted to them. In
my experience, to find an article in a periodical it has proved sufficient
to know the year of publication plus the volume or issue number, and 1
have here too opted for practicality rather than bibliographic or­
thodoxy. The bibliography has been taken to 1976, but it shares the
perennial open-endedness of all encompassing attempts: a few items

285
286 BIBLIOGRAPHY

from 1976 and indeed much earlier vears have proved stubbornly
unfindable, and a few items from 1977 and 1978 have also been in­
cluded. (B) indicates that the item either is a bibliography or contains a
larger one.
The following abbreviations have been used in tides of periodicals:
Ass. Association Proc. Proceedings
Bull. Bulletin Pub(b)l. Pub(b)lications
Ges. Gesel ischaft (liehe) R. Review or Revue
Gesch. Geschichte Trans. Transactions
J- Journal Wiss. Wissenschaft(liche)
Lit(t). Literature or Literary Z. Zeitschrift
(various languages)

I. Theory and General History of SF After Wells


For Russian SF see Bibliography VI, and for Capek Bibliography
VII. See also Biesterfeld, Clarke (1966), Ruyer, and Schwonke in Bib­
liography II, III A, and IV B.
Aldiss, Brian W. Billion Year Spree. London, 1975.
Amis, Kingsley. New Maps of Hell. New York, 1975.
Atheling, William, Jr. (pseud, of James Blish). The Issue at Hand.
Chicago, 1964.
-------- . More Issues at Hand. Chicago, 1970.
Bailey, J. O. Pilgrims Through Space and Time. Westport, CT, 1972 (B).
Barmeyer, Eike, ed. Science Fiction. Munich, 1972 (B by Franz Rot-
tens teiner).
Barron, Neil, ed. Anatomy of Wonder. New York, 1976 (B).
Baudin. Henri. La Science-fiction. Paris, 1971.
Baxter, John. Science Fiction in the Cinema. New York, 1970.
Bretnor, Reginald, ed. Modem Science Fiction. New York, 1953.
----- —. Science Fiction, Today and Tomorrow. Baltimore, 1975.
Bridenne, Jean-Jacques. La Littérature française d’imagination scientifique.
Paris, 1950.
Caillois, Roger. Images, images. . . . Paris, 1966.
Chauvin, Cy, ed.J Multitude of Visions. Baltimore, 1976.
Cinéma d’aujourd’hui, N.S. No. 7 (1976).
Clareson, Thomas. Science Fiction Criticism. Kent, OH, 1972 (B).
Clareson, Thomas D„ ed. Many Futures, Many Worlds. Kent, OH, 1977.
-------- , ed. SF: The Other Side of Realism. Bowling Green, OH, s.a.
[1971],
-------- , ed. Voices for the Future. Bowling Green, OH, 1976.
BIBLIOGRAPHY 387

Clarke, Ian. The Tale of the Future. London, 1972 (B).


Davenport, Basil, ed. The Science Fiction Novel. Chicago, 1969.
Eco. Umberto. Apocalittici e integrati. Milan, 1973.
Europe, 55 (Aug.-Sept. 1977) (B).
Ferrini, Franco. Che cosa e la fantascienza. Milan, 1970.
Gerber, Richard. Utopian Fantasy. New York, 1973 (B).
Gunn, James E. Alternate Worlds. Englewood Cliffs, NJ, 1975.
Gurevich, Georgii. Karta Strany Fantazii. Moscow, 1967.
Handke, Ryszard. Polska prozafantastyczno-naukowa. Wroclaw, 1969.
Harrison, Harry, and Brian Aldiss, eds. SF Horizons. New York, 1975.
Hienger, Jörg. Literarische Zukunftsphantastik. Göttingen, 1972.
Hillegas, Mark R. The Future as Nightmare. Carbondale, IL, 1974.
Johnson, William, ed. Focus on the Science Fiction Film. Englewood Cliffs,
NJ, 1972.
Kagarlitskii, lu. Chto takoe fantastika?. Moscow, 1974 (Kagarlizki, Was ist
Phantastik?. [East] Berlin, 1976).
Ketterer, David. New Worlds for Old. New York, 1974.
Klein, Klaus-Peter. Zukunft zwischen Trauma und Mythos: Science Fiction.
Stuttgart, 1976.
Knight, Damon. In Search of Wonder. Chicago, 1967.
Konstantinova, Elka. Fantastika i beleirislika. Sofia, 1973.
Krysmanski, Hans-Jürgen. Die utopische Methode. Cologne, 1967.
Lem, Stanislaw. Faniastyka ifuturologia, I-1I. Cracow, 1973.
Lewis, C. S. Of Other Worlds. New York, 1967.
Moskowitz, Sam. Explorers of the Infinite. Westport, CT, 1974.
--------- . Seekers of Tomorrow. Westport, CT, 1974.
--------- . Strange Horizons. New York, 1976.
Mullen, R. D„ and Darko Suvin, eds. Science-Fiction Studies: Selected
Articles. . . 1973-1975. Boston, 1976.
Nagi, Manfred. Science Fiction in Deutschland. Tubingen, 1972 (B).
Pagetti, Carlo. Il Senso del Futuro. Rome, 1970.
Pehlke, Michael, and Norbert Lingfeld. Roboter und Gartenlaube.
Munich, 1970.
Rose, Mark, ed. Science Fiction. Englewood Cliffs, NJ, 1976.
Rottensteiner, Franz. The Science Fiction Book. New York, 1975.
Sadoul, Jacques. Histoire de la Science Fiction moderne, I—II. Paris, 1975.
Samuelson, David N. Visions of Tomorrow. New York, 1975.
Scholes, Robert. Structural Tabulation. London, 1975.
-------- , and Eric S. Rabkin. Science Fiction: History—Science—Vision.
New York, 1977.
Tuck, Donald H. The Encyclopedia of Science Fiction and Fantasy, I.
Chicago, 1974.
Van Herp. Jacques. Panorama de la science fiction. Verviers, 1973.
288 BIBLIOGRAPHY

Versins, Pierre. Encyclopédie de l’utopie, de la science-fiction el des voyages


extraordinaires. Lausanne, 1972.
Wessels, Dieter. Welt im Chaos. Frankfurt, 1974.
¿antovská-Murray, Irena, and Darko Suvin. “A Bibliography of Gen­
eral Bibliographies of SF Literature.” Science-Fiction Studies 5 (1978)
(B).

See also the journals:

Algol, ed. Andrew Porter, POB 4175, New York City 17, USA.
Extrapolation, ed. Thomas D. Clareson, Wooster College, Wooster, OH,
USA.
Foundation, ed. Peter Nicholls (as of 1978 Malcolm Edwards), North
East London Polytechnic, Dagenham, Essex, G. Britain.
Quarber Merkur, ed. Franz Rottensteiner, Felsenstr. 20, Miesenbach,
Austria.
Science-Fiction Studies, ed. R. D. Mullen and Darko Suvin, English
Depts., Indiana State University, Terre Haute, IN, USA, and McGill
University, Montreal, Qué, H3A 2T6, Canada.
Science-Fiction Times, eds. Hans Joachim Alpers and Ronald M. Hahn,
Weissenburger Str. 6, 2850 Bremerhaven 1, W. Germany.

Most SF magazines publish current book reviews; see for the USA, H.
W. Hall. Science Fiction Book Review Index, 1923-1973. Detroit, 1975; and
for France, the magazine Fiction.

IL Theory of the Fictional Utopia


Bakhtin, Mikhail. Rabelais and His World. Cambridge, MA, 1968.
Baldissera, Alberto. "Il concetto di utopia,” in Gianni Giannotti, ed.
Concezione e previsione del futuro. Bologna, 1971.
Barthes, Roland. Sade, Fourier, Loyola. New York, 1976.
Berneri, Marie-Louise. Joumey Through Utopia. New York, 1971.
Bloch, Ernst. Das Prinzip Hojfnung, I—II. Frankfurt, 1959 (Le Principe
espérance. Paris, 1976-).
Boas, George. Essays on Primitivism and Related Ideas in the Middle Ages.
New York, 1966.
Brüggemann, Fritz. Utopie und Robinsonade. Weimar, 1914.
Buber, Martin. Paths in Utopia. Boston, 1949.
Cioranescu, Alexandre. L’Avenir du passé. Paris, 1972.
Coli, Edoardo. Il Paradiso Terrestre dantesco. Florence, 1896.
Dubois, Claude-Gilbert. Problèmes de l’utopie. Paris, 1968.
Elliott, Robert C. The Shape of Utopia. Chicago, 1970.
BIBLIOGRAPHY 289

Extrapolation 19 (1977) (B).


Giamatti, A, Bartlett. The Earthly Paradise and the Renaissance Epic.
Princeton, 1966.
Graf, Arturo. Miti, leggende e superstizioni del Medio Evo, I. Bologna,
1965.
Herbrüggen, Hubertus Schulte. Utopie und Anti-Utopie. Bochum, 1969.
Hertzler, Joyce Oramei. The History of Utopian Thought. New York,
1965.
Kerényî, Karl. “Ursinn und Sinn wandel des Utopischen,” in Adolf
Portmann, ed., Vom Sinn der Utopie. Eranos-Jahrhuch 1963. Zurich,
1964.
Krysmanski, Hans-Jürgen. Die utopische Methode. Cologne, 1963.
Lalande, André. “Utopie.” in his Vocabulaire technique et critique de la
philosophie. Paris, 1968.
Lovejoy, Arthur O., and George Boas. Primitivism and Related Ideas in
Antiquity, New York, 1965.
Mannheim, Karl. Ideology and Utopia. New York, 1966.
Manuel, Frank E., ed. Utopias and Utopian Thought. Boston, 1967.
Marin, Louis. Utopiques. Paris, 1973.
Morton, A. L. The English Utopia. London, 1969.
Negley, Glenn, and J. Max Patrick. The Quest for Utopia. College Park,
MD, 1971.
Neusüss, Amhelm, ed. Utopie. Berlin, 1968 (B).
Patch, Howard Rollin. The Other World. New York, 1970.
Revue des sciences humaines. No. 155 (1974).
Ruyer, Raymond. L'Utopie et les utopies. Paris, 1950.
Sauer, Gerda-Karl a. Kindliche Utopien. Berlin, 1954.
Schwonke, Martin. Vom Staatsroman zur Science Fiction. Stuttgart, 1957.
Tillich, Paul. Politische Bedeutung der Utopie für das Leben der Völker.
Berlin, 1951.
Viligradter, Rudolf, and Friedrich Krey, eds. Der utopische Roman.
Darmstadt, 1973.
Voigt, Andreas. Die sozialen Utopien. Leipzig, 1906.
Walsh, Chad. From Utopia to Nightmare. Westport, CT, 1972.

III. History of SF to the Eighteenth Century


A. General (including the general history offictional utopia).
See also Bibliography I and II—in particular Bakhtin, Barron, ed.,
Barthes, Berneri, Bloch (1959), Elliott, Extrapolation, Marin, Morton,
Negley-Patrick, Patch, RSH, Ruyer, Schwonke, Walsh, and Zantovskâ-
Suvin.
290 BIBLIOGRAPHY

Beauchamp, Gorman. “Themes and Uses of Fictional Utopias.”


Science-Fiction Studies 4 (1977) (B).
Beer, Max. The General History of Socialism and Social Struggles, I—II.
New York, 1957.
Biesterfeld, Wolfgang. Die literarische Utopie. Stuttgart, 1974 (B).
Cawley, Robert Ralston. Unpathed Waters. London, 1967.
Ceserani, Gian Paolo. I falsi Adami. Milan, 1969.
Cohen, John. Human Robots in Myth and Science. London, 1966.
Duveau, Georges. Sociologie de l'utopie. Paris, 1961.
Ehrmann, Jacques. “Le dedans et le dehors." Poétique, No. 9 (1972).
Eurich, Neil. Science in Utopia. Cambridge, MA, 1967.
Gibson, R. W., and J. Max Patrick. “Utopias and Dystopias, 1500-1750,”
in R. W. Gibson, Si. Thomas More. New Haven, 1961 (B).
Gove, Philip Babcock. The Imaginary Voyage in Prose Fiction. New York,
1975 (B).
Mucchielli, Roger. Le Mythe de la cité idéale. Paris, 1960.
Mumford, Lewis. The Story of Utopias. New York, 1966.
Nicolson, Marjorie. Science and Imagination. Ithaca, NY, 1962.
Patai. Daphne. “Utopia for Whom.” Aphra, No. 5 (1974).
Philmus, Robert M. Into the Unknown. Berkeley, 1970.
Sargent, Lyman Tower. “Themes in Utopian Fiction in English Before
Wells.” Science-Fiction Studies 3 (1976) (B).
Schlanger, Judith E. “Power and Weakness of the Utopian Imagination.”
Diogenes, No. 84 (1973).
Seeber, Hans Ulrich. Wandlungen der Form in der literarischen Utopie.
Göppingen, 1970.
Simon, Heinrich. "Arabische Utopien im Mittelalter.” PFiss. Z. der
Humboldt-Univ. Berlin, Ges.-Sprachwiss. Reihe 12 (1963).
Trousson, Raymond. Voyages aux pays de nulle part. Bruxelless, 1975.
Tuveson, Ernest Lee. Millennium and Utopia. Berkeley, 1949.

B. Extra-European, Antique, Medieval


See also Bibliography II, 111 C, and III F—in particular Beger, Bloch
(1959), Boas, Brunner, Cioranescu, Coli, Förster, Graf, Lovejoy-Boas,
Morton, Patch, Seibt, Siissmuth, Viligradter-Krey, ed., and Volgin
(1975)—and Bibliography III A.

Ackermann, Elfriede Marie. “Das Schlaraffenland“ in German Literature


and Folksong. Chicago, 1944.
Baldry, H. C. Ancient Utopias. Southampton, 1956.
Barker, Ernest. The Political Thought of Plato and Aristotle. London, 1959.
Bauer, Wolfgang. China und die Hoffnung auf Glück. Munich, 1971.
Bompaire, J. Lucien écrivain. Paris, 1959.
BIBLIOGRAPHY 291

Bonner, Campbell. “Dionysiac Magic and the Greek Land of Cock­


aigne.” Trans, Proc. American Philological Ass. 41 (1910).
Chesneaux, Jean. “Egalitarianism and Utopian Traditions in the East.”
Diogenes, No. 62 (1968).
Cornford. Francis Macdonald. Plato’s Cosmology. London, 1956.
Curtius, Ernst Robert. European Literature and the Latin Middle Ages. New
York, 1963.
Ferguson, John. Utopias of the Classical World. New York, 1975.
Finley, M. I. “Utopianism Ancient and Modern," in Kurt H. Wolff and
Barrington Moore, Jr., eds., The Critical Spirit. Boston, 1967,
Fredericks, S. C. "Lucian’s True Historv as SF." Science-Fiction Studies 3
(1976).
Gatz, Bodo. Weltaller, Goldene Zeit und sinnverwandte Vorstellungen.
Spudasmata XVI. Hildesheim, 1967.
Graus, F. "Social Utopias in the Middle Ages." Past and Present, No. 38
(1967).
Gronau, Karl. Der Staat der Zukunft von Platon bis Dante. Braunschweig,
1933.
Machovcovä, Marketa, and Milan Machovec. Utopie blouznivcü a sek-
täru. Prague, 1960.
Manuel, Frank E., and Fritzie P. Manuel. "Sketch for a Natural History
of Paradise." Daedalus 101 (1972).
Merkelbach, Reinhold. Roman und Mysterium in der Antike. Munich,
1962.
Nuita, Seiji. “Traditional Utopias in Japan and the West," in David W.
Plath, ed. Aware of Utopia. Urbana, IL, 1971.
Nutt, Alfred. “The Happy Otherworld in the Mythico-Romantic Litera­
ture of the Irish," in Ku no Meyer, ed.. The Voyage of Bran .. . , I. New
York, 1972.
Peters, Elizabeth. Quellen und Charakter der Paradiesesvorstellungen in der
deutschen Dielung vom 9. bis 12. Jahrhundert. Breslau, 1915.
Pöhlmann, Robert von. Geschichte der sozialen Frage und des Sozialismus in
der antiken Welt, I—II. Munich, 1925.
Rohde, Erwin. Der griechische Roman und seine Vorläufer. Berlin, I960.
Salin, Edgar. Platon und die griechische Utopie. Munich, 1921.
Sch uh I, Pierre-Maxime. La Fabulation platonicien ne. Paris, 1968.
Swanson, Roy Arthur. "The True, the False, and the Truly False: Lu­
cian’s Philosophical SF.” Science-Fiction Studies 3 (1976).
Vallauri, Giovanna. Evemero di Messene. Univ, di Torino: Pubbl. della
facolta di lettere efilosofia 8, No. 3 (1956).
W’estropp, Thomas Johnson. "Brazil and the Legendary Islands of the
North Atlantic." Proc. Royal Irish Academy 30 (1912).
Winston, David. "Iambulus’ Islands of the Sun and Hellenistic Literary
Utopias.” Science-Fiction Studies 3 (1976) (B).
292 BIBLIOGRAPHY

C. Renaissance Utopias
See also Bibliography I, II, III E, and III F—in particular Bakhtin,
Berneri, Bloch (1959), Brunner, Dubois. Elliott (Shape), Giamatti,
Marin, Morton, Patch. Sainéan, Schwonke, Tuveson, ed., Vickers, and
Villgradter-Krey, eds., and Èantovskâ-Suvîn—and Bibliography III A.

1. GENERAL AND THOMAS MORE

Adams, Robert P. “The Philosophic Unity of More’s Utopia." Studies in


Philology 38 (1941).
---------. “The Social Responsibilities of Science in Utopia, New Atlantis
and After," in P. O. Kristeller and P. P. Wiener, eds., Renaissance
Essays. New York, 1968.
Ames, Russell. Citizen Thomas More and his Utopia. Princeton, 1952.
Beger, Lina. "Thomas Morus und Plato." Z.für die gesamte Staatswiss. 35
(1879).
Chambers, R. W. Thomas More. Ann Arbor, 1968.
Dermenghem, E. Thomas Morus et les Utopistes de la Renaissance. Paris,
1927.
Donner, H. W. Introduction to Utopia. New York, 1969.
Dudok, G. Sir Thomas More and his Utopia. Amsterdam, 1923.
Dupont, V. L'Utopie et le Roman Utopique dans la Littérature Anglaise.
Toulouse, 1941.
Förster, Richard. “Lucian in der Renaissance.” Archiv für Literaturgesch.
14 (1937).
Gallagher, Ligeia, ed. More's Utopia and its Critics. Chicago, 1964.
Heiserman, A. R. “Satire in the Utopia.” PMLA 78 (1963).
Hexter, J. H. More’s "Utopia”. New York, 1965.
Massô, Gildo. Education in Utopia. New York, 1972.
Miles, Leland. “The Literary Artistry of Thomas More.” Studies in En­
glish Lit. 6 (1966).
Morris, William. "Introduction” to Thomas More, Utopia. London,
1893 (rpt. in Science-Fiction Studies 3 [1976]).
Nelson, William, ed. Twentieth Century Interpretations of "Utopia”. En­
glewood Cliffs, NJ, 1968.
Pons, Emile. “Les Langues Imaginaires dans le voyage utopique: . . .
Thomas Morus.” R. de lift, comparée 10 (1930).
Rebhorn, Wayne A. “Thomas More's Enclosed Garden." English Lit.
Renaissance 6 (1976).
Reiss, Timothy. “Utopia and Process." Sub-Stance, No. 8 (1974).
Sanderlin, George. “The Meaning of Thomas More’s Utopia." College
English 12 (1950).
Seibt, Ferdinand. “Utopie im Mittelalter.” Historische Z, 208 (1969).
BIBLIOGRAPHY 293

Sum, Edward L„ S. J. The Praise of Pleasure. Cambridge, MA, 1957.


---------. The Praise of Wisdom. Chicago, 1957.
--------- , and J. H. Hexter. “Introduction” to The Complete Works of St.
Thomas More, 4. New Haven, 1965.
Süssmuth, Hans. Studien zur Utopia des Thomas Morus. Münster, 1967,
Sylvester, R. S., and G. P. Marc’hadour, eds. Essential Articles for the
Study of Thomas More. Hamden, CT, 1977.
L’Utopie à la Renaissance. Brussels, 1963.
Volgin, V. P. Ocherki istorii sotsialislicheskikh idei s drevnosti do kontsa XVIII
v.. Moscow, 1975.
2. OTHER (.'TOPOGRAPHERS

Auerbach, Erich. Mimesis. Garden City, NY, 1957.


Badaloni, Nicola. Tommaso Campanella. Milan, 1965.
Beaujour, Michel. Le Jeu de Rabelais. Paris, 1969.
Bierman, Judah. “Science and Society in the New Atlantis and Other
Renaissance Utopias." PMLA 78 (1963).
Blodgett, Eleanor Dickinson. “Bacon’s New Atlantis and Campanella's
Civitas Solis." PMLA 46 (1931).
Bock, Gisela. Thomas Campanella. Tübingen, 1974.
Diéguez, Manuel de. Rabelais par lui-même. Paris, I960.
Doren, Alfred. “Campanella als Chiliast und Utopist,” in Kultur- und
Universalgeschichte: Walter Goetz zu seinem 60. Geburtstag. . . . Leipzig,
1927.
Farrington. Benjamin. The "New Atlantis" of Francis Bacon. Richmond
Hill, 1965.
François Rabelais: IVe Centenaire de sa Mort. Geneva, 1953.
Gorfunkel, A. Tomazo Kampanela. Moscow, 1969.
Greene, Thomas M. Rabelais. Englewood Cliffs, NJ, 1970.
Grendler, Paul F. “Utopia in Renaissance Italy: Doni’s New World.” J. of
the History of Ideas 26 (1965).
Kaiser, Walter. Praisers of Folly. Cambridge, MA, 1963.
Knights, L. C. Explorations. Westport, CT, 1975.
Lefebvre, Henri. Rabelais. Paris, 1955.
Lefranc, Abel. Les Navigations de Pantagruel. Geneva, 1967.
Marin, Louis. “Les Corps utopiques rabelaisiens." Littérature, No. 21
(1976).
Paris» Jean, Rabelais au futur. Paris, 1970.
Reiss, Timothy J. “Structure and Mind in Two Seventeenth-Century
Utopias: Campanella and Bacon.” Yale French Studies 49 (1973).
Sainéan, Lazare. La Langue de Rabelais, I-IL Paris, 1922-23.
Saulnier, V.-L. Le Dessein de Rabelais. Paris, 1957.
Scholtz, Harald. Evangelischer Utopismus bei J. V. Andreae. Stuttgart, 1957.
294 BIBLIOGRAPHY

Spitzer, Leo. Romanische Slit- und Literaturstudien. Marburg, 1931.


---------. “Ancora sul prologo al Gargantua di Rabelais.” Studi francesi 27
(1965).
White, Howard B. Peace Among the Willows: The Political Philosophy of
Francis Bacon. The Hague, 1968.
Wiener, Harvey S. “ ‘Science or Providence’: Toward Knowledge in
Bacon’s New Atlantis.'' Enlightenment Essays 3 (1972).
3. SHAKESPEARE

Allen, Don Cameron. Image and Meaning. Baltimore, 1960.


Kermode, Frank. “Introduction,” to The Tempest. Arden edn. of the Works
of William Shakespeare. London, 1963.
Kott, Jan. Shakespeare Our Contemporary. Garden City, NY, 1966.
Langbaum, Robert, ed. The Tempest. Signet Classic Shakespeare. New
York, 1964.
Leech, Clifford. Shakespeare’s Tragedies. London, 1950,
Lovejoy, A. O. Essays in the History of Ideas. New York, 1960.
Palmer, D. J., ed. Shakespeare: “The Tempest." London, 1968.
D. The "Planetary Novel” and Cyrano
See also Bibliography [I, III B, III C, III E, and III F—in particular
Atkinson, Bakhtin, Cornelius, Eddy, Graf, Paris, Rozanova, Schwonke.
Toldo, and Tuzet—and Bibliography HI A.

Al cover, Madeleine. La Pensée philosophique et scientifique de Cyrano de


Bergerac. Geneva, 1970.
Blanchot, Maurice. “Cyrano de Bergerac,” in A. Adamov, et al. Tableau
de la littérature française, I. Paris, 1962.
Boas, George. The Happy Beast. New York. 1966,
Brandwajn, Rachmiel. Cyrano de Bergerac. Warsaw, 1960.
Chambers, Ross. “L’Autre Monde, ou le mythe du libertin." Essays in
French Lit, 8 (1971).
Dunlop. Alexander. “The Narrative Function of Ideas in Cyrano’s Estats
et empires de la lune.” Romance Notes 13 (1971).
Harth, Erica. Cyrano de Bergerac and the Polemics of Modernity. New York,
1970.
Harvey, Howard G. “Cyrano de Bergerac and the Question of Human
Liberties." Symposium 4 (1950),
Laugaa, Maurice. “Introduction” to Cyrano de Bergerac, Voyage dans la
Lune. Paris, 1970.
----- —. “Lune, ou l’Autre." Poétique, No. 3 (1970).
Lavers, A. “La Croyance à l’unité de la science dans L'Autre Monde de
Cyrano de Bergerac.” Cahiers du Sud 45(1959).
BIBLIOGRAPHY 395

Lear, John. ‘‘Introduction and Interpretation,” in Kepler's Dream. Berke­


ley, 1965,
Liger, Christian. “Les cinq envois de Cyrano.” Nouvelle R. française 18
(1965).
Mongrédien, Georges. Cyrano de Bergerac. Paris, 1964.
Nicol son. Marjorie Hope. Voyages to the Moon. Havertown, PA, 1973.
Pin tard, René. Le Libertinage érudit dans la première moitié du XVIle siècle.
Paris, 1948.
Ridgely, Beverly S. “A Sixteenth-Century French Cosmic Voyage:
Nouvelles des Regions de la Lune.” Studies in the Renaissance 4 (1957).
Rosenfield, I^onora Cohen. From Beast-Machine to Man-Machine. New
York, 1968.
Spink, John Stephenson. “Form and Structure: Cyrano de Bergerac’s
Atomistic Conception of Metamorphosis,” in Literature and Science:
Proc, of the 6th Congress of IFMLL. Oxford, 1955.
---------, French Free-Thought From Gassendi to Voltaire. New York, 1969.
Thibaudet, Albert. “Réflexion sur la littérature: le roman de l'aven­
ture. ” Nouvelle R. française 13 (1919).
Van Baelen, Jacqueline. “Reality and Illusion in L’Autre Monde." Yale
French Studies 49 ( 1973).
Weber, H. “Introduction” to Cyrano de Bergerac, L'Autre Monde. Paris,
1959.

E. "Gulliver’s Travels”
See also Bibliography I, II, III D, and III F—in particular Adams
(1962), Boas (Happy Beast), Elliott (Shape), Kagarlitskii, Ne gley-Pat rick,
Nicolson (1973), and RSH—and Bibliography III A.

Atkinson, Geoffroy. The Extraordinary Voyage in French Literature, I—II.


New York, 1969.
Bonner, William H. Captain William Dampier. London, 1934.
Brady, Frank, ed. Twentieth Century Interpretations of “Gulliver’s Travels.”
Englewood Cliffs, NJ, 1968.
Brunetti. Giuseppe. “Swift e la satira della scienza,” in Mario Praz, ed.,
English Miscellany 24. Rome, 1973-74.
Carnochan, W. B. Lemuel Gulliver’s Minor for Man. Berkeley, 1968.
Case, Arthur E. Four Essays on "Gulliver’s Travels.” Gloucester, MA,
1958.
Donoghue, Denis, ed. Jonathan Swift. Harmondsworth, 1971.
Eddy, William. Gulliver’s Travels. New York, 1963.
Elliott, Robert C. The Power of Satire. Princeton, 1970.
Forrester, Kent. “They Shoot Horses, Don’t They?” Kentucky Philological
Ass. Bull. (1974).
296 BIBLIOGRAPHY

Foster, Milton P., ed. A Casebook on Gulliver among the Houyhnhnms. New
York. 1961.
Frantz, R. W. "Swift’s Yahoos and the Voyagers.” Modem Philology 29
(1931).
Greenberg, Robert A., ed. Gulliver’s Travels. New York, 1970 (B).
Jeffares, A. Norman, ed. Fair Liberty was all his Cry. London, 1967.
---------, ed. Swift. London, 1968.
Kiernan, Colin. "Swift and Science.” Historical J. 14 (1971).
Price, Martin. Swift's Rhetorical Art. Carbondale, IL, 1973.
Quintana, Ricardo. Swift. London, 1962.
Sacks, Sheldon. “Fiction” and the Shape of Belief. Berkeley, 1966.
Stanzel, Franz. “Gulliver's Travels: Satire, Utopie, Dystopie.” Moderne
Sprachen 7 (1963).
Sutherland, John H. “A Reconsideration of Gulliver's Third Voyage.”
Studies in Philology 54 (1957).
Sutherland, W. O. S., Jr. The Art of the Satirist. Austin, 1965.
Toldo, Pietro. Les Toyages merveilleux de Cyrano de Bergerac et de Swift et
leur rapports avec l’oeuvre de Rabelais. Paris, 1907.
Tuveson, Ernest, ed. Swift. Englewood Cliffs, NJ, 1964.
Vickers, Brian. The Satiric Structure of “Gulliver's Travels” Ö1 More's
“Utopia.” Oxford, 1968.
---------, ed. The World ofJonathan Swift. Oxford, 1968.
Voigt, Milton. Swift and the Twentieth Century. Detroit, 1964.
Williams, Kathleen. Jonathan Swift and the Age of Compromise. Lawrence,
KS, 1958.

F. Other Works of the Seventeenth and Eighteenth Centuries


See also Bibliography I, 11, Ill C, III D, III E, and VI—in particular
Atkinson, Bailey, Barthes (1976), Berneri, Bloch (1959), Bruggemann,
Chistov, Cioranescu, Clarke, Dupont, Fedosov, Herbriiggen, Kagar-
litskii, Kizevetter, Morton, Nicolson (1973), RSH, Ruyer, Schwonke,
Sipovskii (1909-10), Sipovskii (1924), Spink (1969), Suvin (1976), Sviat-
lovskii, Thibaudet, and Volgin (1975)—and Bibliography III A.

Adams. Percy G. Travelers and Travel Liars, 1660-1SOO. Berkeley, 1962.


Baczko, B. “Lumières et utopie.” Annal« 26 (1971).
Bélaval, Yvon. "Le Conte philosophique," in W. H. Barber, et al., eds.,
The Age of Enlightenment. Edinburgh, 1967.
Blitzer, Charles. An Immortal Commonwealth: The Political Thought of
James Harrington. Hamden, CT, 1970.
Brunner, Horst. Die poetische Insel. Stuttgart, 1967.
Chérel, Albert. De Télémaque à Candide. Paris, 1958.
BIBLIOGRAPHY 297

Chinard, Gilbert. L’Amérique et le rêve exotique dans la littérature française


au XVIIe et au XVIIIe siècle. Geneva, 1970.
Coe, Richard N. Morelly. [East] Berlin, 1961.
Cornelius, Paul. Languages in Seventeenth- and Early Eighteenth-Century
Imaginary Voyages. Geneva, 1965.
Coste, B.Mably. Paris, 1975.
Courbin, J.-Cl. Le Monde de Restif. Paris, 1962.
Dalnekoff, Donna Isaacs. “The Meaning of Eldorado: Utopia and Sat­
ire in Candide.” Studies on Voltaire and the Eighteenth Century, No. 127
(1974).
Dédéyan, Charles. Le Télémaque de Fénelon. Paris. 1967.
Elliott, Robert C. “The Costs of utopia," Studies on Voltaire and the
Eighteenth Century, No. 151 (1976).
Ellison, Lee Monroe. “Gaudentio di Lucca." PMLA 50 (1935).
Girsberger, Hans. Der utopische Sozialismus des 18. Jahrhunderts in Frank­
reich. Wiesbaden, 1973.
Gordon, L. S. “Zabytyi Utopist XVIIlogo veka Tifèn de Larosh," in
Isloriia sotsialisticheskikh uchenii. Moscow, 1962.
Hafter, Monroe Z. “Toward a History of Spanish Imaginary Voyages.”
Eighteenth Century Studies 8 (1975).
Heuvel, J. Van den. Voltaire dans ses contes. Paris, 1967.
Hinterhäuser, Hans. Utopie und Wirklichkeit bei Diderot. Heidelberg,
1957.
Holiendahl, Peter Uwe. “Zum Erzählprobiem des utopischen Romans
im 18. Jahrhundert," in Helmut Kreuzer, cd., Gestaltungsgeschichte
und Gesellschaftsgeschichte. Stuttgart, 1969.
loannisian, A. R. “Restif de la Bretonne ei le communisme utopique.”
La Pensée (March-April 1958).
Krauss, Werner. Fontenelle und die Aufklärung. Munich, 1969.
Lachèvre, Frédéric. Les Successeurs de Cyrano de Bergerac. Paris, 1922.
Lanson, Gustave. “Les Origines et les premières manifestations—” and
"Formation et développement—de l’esprit philosophique dans la lit­
térature française." R. des cours et conférences 16-18 (1907-10).
Le Breton, A. Le Roman au XVIIIe siècle. Geneva, 1970.
Le Flamanc, Auguste. Les Utopies prérevolutionnaires et la philosophie du
XVHIe siècle. Paris, 1934.
Lichtenberger, André. Le Socialisme au XVIIIe siècle. Osnabrück, 1970.
Mayer, Hans. "Die alte und neue epische Form." in his Von Lessing Ins
Thomas Mann. Pfullingen, 1959.
McNelis, James I., Jr. “Introduction” to Ludvig Holberg, The Journey of
Niels Klim to the World Underground. Lincoln, NE, 1960.
Messac, Régis. “Voyages modernes au centre de la terre.” R. de litt,
comparée 0 (1929).
298 BIBLIOGRAPHY

Mühll, Emanuel von der. Denis l'eiras et son "Histoire des Sévarambes”
(1677-1679). Paris. 1938.
Nerval, Gérard de. Les Illuminés. Verviers, 1973.
Paludan, Julius. Om Holbergs Niels Klim. Copenhagen, 1878.
Patrick, J. Max. “A Consideration of ¿a Terre Australe Connue by Gabriel
de Foigny." PMLA 61 (1946).
Pons, Emile. ‘Le ‘Voyage,’ genre littéraire au XVII le siècle.” Bull,
faculté des lettres de Strasbourg 4 (1925-26).
---------. “Les Langues imaginaires dans le voyage utopique: . .. Vairasse
et Foigny.” R. de litt, comparée 12 (1932).
Poster, Mark. The Utopian Thought of Restif de la Bretonne. New York,
1971.
Rihs, Charles. Les Philosophes utopistes. Paris, 1970.
Rozanova, A. A. Sotsial’naia i nauchnaia fantastika v klassirheskoi frantsuz-
koi literature XVI-XIX w.. Kiev, 1974.
Sainéan, Lazare. L'Influence et la réputation de Rabelais. Paris, 1930.
Sareil, Jean. Essai sur “Candide." Geneva, 1967.
Schmidt, Arno. Dya Na Sore. Karlsruhe, 1959.
Tuzet, Hélene. Cosmos et imagination. Paris, 1965.
Venturi, Franco. Utopia e riforma neWIlluminismo. Turin, 1970.
Volgin, V. P. Franlsuzkn utopieheskii kommunizm. Moscow, 1960.
Wade, Ira O. Voltaire’s “Micromégas", Princeton, 1950.
Wijngaarden, Nicolaas van. Les Odyssées philosophiques en France entre
1616 et 1789. Haarlem, 1932.

IV. History of SF from the French Revolution to H. G. Wells


A. Radical Rhapsody and Romantic Recoil
See also Bibliography I, II, III A, III C, and III F, in particular
Aldiss, Bailey, Barron, ed., Barthes, Beer, Berneri, Bloch, Clarke,
Cohen, Dupont, Elliott (Shape), Ketterer, Marin (1973), Morton,
Mumford, Negley-Patrick, Nicolson, Philmus, Sargent, Schwonke,
Tuzet, and 2antovska-Suvin.
1. IN FRANCE

Angenot, Marc. "Science Fiction in France Before Verne.” Science-


Fiction Studies 5 (1978).
Ansart, Pierre. Sociologie de Saint-Simon. Paris, 1970.
Béciard, Léon. Sébastien Mercier. Paris, 1903.
Cole, G. D. H. A History of Socialist Thought, I—V. New York, 1953-58.
Debout Oleskiewicz, Simone. “Introduction” to Oeuvres complètes de
Charles Fourier, 1. Paris, 1966.
BIBLIOGRAPHY 299

--------- . “Préface" to Oeuvres completes de Charles Fourier, VII. Paris, 1967.


Desanti, Dominique. Les Socialistes de l'utopie. Paris, 1970.
Desroche, Henri. La Société festive. Paris, 1975.
Durkheim, Emile. Socialism and Saint-Simon. Yellow Springs, OH, 1958.
Engels, Friedrich. “Socialism: Utopian and Scientific,” in Karl Marx
and Friedrich Engels, Selected Works In One Volume. New York, 1968.
Lehouck, Emile. Fourier, aujourd’hui. Paris, 1966.
Leroy, Maxime. Histoire des idées sociales en France, I-III, Paris, 1947-54.
Majewski, Henry F. "Grainville’s Le Dernier Homme.” Symposium 17
(1963).
Manuel, Frank. The New World of Henri Saint-Simon. Cambridge, MA,
1956.
---------. The Prophets of Paris. New York, 1965.
Riasanovsky, Nicholas V. The Teaching of Charles Fourier. Berkeley, 1969.
Volgin, V. P. Sen-Simon i sen-simonizrn. Moscow, 1961.

2. GENERAL AND IN ENGLAND

Auden, W. H. The Enchaféd Flood. New York, 1967.


Ault, Donald D. Visionary Physics. Chicago, 1974.
Bloom, Harold. The Ringers in the Tower. Chicago, 1971.
Brailsford, H. N. Shelley, Godwin and Their Circle. London, 1954.
Bronowski, Jacob. William Blake and the Age of Revolution. London, 1972.
Buchen, Irving H. "Frankenstein and the Alchemy of Creation and
Evolution.” The Wordsworth Circle 8 (1977).
Cameron, Kenneth Neill. The Young Shelley. New York, 1973.
Curran, Stuart, and Joseph Anthony Wittreich, Jr., eds. Blake’s Sublime
Allegory. Madison, 1973.
DiSalvo, Jackie. “Blake Encountering Milton,” in Joseph Anthony Witt­
reich, Jr., ed., Milton and the Line of Vision. Madison, 1975.
Erdman, David V. Blake: Prophet Against Empire. Princeton, 1969.
Fisher, Peter F. The Valley of Vision. Toronto, 1961.
Frye, Northrop. Fearful Symmetry. Princeton, 1969.
Goldberg. M.A. “Moral and Myth in Mrs. Shelley’s Frankenstein.” Keats-
Shelley J. 8 (1959).
Grabo, Cari H. A Newton Among Poets. New York, 1968.
Hughes, A. M. D. The Nascent Mind of Shelley. Oxford, 1971.
Joseph, M. K. “Introduction” to Mary W. Shelley, Frankenstein or the
Modern Prometheus. London, 1971.
Kayser, Wolfgang. The Grotesque in Art and Literature. Gloucester, MA,
1963.
Kreutz, Christian, Das Prometheussymbol in der Dichtung der englischen
Romantik. Palaestra Bd. 236. Göttingen, 1963.
300 BIBLIOGRAPHY

Levine, George, "Frankenstein and the Tradition of Realism.” Novel 7


(1973).
Lewis, C. S, Rehabilitations. London, 1939.
Massey, Irving. The Gaping Pig. Berkeley, 1976.
McNeice, Gerald. Shelley and the Revolutionary Idea. Cambridge, MA,
1969.
Morton, A. L. The Everlasting Gospel. Folcroft, PA, 1974.
Mumford, Lewis, Technics and Civilization. New York, 1963.
Palacio, Jean de. Mary Shelley dans son oeuvre. Paris, 1969.
Pollin, Burton R. "Philosophical and Literary Sources of Frankenstein.”
Comparative Lit. 17 (1965).
Ridenour, George M„ ed. Shelley. Englewood Cliffs, NJ, 1965.
Rieger, James, The Mutiny Within. New York, 1967.
Rubenstein, Marc A. “ ‘My Accursed Origin’: The Search for the
Mother in Frankenstein.” Studies in Romanticism 15 (1976).
Sa mb rook, A. J. “A Romantic Theme: The Last Man." Forum for
Modern Language Studies 2 (1966).
Schorer, Mark. William Blake: The Politics of Vision. New York, 1959.
Sir Walter Scott on Novelists and Fiction. Ed. 1. Williams. London, 1968.
Small, Christopher. Mary Shelley's "Frankenstein." Pittsburgh, 1973.
Smith, Elton E., and Esther G. Smith. William Godwin. New York, 1966.
Walling, William A. Mary Shelley. New York, 1972.
Ward, Aileen. “The Forging of Orc; Blake and the Idea of Revolution.”
Triquarterly, No. 23/24 (1972).
Wasserman, Earl R. Shelley’s “Prometheus Unbound." Baltimore, 1965.
Wilner, Eleanor. Gathering the Winds. Baltimore, 1975.
Woodings, R. B., ed. Shelley. London, 1968.
3. IN THE UNITED STATES

Adams, Richard P. “Hawthorne.” Tulane Studies in English 8 (1958).


Bailey, J. O. "Sources of Poe’s Arthur Gordon Pym, ‘Hans PfaaF and
Other Pieces.” PMLA 57 (1942).
Beaver, Harold. “Introduction" and “Commentary," in idem, ed., The
Science Fiction of Edgar Allan Poe. Harmondsworth, 1976 (B).
Carlson, Eric W., ed. The Recognition of Edgar Allan Poe. Ann Arbor,
1966.
Chase, Richard. Herman Melville. New York, 1949.
Davidson, Edward H. Poe. Cambridge, MA, 1957.
Davis, Merrell R. Melville’s Mardi. New Haven, 1952.
Falk, Doris V'. “Poe and the Power of Animal Magnetism." PMLA 84
(1969).
Fiedler, Leslie. Love and Death in the American Novel. London, 1967.
Fogle, Richard Harter. Hawthorne's Fiction. Norman, OK, 1964.
BIBLIOGRAPHY 301

Franklin, H. Bruce. The Wake of the Gods. Stanford, 1963.


---------. Future Perfect. New York, 1968.
Halliburton, David. Edgar Allan Poe. Princeton, 1973.
Howarth, William, ed. Twentieth Century Interpretations of Poe’s Tales. En­
glewood Cliffs, NJ, 1971.
Kaul, A. N. The American Vision. New Haven, 1963.
Ketterer, David. “The S-F Element in the Work of Poe.” Science-Fiction
Studies 1 (1974) (B).
Levin, Harry. The Power of Blackness. New York, 1960.
Lynen, John. The Design of the Present. New Haven, 1969,
M atth ie s s e n, F. O. American Renaissance. New York, 1941.
Parrington, Vernon Louis. Main Currents in American Thought, I—Ill-
New York, 1958.
Pollin, Burton R. "Rappaccini’s Daughter—Sources and Names.” Names
14 (1966),
Vernon, John. “Melville's The Bell Tower." Studies in Short Fiction 7
(1970).

B. Verne and the Space-Binding Machines


See also Bibliography I, IL Hl A, III F, and IV A, in particular
Aldiss, Angenot, Auden, Bailey, Barron, ed., Beer, Bloch (1959),
Bridenne, Cohen, Elliott (1970), Gove, Lem, Messac, Moskowitz (Ex­
plorers), Mumford (Story), Nagl.^Nicolson (1973), Philmus, Sargent,
Scholes-Rabkin, Schwonke, and Zantovskâ-Suvin, as well as the three
bibliographic surveys on Verne cited in note 3 to chap. 7.

1. VERNE

Ailott, Kenneth. Jules Verne. Port Washington, NY, 1974.


Allotte de la Fuve, Marguerite. Jules Verne. Paris, 1966.
Andreev, Kirill. Tri zhizni Zhiulia Verna, Moscow, 1960.
L’Arc, No. 29 (1966).
Barthes, Roland. “Par où commencer?" Poétique. No. 1 (1970).
---------. “The Nautilus and the Drunken Boat,” in his Mythologies. St.
Albans, Herts., 1973.
Butor, Michel. “Le Point suprême et l’âge d'or à travers quelques
oeuvres de Jules Verne,” in his Essais sur les modernes. Paris, 1964.
Cahiers de l'Heme, No. 25 (1974).
Chesneaux, Jean. The Political and Social Ideas of Jules Verne. London,
1972.
Cohen, M. “Remarques à propos de la manière d’écrire de Jules
Verne," in Im Dienste der Sprache: Festschrift V. Klemperer. Halle, 1958.
Compère, Daniel. Approche de File chez Jules Verne. Paris, 1977.
302 BIBLIOGRAPHY

Diesbach, Ghislain de. Le Tour de Jules Verne en quatre-vingt livres. Paris,


1969.
Escaich, René. Voyage au monde de Jules Verne. Paris, 1955.
Europe, No. 112/113 (1955).
Huet, Marie-Thérèse. L'Histoire des “Voyages extraordinaires". Paris, 1973.
Jules-Verne, Jean. Jules Verne. Paris, 1973.
Macherey, Pierre. “Jules Verne ou le récit en défaut,” in his Pour une
théorie de la production littéraire. Paris, 1966.
Martin, C.-N. Jules Verne et son oeuvre. Lausanne, 1971.
More, Marcel. Le très curieux Jules Verne. Paris, 1960.
—------ . Nouvelles explorations de Jules Verne. Paris, 1963.
Paroiénie, A., and C. Bonnier de la Chapelle. Histoire d’un éditeur et de
ses auteurs: P. J. Hetzel. Paris, 1953.
Portuondo, José Antonio. “Jules Verne's America." Americas 9 (1957).
Poulet, Georges. Metamorphoses of the Circle. Baltimore, 1966.
Serres, Michel. Jouvences sur Jules Verne. Paris, 1974.
Vierne, Simone. L’Ile mystérieuse de Jules Verne. Paris, 1973.
---------.Jules Verne et le roman initiatique. Paris, 1974.

2. OTHERS

Amis, Kingsley. “Afterword” to Samuel Butler, Erewhon. New York,


1960.
Christensen, Allan Conrad. Edward Bulwer-Lytton. Athens, GA, 1976.
Clarke, I. F. “The Nineteenth-Century Utopia.” The Quarterly R. 296
(1958).
---------. Voices Prophesying War 1763-1984 [sic], London, 1966 (B).
Cole, G. D. H. Samuel Butler. London, 1961.
Deenen, Maria. Le Merveilleux dans l’oeuvre de Villiers de l’Isle Adam. Paris,
1939.
Holt, Lee E. Samuel Butler. New York, 1964.
Parks, M. G. “Strange to Strangers Only." Canadian Lit., No. 70 (1976).
Schepelmann, Wolfgang. Die englische Utopie im Uehergang: von Bulwer-
Lytton bis H. G. Wells. Vienna, 1975.
Schlösser, Anselm. “Der viktorianische Gulliver.” Z. für Anglistik und
Amerikanistik 9 (1961).
Seeber, Hans Ulrich. “Gegenutopie und Roman: B ul wer-Lytton's The
Coming Race (1871).” Deutsche Vierteljahresschrift für Lit.wiss. und Geis-
tesgesch. 45 (1971).
Sussman, Herbert L. Victorians and the Machine. Cambridge, MA.,
1968.
Willey, Basil. Darwin and Butler. London, 1960.
Woodcock, George. “De Mille and the Utopian Vision.’’^. of Canadian
Fiction 2 (1973).
BIBLIOGRAPHY 303

C. Anticipating the Sunburst


See also Bibliography 1, II, 111 A, HI C, III F, IV A, and IV B, in
particular Bailey (1972), Barron, ed., Beauchamp, Beer, Bloch (1959),
Bridenne, Dupont, Elliott (1970), Elliott (1976), Extrapolation (1977),
Franklin (1968), Kaul, Ketterer (New Worlds), Lewis (1939), Morton
(1969), Mumford (1963), Mumford (1966), Parrington, Patai, Sargent,
Schepelmann, and Zantovska-Suvin.

1. GENERAL AND THE L'TO POG RAP HERS

Aaron, Daniel. Men of Good Hope. New York, 1961.


Bleich, David. “Eros and Bellamy.” American Quarterly 16 (1964).
Bowman, Sylvia E. The Year 2000. New York, 1958.
--------- , et al. Edward Bellamy Abroad. New York, 1962.
Brantlinger, Patrick. "News From Nowhere." Victorian Studies 19 (1975).
Brooks, Van Wyck. New England: Indian Summer. New York, 1950.
Calhoun, Blue. The Pastoral Vision of William Morris. Athens, GA, 1975.
Cole, G. D. H. “Introduction" to William Morris, Selected Writings. Lon­
don, 1948.
Cornet, Robert J. "Rhetorical Strategies in Looking Backward." Markham
R. 4 (1974).
Forbes, Allyn B. “The Literary Quest for Utopia, 1880-1900.”/. of
Social Forces 6 ( 1927/28).
Goode, John. “William Morris and the Dream of Revolution,” in John
Lucas, ed.. Literature and Politics in the Nineteenth Century. London,
1971.
Henderson, Harry B., III. Pistons of the Past. New York, 1974.
Hicks, Granville. Figures of Transition. Westport, CT, 1969.
Hofstadter, Richard. Social Darwinism in American Thought. New York,
1959.
Howells, W. D. “Edward Bellamy," in Edward Bellamy, The Blindmnn's
World. London, [1898],
Jaher, Frederic Copie. Doublers and Dissenters. Glencoe, IL, 1964,
Kasson, J. F. Civilizing the Machine. New York, 1976.
Kocmanova, Jessie. "Two Uses of the Dream Form as a Means of Con­
fronting the Present With the Past." Brno Studies in English 2 (1960).
Lewis, Arthur O. “Introduction,” in idem, ed., American Utopias, New
York, 1971.
Lindsay, Jack. William Morris, Writer. London, 1961.
Lokke, Virgil. "The American Utopian Anti-Novel,” in Ray Browne et
al. eds.. Frontiers of American Culture. Lafayette, IN, 1968.
Martin, Jay. Harvests of Change. Englewood Cliffs, NJ, 1967.
Meier, Paul. La Pensée utopique de William Morris. Paris, 1972.
304 BIBLIOGRAPHY

Middlebro', Tom. “Brief Thoughts on News From Nowhere.” J. of the


William Morris Society 2 (1970).
Parrinder, Patrick. “News From Nowhere, The Time Machine, and the
Break-Up of Classical Realism.” Science-Fiction Studies 3 (1976).
Parrington, Vernon L., Jr. American Dreams. New York, 1964 (B).
Poli, Bernard. Le Roman américain 1865-1917. Paris, 1972.
Redmond, James. “Introduction" to William Morris, News From Nowhere.
London, 1970.
Ridge, Martin, Ignatius Donnelly. Chicago, 1962.
Roemer, Kenneth N. The Obsolete Necessity. Kent, OH, 1976 (B).
Sadler, Elizabeth. “One Book’s Influence." New England Quarterly 17
(1944).
Saxton, Alexander. "Caesar’s Column." American Quarterly 19 (1967),
Schiffman, Joseph, “Edward Bellamy's Altruistic Man." American Quar­
terly 6 (1954).
-------- . “Edward Bellamy and the Social Gospel,” in Cushing Strout,
ed., Intellectual History in America, I. New York, 1968.
Shurter, Robert L. "The Literary Work of Edward Bellamy." American
Lit. 5 (1933).
Taylor, Waiter F. The Economic Novel in America. New York, 1964.
Thomas, John L. “Introduction" to Edward Bellamy, Looking Backward,
2000-1887. Cambridge, MA, 1967.
Thompson, E. P. William Morris, Romantic to Revolutionary. New York,
1977.
Towers, Tom H. “The Insomnia of Julian West.” American Lit. 47
(1975).
Williams, Raymond. Culture and Society 1780-1950. Harmondsworth,
1966.
-------- . The Country and the City. New York, 1973.
2. TWAIN AND OTHERS

Budd, Louis. Mark Twain: Social Philosopher. Port Washington, NY,


1973.
Cox, James M. Mark Twain: The Fate of Humor. Princeton, 1966.
Foner, Philip S. Mark Twain: Social Critic. New York, 1966.
Griffin, Martin 1, J.FranA R. Stockton. Pon Washington, NY, 1965.
Griffith, Clark. "Merlin’s Grin." New England Quarterly 48 (1975).
Kaplan, Justin. "Introduction” to Mark Twain, A Connecticut Yankee in
King Arthur’s Court. Harmondsworth, 1971.
Long, E. Hudson. Mark Twain Handbook. New York, 1957.
Salomon, Roger B. Twain and the Image of History. New Haven, 1961.
Smith, Henry, Nash. Mark Twain's Fable of Progress. New Brunswick,
1964.
BIBLIOGRAPHY 305

---------. Mark Twain; The Development of A Writer. New York, 1972.


Taylor, Walter Fuller. “Mark Twain and the Machine Age." The South
Atlantic Quarterly 37 (1938).

V. H.G. Wells and his SF Context


See also Bibliography I, II, 111 A, III D, III F, IV B, and IV C, in
particular Aldiss, Bailey, Barron, ed., Bloch (1959), Clareson, ed.
(1971), Clarke (1966), Hillegas, Kagarlitskii, Morton, Moskowitz (1976),
Mumford (1966), Nagi, Nicolson (1973), Parrinder (1976), Philmus,
Schepelmann, Scholes-Rabkin, Schwonke, Tuzet, and the notes to
chap. 10.
Bellamy, William. The Novels of Wells, Bennett and Galsworthy: 1890-1910.
London,1971.
Bergonzi, Bernard. The Early H. G. Wells. Manchester, 1969.
-------- . “Wells, Fiction and Politics,” in his The Turn of a Century. New
York, 1973.
---------, ed. H. G. Wells. Englewood Cliffs N.J., 1976.
Borges, J. L. “The Flower of Coleridge” and “The First Wells," in his
Other Inquisitions 1937-1952. New York, 1968.
Brooks, Van Wyck. The World ofH. G. Wells. St. Clair Shores MI, 1970.
Cantril, Hadley. The Invasion from Mars. New York, 1966.
Caudwell, Christopher. “H. G. Wells: A Study in Utopianism," in his
Studies and Further Studies in a Dying Culture. New York, 1971.
Hammond, J. R. Herbert George Wells. New York, 1977 (B).
Hillegas, Mark R. “Cosmic Pessimism in H. G. Wells's Scientific Ro­
mances." Papers of the Michigan Academy of Sciences, Arts, and Letters 46
(1961).
--------- . “Introduction” to H. G. Wells, A Modem Utopia. Lincoln, NE,
1967.
---------. “Martians and Mythmakers,” in Ray B. Browne et al., eds.,
Challenges in American Culture. Bowling Green, OH, 1970.
---------. “Victorian ‘Extraterrestrials',” in Jerome H. Buckley, ed., The
Worlds of Victorian Fiction. Harvard English Studies 6. Cambridge,
MA, 1975.
Hughes, David Y. '‘The War of the Worlds in the Yellow Press." Jo umalism
Quarterly 43 (1966).
Kagarlitski, Julius. The Life and Thought of H. G. Wells. London, 1966;
augmented edn. H. G. Wells: La vita e le opere. Milan, 1974.
Locke, George. “Wells in Three Volumes?” Science-Fiction Studies 3
(1976).
Lodge, David. The Novelist at the Crossroads. Ithaca, 1971.
306 BIBLIOGRAPHY

Mackenzie, Norman and Jeanne. The Time Traveller. London, 1973; in


USA as//. G. Wells: A Biography. New York, n.d. [1973],
Morton, Peter R, “Biological Degeneration.” Southern R., No. 9 (1976).
Moskowitz, Sam. “George Griffith,” in George Griffith, The Raid of “Le
Vengewr." London, 1974.
Nicholson, Norman.//. G. Wells. Folcroft. PA, 1973.
Parrinder, Patrick, ed. H. G. Wells: The Critical Heritage. London, 1972.
---------. H. G. Wells. New York, 1977.
Philmus, Robert M. “Revisions of the Future: The Time Machine." J. of
General Education 28 (1976).
Philmus, Robert M., and David Y. Hughes, eds. Early Writings in Science
and Science Fiction by H. G. Wells. Berkeley, 1975.
Platzner, Robert L. “H. G. Wells's ‘Jungle Book’; The Influence of
Kipling on The Island of Dr. Moreau." Victorian Newsletter 36 (1969).
Pritchett, V. S. The Living Novel. London, I960.
Raknem, Ingvald. H. G. Wells and His Critics. Oslo, 1962.
Suvin, Darko, with Robert M. Philmus, eds. H. G. Wells and Modem
Science Fiction. Lewisburg, PA, 1977 (B).
Vernier, Jean-Pierre. H. G. Wells el son temps. Rouen, 1971 (B).
---------. Wells at the Turn of the Century. Wells Society Occasional Papers.
London,1973.
Wagar, W. Warren. H. G. Wells and the World State. Freeport, NY, 1971.
West, Geoffrey. H. G. Wells. Folcroft, PA, 1973.
Woodcock, George. “The Darkness Violated by Light.” Malahat R. 26
(1973).
Zamyatin, Yevgeny. “H. G. Wells,” in Mirra Ginsburg, ed., A Soviet
Heretic. Chicago, 1970.

VI. Russian SF to 1958


See also Bibliography I, II, III A, III C, and V, in particular Bar-
meyer, ed., Beer, Bowman et al, Elliott, Gurevich, Hillegas, Kagarlitskii,
Scholes-Rabkin, Suvin-Philmus, eds., Villgradter-Krey, eds. and 2an-
tovska-Suvin. For a more detailed bibliography that goes to 1974, see
Suvin (1976) in the present section. In view of the widespread Russian
publishing practice of using only the first initial of the author’s given
name, I have standardized this for all Russian items in this section.

Bel'chikov, N. Dostoevskil v protsesse petrashevtsev. Leningrad, 1936.


---------. " ‘Zolotoi vek’ v predstavlenii F. M, Do s toe vs ko go,” in V.
Kuleshov et al., eds., Problemy teorii i istorii literatury. Moscow, 1971.
BIBLIOGRAPHY 307

Brandis, E., and V. Dmitrevskii, Cherez gory vremeni. Moscow, 1963.


---------. “Problemy izobrazheniia budushchego v svete ieninskikh pred-
videnii,” in 0 literature dlia detei, issue 15. Leningrad, 1970.
Britikov, A. Russkii sovetskii nauchno-fanlasticheslai roman. Leningrad,
1970 (B by B. Liapunov).
-------- . [sub-chapters 111/14-16, IV/17 and Vil 1/8-9] in Russkii sovetskii
rasskaz. Leningrad, 1970.
Chistov, K. Russkie narodnye sotsial'no-utopieheskie legendy XVI-XIX vekov.
Moscow, 1967.
Collins, Christopher. Evgenij Zamjatin. The Hague, 1973.
Fedosov, I. Iz istorii russkoi obshchestvennoi mysli XVIII sloletiia: M. M.
Sheherbaton. Moscow, 1967.
Fiaker, .Aleksandar. Heretici i sartjari. Zagreb, 1958.
Gregg, R. A. “Two Adams and Eve in the Crystal Palace.” Slavic R. 24
(1965).
Grille, D. Lenins Rivale. Cologne, 1966.
Gromova, A. “Dvoinoi iik griadushchego,” in NF: Al’manakh nauchnoi
fanlastiki. Moscow, 1964.
---------. "Introduction: At the Frontier of the Present Age,” in C. G.
Bearne, ed. Vortex. London, 1971.
Jackson, Robert L. Dostoevskij’s Underground Man in Russian Literature.
The Hague, 1958.
Kizevetter, A. “Russkaia utopiia XVI11 st.,” in his Istoricheskie ocherki.
The Hague, 1967.
Liapunov, B. Aleksandr Beliaev. Moscow, 1967.
-------- . V mire fantasliki. Moscow, 1975 (B).
Maksimov, D. Briusov. Leningrad, 1969.
Miliavskii, B. Satiric i vremia. Moscow, 1963.
Mochulskii, K. Valerii Briusov. Paris, 1962.
Naumova, N. Roman N. G. Chemyshevskogo “Chto delat'I", Leningrad,
1972.
Nazirov, R. "Vladimir Odoevskü i Dostoevskii. ” Russkaia lit., No. 3
(1974).
Nudel’man, R. “Conversation in a Railway Compartment.” Science-
Fiction Studies 5 (1978).
---------. “Fantasdka, rozhdennaia revoliutsiei.” in Fantastika 1966, issue
3. Moscow, 1967.
Pil'nen’kii, Sergii. Kriz’ desiatilittia. Kiev, 1973.
Poliak, L. Aleksei Tolstoi—khudozhnik. Moscow, 1964.
Richards, D, J. Zamyatin. London, 1962.
Ripellino, Angelo Maria. Majakovski et le théâtre russe d’avant-garde.
Paris, 1965.
308 BIBLIOGRAPHY

Riurikov, lu. Cherez 100 i 1000 let. Moscow, 1961.


Rullkötter, Bernd. Die Wissenschaftliche Phantastik der Sowjetunion. Bern,
1974.
Rynin, N. Interplanetary Flight and Communication. Springfield, VA, 1972
(B).
Sakulin, P. Iz istorii russkogo idealizma: Kniaz P. F. Odoevskii. Moscow,
1913.
---------. Russkaia literatura i sotsializm, 1. Moscow, 1924.
Shane, Alex M. The Life and Works of Evgenij Zamjatin. Berkeley, 1968
(B).
Siniavskii, A. “Sovremennyi nauchno-fantasticheskii roman,” in Puli
razvitiia soiremennogo sovetskogo romana. Moscow, 1961.
Sipovskii. V. Ocherki iz istorii russkogo romana, I-11. St. Petersburg, 1909-
10.
---------. Etapy russkoi mysli. Petrograd, 1924.
Skaftymov, A. Nravstvennye iskaniia russkikh pisatelei. Moscow, 1972.
Smelkov, lu. Fantastika—ochemona*. Moscow7, 1974.
Suvin, Darko. Russian Science Fiction 1956-1974. Elizabethtown NY,
1976 (B).
Sviatiovskii, V. RusskH utopicheskii roman. Petersburg, 1922 (B).
Trotsky, Leon. Literature and Revolution. Ann Arbor, 1966.
Veksler, N. Aleksei Nikolaevich Tolstoi. Moscow, 1948.
Vorob’ev, B. “Nauchnaia fantastika v trudakh K. E. Tsiolkovskogo," in
K. Tsiolkovskii. Put' k zvezdam. Moscow, 1960 (French in C. Tsiol-
kowski, Le Chemin des étoiles. Moscow, 1963).
Voronskii, A. “Evgeny Zamyatin.” Russian Lit. Triquarierly, No. 2 (1972).

VII. Karel Caper and His SF


See also Hillegas, Philmus, and Smail in Bibliography I, III A, and
IV A.
Bernshtein, I. A. Karel Chapek. Moscow, 1969.
Burianek, Frantiiek. “Predmluva," in Karel Capek, Valka s mloky.
Prague, 1955.
Harkins, William E. Karel Capek. New York, 1962.
Janaszek-Ivaniêkovâ, Halina. Karol Capek. Warsaw', 1962.
Klima, Ivan. Karel Capek. Prague, 1962.
Malevic, Oleg. “Tovarna na Absolutno po 40 letech . .. . " Ceska lit., No.
3 (1962).
Malevich, O. Karel Chapek. Moscow, 1968.
Matushka, Alexander. Karel Capek. Prague and London, 1964.
Mukafovsky, Jan. Kapitoly z ieske poetiky, II. Prague, 1948.
BIBLIOGRAPHY 309

---------. "O Karlu Capkovi i jeho Krakatitu,” in Karel Capek, Krakatil.


Prague, 1948.
Nikol’skii, S. Roman K. Chapeka “Vaina s salamandrami”. Moscow, 1968.
---------. Karel Chapek—faniast i satirih. Moscow, 1973.
Piäa, A. M. Sm¿ry a eile. Prague, 1927.
Strohsová, Eva. “Román pro sluzky a Capkovo smêfovani k epicnosti,”
in Struktura a smysl hterámího dita. Prague, 1966.
Wellek, René. “Karel Öapek,” in his Essays on Czech Literature. The
Hague, 1963.
Index

Only proper names of non-imaginary persons have been retained: also excluded are
names in prefatory matter and Bibliography, names cited only in notes referring to the
Bibliography, names in titles of literary works, and names of editors of works by other
authors.

Abbott, Edwin A.. 29, 167 Bellamv, Edward, 54.76. 87, 120. 144, 165.
Adamov, Grigory B., 264-65 169, 171-80, 182-84, 188-94, 198-201,
Adams. Richard Locke, 142 203. 218, 220, 247, 251, 267, 279
Adorno, Theodor W., 82n Belyaev, Alexander R., 260, 263
Aeschylus. 87 Benoit. Pierre. 261
Aesop, 215, 279 Berger, Harry , Jr., 52n
Aldiss, Brian W., 242 Bergonzi, Bernard, 222n, 223n
Allende, Salvador, 75 Bergson, Henri, 81 and n
Andersen. Hans Christian. 183 Berneri. Marie Louise, 41 and n, 60
Angenot. Marc. 147 and n, 163n, 206» Bidney, David, 35 and n
Antonius Diogenes, 87 Bierce, Ambrose, 203-04
Apollinaire. Guillaume. I6J Blake, William. 74, 1)3, 115, 117-19.
Appleton, Jane S.. 177 121-25. ¡34-35, 144. 183-84. 197,229
Ac]bar (emperor). 38 Blavatsky, Helena P., 202
Arelsky, N., 261 Blish, James. 22, 68. 143, 279
Aristophanes. 55, 87. 94, 99 Bloch. Ernst. In. 12, 39 and tt, 42 and n,
Aristotle, 18, 48, 82, 242n 52n, 54n, 57n, 58 , 61-62, 64 and n, 81-
Arkwright, Richard, 194 82n. 84, 98n. 148n, I92n
Asevev, Nikolai N.. 253 Blök. Alexander A„ 251,256
Asimov, Isaac, 13, 23, 29, 82, 120, 136, Bobrov, Sergei P., 254
242.279 Boccaccio. Giovanni, 18
Astor, John Jacob. 203 Bogdanov (pseud, of Malinovsky), Alexan­
Athanasius of Alexandria, 45n der A.. 252-53, 257, 260-61
Attila, 256 Bogdanov, Fyodor M., 262
Augustine of Hippo. 21,93 Boguslaw, Robert. 60n
Bonner, Anthony, 153
Babeuf, François Noël "Gracchus", 118-19 Borges, J. L.. 29^-30. 59. 65n, 220
Bacon. Francis, 23, 45, 65 , 96, 100-02, Bougainville. Louis Antoine de. 59
104, 108-10. 115. 129, 194. 197, 259 Bradbury, Ray. 68, 141
Baird, Theodore, 224n Bradshaw, William R„ 165
Bakhmetyev, Vladimir M., 251 Braine, Robert D.. 203
Bakhtin, Mikhail, 54n, 56«, 79» Bray, John Francis. 138
Ball, John, 94 Brecht, Bertolt, 6 and n, 7 and n, 18, 54n,
Ballard, J. G., 67, 242 71, 74, 80n. 82n.278
Balzac, Honoré de, 69, 161 Brown, Charles Brockden, 139
Barlow, Joel. 139 Brunner, John, 27, 280
Barnum. P. T., 197 Bruno, Giordano, 98
Barthes. Roland. 18, 43 and n. 50-5In, Brutus, Marcus I unius, 110, 113
54n,61n.147 Brvusov, Valery Ya., 250-53,256, 260
Baudelaire. Charles. 168 Buber, Martin, 39 and n
Beaumarchais, Pierre Caron de, 114 Bulgakov, Mikhail A.. 254
Bell. Alexander Graham. 194 Bulgarin. Thaddeus, 244-45, 247
311
312 INDEX

Bulwer-Lvtlon. Edward, 69, 164-65, 169, Corelli, Marie. 69. 202


188,202, 220 Cornford, F. M., 25 and n
Bunyan, John. 217 Cromie. Robert. 203
Burke, Kenneth, 51 and n Culler, A. Dwight. 225»
Burne-Jones, Georgiana, )88n Cuvier, Georges. 150
Burroughs, Edgar Rice. 29. 116. 162, 178. Cyrano "de Bergerac," Savinien. 5, 10, 12,
205, 260-61, 282 25, 88, 98. 100». 103-08, Il I. 114, 116,
Buder, Samuel, 5, 165-66, 169, 257 127, 162. 171, 202
Butor. Michel, 147. 154n
Byron. George Gordon, Lord. 128. 137- Däniken. Erich von, 83
38. 150, 153. 168 Dante Alighieri. 33. 55», 57. 93. 125, 132,
248
Gäbet. Etienne. 138-39. 175, 177,248 Darwin. Charles, 30, 36, 149, 164, 166.
Cade, Jack, 195. 197 209-17, 223-35, 237-39, 241
Campanella, Tommaso. 96. 100 and n. Defoe. Daniel, 111. 113
102-03, 259 Defontenay, C. !.. 121, 138, 202
Camus. Alben. 251 Déjacque. Joseph, 138
Canning. George. 135» Delany. Samuel R., 29, 35, 52», 74. 121
Cannon. Walter F., 224» De Mille, James, 165
Capek, Josef, 270. 272. 282 Descartes, René (Cartestus), 65. 151
¿apck, Karel. 29, 30». 74, 137, 229, Dick, Philip K., 13, 30», 201-02
270-83 Dickens. Charles. 175, 197, 208, 239»
Carlyle, Thomas, 224» Diderot. Denis. 30. 59, 88, 114. 123, 131,
Carnegie, Andrew. 225». 229 244
Cassirer, Ernst, 31-32», 34-35» Dimitrov, Georgi, 247
Castro, Fidel, 38 Dionysius of Syracuse. 28
Catherine 11 (empress), 244 Disch, Thomas M„ 74
Caudwell, Christopher (pseud, of Christ­ Disraeli. Benjamin. 171
opher St. J. Sprigg), 212», 239-40» D'lvoi, Paul. 162
Cervantes, Miguel de, 248 Dixie. Florence, 203
Chaucer, Geoffrey, 184 Dmitriev-Mamonov, Fyodor !.. 244
Chekhov. Anton P., 243, 250 Dneprov, A. (pseud, of Anatoly P.
Chernyshevsky, Nikolai G,, 102, 164 , 247- Mitskevitch), 230. 262
49, 252-53, 257,266, 268 Dolgushin, Yury A., 265
Cherry, Colin. 75» Dominik. Hans. 162
Chesney, George T,, 166 Doni, Anton Francesco, 100
Chesterton, G. K., 261 Donnelly, Ignatius, 178
Chikolev, V„ 251 Dorémicux. Alain. 239»
Christopher. John (pseud, of C. S. Youd), 9 Dorson, Richard M.. 31»
Cioranescu. Alexandre, 47-48. 56-57», Dos Passos, John. 280
58» Dostoevsky, Fyodor M., 20, 164, 243,
Clarke. Arthur C-. 122 248-50, 256-57, 259. 274
Cobden, Richard. 146 and n Douglas. David. 207»
Coleridge, Samuel T.. 124, 133 Dovzhenko, Alexander P., 264
Collins. Wilkie. 70» Doyle. Arthur Conan, 162, 167», 279
Columbus, Christopher, 6, 22, 59 Dubfek, Alexander, 75
Condorcet. Antoine Nicolas de. 117-19, Dubois, Claude-Gilbert, 52»
123,128 Dumas. Alexandre (père), 29, 148
Conrad, Joseph, 227». 276
Cooper, J. Fenimore, 139. 148, 156, 162, Edison, Thomas Alva. 168
193 Edwards, Bryan, 135»
Copernicus. Nicolaus, 103 Ehrenburg. Ilya, 253, 261
INDEX 3J3

Einstein, Albert, 14, 68, 74, 255, 269 135. 139, 171
Eiseley, Loren. 225n Gogol, Nikolai V., 8,242
Eisenstein. Sergei M., 74 Goncharov, Viktor A.. 261
Eliade, Mircea, 58 and n Gorky, Maxim (pseud, of Alexei M. Pyesh-
Eliot. T.S.. 14. 141 and n. 220 kov). 17, 250
Elliott. Robert C.. 50 and n, 52n, 54n. 55n, Grafftgny, Henri de. 162
57n. 61n, 108n.24 In Grainville, J. B. Cousin de, 137, 202
Elisbcrg. Daniel. 239n Gramsci, Antonio, 82n. 206n
Emerson, Ralph Waldo, 191n Grave, Sergei L„ 261
Emin. Fyodor, 244 Grebnev. Grigory N., 264
Enfantin. Barthélémy Prosper, 157n Griffith. George (pseud, of George C. G.
Engels, Friedrich. 23,39. 112n, 148n. 199n Jones). 162
Epicure, 106, 181 Griffith, Mary, 176-77
Erasmus of Rotterdam, 103 Grigoryev, Sergei T„ 262
Erlich, Victor. 6n Grimm. Jakob and Wilhelm. 183
Euclid. 52 Gutenberg, Johann. 194
Euhemcrus, 87, 94
Haeckel. Ernst. 232
Falke. Rita, 4 0n Haggard. Henry Rider, 141, 165
Farmer, Philip José. 239n Haldane. J. B. S„ 226n
Faulkner. William. 26 Haie. Edward Everett, 162, 177
Fawcett, E. D-. 162 Harbou, Thea von, 218
Fénelon, François de Salignac de la Mothe, Hardy, Thomas, 1 In, 70n
45.88. 111. 177, 244 Harrington, James. 111
Flammarion, Camille. 137, 178, 202-03. Harrison. Jane Ellen. 25n
220,251-53 Harun-al-Rashid. 195
Flaubert, Gustave, 269 HaSek. Jaroslav. 282
Fleming, Donald, 224n Hawthorne, Nathaniel, 139-40, 143, 162,
Foht, Ivan, 66n 176-77
Foignv, Gabriel de, 88. 107 Hay, W. D., 177
Foucault, Michel, 13 In Hecataeus of Abdera, 87
Fourier, Charles, 50-51, 54 . 59. 74 . 95, Hegel. Georg W. F„ 21. 78, 80
102, 119-21, 124, 138, 140, 165, 175, Heinlein. Robert A., 13, 22, 78, 143, 193.
187. 192 and n, 202, 248 201,218,242, 279
France. Anatole (pseud, of Jacques- Henkin, 1-eoJ., 225n
Anatole Thibault}, 30, 138, 279 Henningsen, Charles, 138
Freud, Sigmund, 134 Herbrüggen, Hubertus Schulte, 40n, 46
Friedrich II (emperor), 38 and n, 47 and n
Frye. Northrop, 12, 16 and n. 18. 26 and Heruler, Joyce Oramel, 41-42
n, 31-35, 46-47n, 49 and n. 51 and n, Hesiod, 187
54n, 55n, 60n, 147n Hillegas, Mark R., 147 and n, 222n
Hirschhorn (Girshgom), V., 262
Galileo Galilei. 6. 9, 103-04 Hitler, Adolf, 276-77, 281
Geoffroy, Louis, 138, 166 Hoffmann, E. T. A.. 131, 168, 245, 282
Gerber, Richard. 43n, 49-50n Holberg. Ludvig, 114, 165. 244
Gernsback, Hugo. 23, 65, 163, 204-05. Holquist, J. Michael, Sn, 52n
267,282 Homer, 25, 92, 97, 145, 153, 187, 275
Glebov, Anatoly G., 262 Horrell. Joseph, 242n
Godwin. Francis, 88, lOOn, 103 Howells, W. D.. 172n, 177-178, 197 and n
Godwin, William, 123, 127, 129-30, 134, Hudson, W. H., 187-88, 193
136, 273 Hugo, Victor. 153
Goethe, Johann Wolfgang von. 117. 130, Huxley, Aldous. 83. 279
314 INDEX

Huxley, Julian. 226n Lang, Herrmann (pseud.), 166


Huxley, Thomas H.. 23, 209, 222-31,233. Langer, Susanne, 32
238-39n. 242n Langland, William, 184, 248
Hyman, Stanley Edgar. 26n. 31n, 224n Larri. Yan, 253
Lasswilz, Kurd. 252. 261
Iambulus, 87, 94-95, 100 Laurie. André (pseud, of Paschal Grous-
Ibsen. Henrik, I In set), 162
Irving. Washington, 139. 177 Lavrenyev, Boris A.. 254, 261
Isaiah. 117-18 Lawrence. D. H., 127
Ilin, Vivian A., 253 1 Lear, Edward, 215
Ivanov. Vsevolod V.. 254 Le Guin, Ursula K„ 29, 30n. 68, 74, 79,
Ivan the Terrible (emperor), 243 121, 193, 199-200,280
Leinster, Murray (pseud, of Will F. Jen­
James, Henry, 49 and n kins). 268
Jameson. Fredric R.. 76n Lem. Stanislaw, 3n, 29, 30n, 36, 67n, 74,
Jauss. Hans Robert. 80n 83n. 202, 230, 269, 280
Jefferies, Richard. 187—88. 193 Lenin. Vladimir Ilyich (V. 1. Ulyanov), 38,
Jefferson, Thomas, 38, 175 158n, 219, 247, 253, 260
Joachim Di Fiore. 118 Leonov, Leonid M.. 264
Jokai, Maurus, 162 Le Rouge. Gustave, 162
Joyce, James, 280 Levin. Harry, 26n
Lévi-Strauss, Claude, 59, 233
Kafka, Franz. 26, 29-30. 65n, 282 Levshin. Vassilv A., 244
Kaiser, Georg. 282 Lewis. C. S„ 26,69, 138, 178. I89n, 193
Kam, Immanuel, 60, 83, 202 I^wis, Sinclair, 280
Karpov, Nikolai A„ 262 Lincoln. Abraham, 175
Katayev, Valentin P., 253-54 Linnaeus (Linné), Carl von, 227
Kaverin. Veniamin A.. 254 Ussovoy, P., 264
Kazantsev. Alexander P., 264-65 London. Jack. 10. 25. 27, 178, 185, 193,
Keller, I.. 262 199, 207 and n, 230, 240n, 250, 255,
Kent, Edward, 177 262,276
Kepler, Johannes. 88. lOOn, 103,219 Longfellow, Henrv Wadsworth, 146
Keynes, John Maynard, 219 Lotman. Jurij. 70n, 74 and n
Kheraskov, Mikhail M., 244 Lovecraft. H. P„ 8, 68. 141
Kipling. Rudyard, 214. 231,276 Lowell. Percival. 260
Klein. Gérard, ¡6 hi Lucian of Samosata. 5, 10, 12, 25, 54, 74.
Knight. Damon. 36, 143 87-88,97-98, 107. 116, 211,219
Komarov. N. S., 251 Lucretius Carus, Titus, 126
Kornbluth, Cyril M., 27, 78 Lukács, György (Georg), 14, 73n,222
Kraft, Robert. 162 Lunacharsky. Anatoly V., 262
Krleïa, Miroslav, 127 and n Lunts, Lev N., 254
Krysmanski. Hans-Jürgen, 47-48, 60 Lvov. Pavel L, 244
Kryzhanovskaia, Vera I., 251 Lyell, Charles, 227
Kubrick, Stanlev, 76
Kuhn, Thomas S., 222 Mably, Gabriel Bonnot de, 118, 123
Kuprin. Alexander I„ 251 Macaulay, Thomas Babington, 42 and n
Macnie, John. 177
Lafargue. Paul, 262 MacOrlan, Pierre, 279
Lagin. 1-azar I.. 262 Maitland. Edward, 177
Lalande. André. 43-44n, 52n. 54n Malory , Thomas, 195, 197
Lamartine, Alphonse de. 138 Malthus. Thomas, 229 and n
Lane, Mary Bradley, 165 Mandeville, John, 94
INDEX 315

Mannheim, Kari, 39 and n. 48 Nemtsov, Vladimir I., 265


Mao Tse-tung, 38 Nero (emperor). 121
Marivaux, Pierre de, 113 Newton, Isaac. 87, 109, 122, 210
Marvell. Andrew, 248 Nietzsche, Friedrich, 22, 229, 272, 275
Marx. Karl, 21, 39, 68. 73n, 74. 77n, 112n, Nikolsky. V'. D., 253, 264
118-19, 121. 175. 199n, 230, 247, 256, Nudel’man, Rafail. 67n. 68n. 79 and n
266
Masaryk, Thomas G., 281 O'Brien, Fitz-James, 144, 201
Mayakovsky, Vladimir V.. 127 and n, Obruchev. Vassily A.. 263
254-55, 258, 260-61,264, 266 Odoevsky. Vladimir F., 137, 245-47
Melville. Herman. 137. 139. 143-44, 170 Okunev, Yakov M.. 253
Mercier. Louis-Sébastien, 27, 118. 177, Olesha, Yury K., 264
179, 245 Ong, Walter J., 242n
Merritt, Abe, 68 Orlovsky (pseud, of Grushvitsky), Vladimir
Meyerhold, Vsevolod E., 255 E.. 262
Meyrink. Gustav. 282 Orwell, George. 76, 83, H2n, 129, 242,
Micah, 118 256
Michelet. Jules, 153 Owen, Robert, 165, 248
Miller. George Noyes. 203
Miller, Walter, Jr., 27 Paley, Abram R.. 254
Milton, John. 33, 70. 122, 132.134.217 Paltock. Robert. 88. 127, 165
Mitchell. Edward Page. 167 and n Pascal, Blaise. 106
Moliere (pseud, of Jean-Baptiste Poquelin), Pasternak, Boris, 253
114 Patrick. J. Max. 46-47n, 54n
Montaigne, Michel Eyquem de, 7, 99. 114 Pavlenko. Pyotr A.. 265
Montesquieu. Charles de Secondât. 114, Peirce, Charles Sanders, 77 and n
245 Pemberton. Max, 162
More, Thomas. 10, 12-13, 37,40-43, 48, Perelman. Ya. 1„ 264
50-54 , 58, 60, 73-74, 83 . 88. 90-103, Peresvetoy, Ivan S., 243
IO6n, 107, 110-11, 113, 115-116, 123, Pericles, 38
131, 165, 171, 179n, 180, 213, 222, 230, Peter the Great (emperor), 245
235 and n. 237, 242 and n. 244 4 5. 249. Petrarca. Francesco (Petrarch). 43
257, 267 Philmus. Robert M., 65 and n, 67n, 76n,
Morgan, Arthur E., 177n 80, 222n, 232n
Morgan, Joseph, 139 Picasso, Pablo. 74
Morris. William. 9. 38. 45, 50-51, 54, 76. Plato. 28. 37 and n. 39, 42. 45. 83 . 87-88,
95. 113, 169. 171, 175, 177-94, 198-99, 93-94 , 97, 101-02, 111, 118, 123, 166,
201, 203, 212, 217, 219-20, 230, 235- 168, 176. 220, 229-30, 246, 257, 267
237,240n, 266 Poe, Edgar Allan, 22, 36. 82, 124, 132n,
Morse, Samuel, 194 137, 139-43, 148-49, 162, 165, 168,
Morton. John (cardinal), 29 201,204. 220. 251
Mukhanov, N. I., 261 Pohl. Frederik. 13. 27, 78, 218, 242
Mullen, Richard Dale. 239n Polo. Marco. 243
Mumford. Lewis, 12. 73n, 93n, 13In, 141n, Porphyrius, 242n
IBOn, 18ln Prévost, Antoine-François (abbé), 49
Münzer, Thomas. 38 Proust, Marcel, 49
Murray, Gilbert, 25 and n Ptolemy (Claudius Ptolemaeus), 6, 87
Mussorgsky, Modest P., 243 Pushkin. Alexander S.> 243, 245

Napoleon 1 Bonaparte, 119, 150, 153, 245, Rabelais, François. 10, 12. 57. 88, 96-98,
273 101-02, 106-07, 121,127,143,153, 249,
Negley, Glenn, 46-47n, 54a 259
316 INDEX

Radishchev, Alexander N., 244 Smith. E. E., 122


Renard. Maurice, 251-52, 263 Solovyev, Vladimir A.. 251
Renouvier. Charles, 138 Sombart. Werner, 73«
Reslif “de la Bretonne", Nicolas-Edmé, 9, Souvestre, Emile, 138
114. 138 Spence, Thomas. 122
Richards, G. C„ 90 Spencer, Herbert. 223, 225n, 229
Richardson, Samuel. 49 Spengler. Oswald, 29. 279
Riurikov, Yury B„ 264 Stalin, Joseph (Yosif V. Djugashvili). 9,
Robida, Albert, 162, 167n 119. 243, 253, 264-65, 267
Rockefeller, John D„ 225n. 229 Stapledon. W. Olaf. 23. 27-28, 60. 78,
Rodnvkh, .Alexander A.. 251 121-22. 138. 201-02. 242
Rogers, B. B„ 94 Stauffer. Donald A.. 31«
Roosevelt. Theodore, 240n Stendhal (pseud, of Henri Beyle), 20
Rosnv Aîné (pseud, of Joseph-Henri Stephenson. George, 194
Boêx). 30. 207, 239n Stevenson. Robert Louis, 69
Rostand. Edmond. 107n Stockton. Frank. 203-04
Rousseau. Jean-Jacques. 114, 128.249 Strachan, Geoffrey, 103, 106
Ruskin, John, 217 Strugatsky. Arkady N. and Boris N., 199,
Russ. Joanna, 67n 230, 269
Ruyer, Raymond. 42-44n. 52n, 55n, 264 Swedenborg. Emanuel, 70
Rynin, Nikolai A., 264 Swift, Jonathan. 5-6. 10, 12, 25.30. 59-60,
73-74,88.98-99, 103, 105, 107-14. 120,
Sade, Donatien Alphonse François. Mar­ 123, 126, 130-31, 137-38, 148, 162. 164,
quis de. 114 166, 174, 196, 201, 211. 213-15, 219,
Saint-Simon. Henri de, 22, 45, 59, 101, 236«. 241-42,i, 275
119-20. 148 indu, 157-58, 175 Symmes, John Cleves (putdtive author of
Salgari. Emilio. 162 SynuKonia), 132«, 142
Sand. George (pseud, of Aurore Dupin),
153,248 Taylor, Frederick W., 254, 256, 273
Savchenko. Vladimir 1., 262 Tennyson. Alfred, 145-46. 163, 169, 183,
Scheerbart, Paul, 207 193,195
Schelling. Friedrich Wilhelm. 245 Thackeray, William Makepeace, 248
Scott. Walter, 148, 193,269 Thales, 162
Semenov, V., 252 Theopompus, 87
Shaginyan, Marietta, 253 Thom, Rene, 75n
Shakespeare, William, lin. 18, 98-99, 174 Thomson. David, 146n
Sharapov, S., 251 Thomson. George. 25«
Shaw, G. B„ 193 Thoreau, Henry David, 173, 180
Shcherbatov, Mikhail ,M„ 244 Tillich. Paul. 39 and n
Sheckley, Robert, 200 Tolstoy. Alexei N„ 193, 252, 254, 260-6)
Shefner. Vadim S„ 269 Tolstoy, Leo (Lev N.). 243. 245. 250,
Shellev, Man W„ 12. 113, 115, 117, 124, 269,272
127-39. 141, 143, 166, 168-70, 198, 220, Toman, Nikolai V„ 262
246 Traugott. John, lln
Sheilev, Percy Bysshe. 113, 115, 117-18, Tredyakovsky, Vassily K„ 244
121-30, 132-36. 138, 144, 153, 168, 171, Trollope, Anthony, 166
183, 229, 259 Trzynadlowski. Jan,, 67«
Shishko, Anatoly V., 254, 261 Tsiolkovsky, Konstantin E., 251-52, 257,
Shklovsky, Viktor B., 6 and n. 54n, 71,254 259. 263
Simak. Clifford D„ 9 Tucker, George, 142
Simmel, Georg, 73n Turov, B., 261
INDEX 317

Twain, Mark (pseud, of Samuel L. Cle­ Watson. H. C. M„ 177


mens), 30,88, 171, 193-203, 220 Watt. James. 194
Tyler, Wat. 195, 197 Wellek. René, 33n
Welles. Orson, 214
Uminsky, 252 Wells. H. G., 5. 9-10, 12. 23.25. 27-28, 30
Uspensky, L. V., 264 and n, 36, 45, 49n, 54, 70n, 71-72, 76n,
79, 87, 92, 95, 103, 106n, 116, 120. 130,
Vaihinger, Hans. 52n 134-37, 139, 143, 162-63, 165-67, 167n,
Vairasse d’Allais, Denis. 88, 107 171, 177-78, 185, 193, 199-204, 206-42,
Valéry, Paul. 159 245, 247, 251-57, 259-61, 263-64, 266,
Van Vogt. A. E., 22. 26. 79. 122, 279 272, 279-80, 282
Varshavsky, Ilya I., 262. 269 Wheelwright, Philip, 32n
Vercors (pseud, of Jean Bruller), 239n Whitney, Eli, 194
Verne, Jules, 5, 9-10, 12. 22-23, 35. 59. Williams, Basil, 209n
65, 72, 73n, 87-88, 116-17. 120, 123, Williams, Raymond, 77 and n, 186, 206n
132n, 143, 147-65, 167-71, 198, 203, Wise. P. C.. 177
209-1 In, 217, 220, 236n, 239, 244, Wollstonecraft. Mary. 126-27
251-52, 254,260-61,263-64 Wordsworth. William, 123, 144
Verne, Michel, 162
Vespucci, Amerigo. 95 Xenophon.153
Villiers de i'Isle Adam, Auguste, 162,
167-71, 239n
Villon, François. 97 Yaroslavski, Alexander B., 262
Virgil. 87. 118 Yazvitsky, Valery I., 261
Voigt, Andreas. 41 and n Yefremov. Ivan A„ 9, 27, 35.78. 252,261,
Volokhov (pseud, of Pervukhin), V., 252 265-69
Voltaire (pseud, of François Marie de
Arouet), 30, 114, 139, 244. 275 Zamyatin, Evgeny, 10, 13 , 25-27 , 54 . 74,
Vonnegut. Kurt, Jr.. 200, 242 185, 202, 218, 230, 254-61, 264, 280
Zelazny. Roger, 26
Warren, Austin, 33n Zelikovich. Emmanuil S., 253
Watlock, W. A., 203 Zuyev-Ordynets, M. E., 261

You might also like