100% found this document useful (1 vote)
39 views292 pages

MORRIS, Gay (2006) - A Game For Dancers - Performing Modernism in The Postwar Years, 1945-1960. Connecticut, Wesleyan University Press

Uploaded by

susana tambutti
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
100% found this document useful (1 vote)
39 views292 pages

MORRIS, Gay (2006) - A Game For Dancers - Performing Modernism in The Postwar Years, 1945-1960. Connecticut, Wesleyan University Press

Uploaded by

susana tambutti
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/ 292

3

A Game for Dancers


A GAME FOR
DANCERS
Performing Modernism in
the Postwar Years, 1945-1960

GAY MORRIS

WESLEYAN UNIVERSITY PRESS


Middletown, Connecticut
Published by Wesleyan University Press, Middletown, CT 06459
www.wesleyan.edu/wespress
© 2006 by Gay Morris

All rights reserved

Printed in the United States of America


54321

Text by George Balanchine used by permission of The George Balanchine Trust.


Balanchine is a trademark of The George Balanchine Trust.

Library of Congress Cataloging-in-Publication Data


Morris, Gay, 1940-
A game for dancers : performing modernism in the postwar years, 1945-1960 /
Gay Morris,
p. cm.
Includes hibliographical references and index.
ISBN-13; 978-0-8195-6804-5 (cloth ; alk. paper)
iSBN-io: 0-8195-6804-X (cloth : alk. paper)
ISBN-13: 978-0-8195-6805-2 (pbk. : alk. paper)
iSBN-io: 0-8195-6805-8 (pbk. : alk. paper)
I. Modern dance—History. I. Title.
GV1619.M67 2006
792.8o9'o45—dc22 2006000181
For G. G.
1 ^'’>3^. 4

»^ 4. ■ * • rf - 1-* ^

\
’'S- <*• :%•^.^
bi •:

.#<liil»%:ti

*»•; j ^Vi^.,‘'^>* •' ’- r>^(i Mi


■'• 1’'. * jj ^

■i
■<••,• -w ^ ■- ' •■5»-

't =-»•
■> >-i ) i
V'
V;% .
-trr

■# ,.'■ '» ft. •. '>VV'


r.^'
■ 1
^ ■ J. .
‘ SP''
f. 1■
*■. ' ' -i,
u

f
a.

■,-; .w.,
I - . .^'*
A
■W*1Fd

.S''!' • ■
*'J-^

. J '• 3 V .
si'-
•V .> y'SL.i9BSiiai

t~
Contents

List of Illustrations ix
Acknowledgmmts xi
Introduction xiii

Chapter i
The Trouble with Modern Dance I

Chapter 2
Ballet’s Challenge 38

Chapter 5
Modernist Theory: John Martin, Edwin Denby, John Cage 64

Chapter 4
Embodying Community 87

Chapter 5
Airican-American Vanguardism: 1940s 114

Chapter 6
African-American Vanguardism: 1950s 147

Chapter 7
Objectivism’s Consonance 166

Notes 205
References 217
Index 243
riTifil rKD'ji.

4»*
•’/rjjjoy'' ■"• -5
I >,T 4»'»
./ .-.;V
■ .A‘‘ \ r<V(vO 'i
rf?»v iii'r* i

*.

.'■•t* ; •iJirL • ^ THfij 6"'^'xi

’if!
H.,
■» ^
i
.■ J■ i < ,4 njt- tfl' w t » ■.' I''!.
i ,

-^: '■ ■f ' '■’ • ]


.'**- ‘'s
JK'v v" 2
■j, V’..rsfI ^ ’i
^ * ,■
«-•
.**. ■ < r*idr?»»HK>
' ■ -t ■?' %, A'; -"
•irf',.
ii^
-.V/T

'*:->■■-. ’ .... * -■ . .' V « X’^ovu.


«-
^ ,. ./ .. ;i
■ ;-S

■ «« - », »

'll ; ^
Illustrations

“The Theater of Martha Graham” 14

Martha Graham in Frontier (1935) 22


Martha Graham as Jocasta in Night Journey (1947) 23

Valerie Bettis and Dimcan Noble in Bettis’s Daisy Lee (1944) 26


Lucas Roving, Jose Limon, and Pauline Koner in Limon’s
The Moor's Pavane 29
Lincoln Kirstein and George Balanchine in the 1950s 39
The Phlegmatic variation with Todd Bolender and corps de
ballet in the original Ballet Society production of George
Balanchine’s The Four Temperaments (1946) 46
The Melancholic variation from George Balanchine’s The
Four Temperaments 48
Mary Ellen Moylan and Fred Danieli in the Sanguinic
variation in the original Ballet Society production of
George Balanchine’s The Four Temperaments (1946) 50
Jason Fowler and Jennifer Tinsley in the third theme of
George Balanchine’s The Four Temperaments 55

John Martin 66

Edwin Denby 75
John Cage 79
Jane Dudley, Sophie Maslow, and Ronne Aul in Maslow’s
The Village I Knew (1950) 91

Anna Sokolow in Kaddish (1945) 95


Martha Graham in Lamentation (1930) 96
X Illustrations

The Juilliard Dance Ensemble in the opening dance from a


1977 production of Anna Sokolow’s Rooms (1955) 99
The Juilliard Dance Ensemble in a 1977 production of
Anna Sokolow’s Rooms (1955) 104
Lynne Eippinger and men of the Juilliard Dance Ensemble
in Anna Sokolow’s Opus ’6^ (1963) 111
Katherine Dunham as the Woman with the Cigar in Tropics—
Shore Excursion (1940) 118
Lucille Ellis and James Alexander in the “Eertility Ritual”
section of Katherine Dunham’s Rites de Passage (1943) 124
Pearl Primus in Hard Times Blues (1943) 133
Talley Beatty as a young dancer in New York 138
Kim Bears-Bailey and members of the Philadelphia Dance
Company in Talley Beatty’s Southerm Landscape (1948),
which Beatty revived for the company in 1992 141
An early cast of the men in Donald McKayle’s Rainbow Round
My Shoulder 154

Donald McKayle and Shelley Frankel in a 1962 photograph


of McKayle’s Rainbow Round My Shoulder (1959) 156
Enid Britten and Ulysses Dove as the central couple in an
Alvin Ailey American Dance Theater production from
1970 of Talley Beatty’s The Road of the Phoebe Snow (1959) 162
Jo Anne Melsher, Merce Cunningham, Carolyn Brown, and
Viola Farber rehearsing Cunningham’s Septet (1953) 169
Viola Farber and Carolyn Brown in Merce Cunningham’s
Summerspace (1958) 179

Alwin Nikolais speaking to students at the Henry Street


Playhouse in the early 1950s 185

Murray Louis, Gladys Bailin, and Coral Martindale in


“Straps” from Alwin Nikolais’s Kaleidoscope (1956) 191

Murray Louis, Phyllis Lamhut, Gladys Bailin, Dorothy


Vislocky, Beverly Schmidt, Bill Frank, and Coral
Martindale in “Discs” from Alwin Nikolais’s Kaleidoscope
(1956) 193
Acknowledgments

I n a research project that spans many years, the number of people


who contribute to it grows accordingly. This project lingered in the
back of my mind for most of a career that has involved both art and
dance criticism. It began to come to fruition with a dissertation in the
Sociology Department of Goldsmiths College, University of London,
that was overseen by Helen Thomas. During those happy years I often
lived in a flat, which overlooked an eighteenth-century square, that
was loaned to me by Sue Merrett, my editor at the Dancing Times. It is
safe to say there would be no book today without the aid of these two
remarkable women. Helen was a mentor and became a close friend; Sue
is now retired and floats about the canals and rivers of England on her
barge. Other colleagues in the United Kingdom who have given advice
and encouragement over the years include Stephanie Jordan, Stacey
Prickett, and especially Dee Reynolds, who read parts of the book man¬
uscript as it developed.
In the United States I must particularly thank Marcia Siegel and
Mark Franko, who have each in important ways helped sort out my
thoughts over the course of the project and have been friends as well as
colleagues. Others who have helped with information and advice in¬
clude Claudia Gitelman, Dawn Lille, Murray Louis, Ze’eva Cohen, Joe
Nash, George Dorris, Jack Anderson, Doris Hering, Donald McKayle,
Susan Buirge, Phyllis Lamhut, Janet Soares, Deborah Jowitt, Lynn
Garafola, Joan Brown, Delores Brown, David Vaughan, Naima Prevots,
George Jackson, Elizabeth Zimmer, Don McDonagh, and Leslie Getz.
The staffs of research libraries and archives are the unsung heroes of
scholarship, and my work was aided immeasurably by them. This was
especially the case with all the people at the Dance Collection of the
Library for the Performing Arts and the Schomburg Center for Research
xii Acknowledgments

in Black Culture, both of which are research arms of the New York
Public Library, where much of my research was done. Phil Karg at the
Dance Collection was particularly helpful with all aspects of locating
and obtaining photographs. I also must thank David Vaughan (this time
in his role as archivist) and Stacy Sumpman at the Cunningham Foun¬
dation Archives; Stephen McLeod at the University of California-Irvine
Library where Donald McKayle’s archives are housed; Steve Siegel,
archivist at the 92nd Street Y; Beth Olsen and Robert Tracy at the Alvin
Ailey Dance Theater; and the staff of the New York City Ballet, its
archives, and the Balanchine Trust. No scholar could do her work
without these dedicated and generous people.
I would like to thank my editor, Suzanna Tamminen, production
editors Jessica Stevens and Elizabeth Rawitsch, marketing manager
Leslie Starr, and the staff at Wesleyan University Press for their un¬
stinting support throughout the course of this book. A special thanks,
too, to John Bennett, computer guru extraordinaire.
Three chapters in A Game for Dancers include material from articles
that appeared elsewhere; chapter i includes sections from “Bourdieu,
the Body, and Graham’s Postwar Dance” that appeared in Dance Re¬
search 19:2 (Winter 2001); chapter 2 includes substantial material from
“Balanchine’s Bodies” that was included in Body and Society (December
2005); chapter 3 is an enlarged and revised version of “Modernism’s
Role in the Theory of John Martin and Edwin Denby” that appeared in
Dance Research 22:2 (Winter 2004). I thank these journals for permis¬
sion to reprint.
Einally, I would like to thank my family. There must be a special cor¬
ner of heaven for these much-put-upon people who listen for years to
stories and complaints of an arcane nature and who patiently take in
stride lunches cancelled and events missed. As for spouses, they are in a
saintly class all their own.
I NTRODUCTION

W hen I first imagined A Game for Dancers, I thought it would deal


with the shift in aesthetics that occurred in American modern
dance after World War II, a time during which dancers moved away
from the idea of communicating essential emotions to a so-called pure
dance. I wanted to investigate this shift, which seemed to have had
much to do with the advent of the Cold War and the resulting chill on
overt pohtical expression. However, compelling as this topic was, I even¬
tually reahzed that it was only part of a larger and more troubling issue.
This concerned modernism itself and what was possible in that late
modernist moment. Modernism, which largely had been intended to
give artists control over their work, was by 1945 being actively co-opted
by the very forces it had been meant to counter. During the 1940s and
’50s, dancers did not yet question the rules of the game they played;
they continued to attempt to find solutions within it. Yet how could
modem dancers fulfill the aims of a modernism that was ever more com¬
promised? And how could modernism serve the needs of a genre that
was itself facing institutionalization? It is the postwar dancers’ struggles
to solve their considerable problems within the framework of modernism
that is finally the subject of this book.
Although American postwar modernism has received a great deal of
attention in fields such as art history, cultural studies, and social criti¬
cism, the emphasis has generally been on the impact larger social forces
have made on art. A Game for Dancers differs from these studies in that
it considers a specific art field’s history and dispositions as cmcial fac¬
tors in how that field intersects with broader social issues. In particular,
competition for position within the field is a key element.
In dance studies itself, neither the postwar period nor the problems
of late modernism have received extensive investigation. The majority
XIV Introduction

of dance research has dealt with issues across a spectrum of modern


dance rather than those particular to the postwar period.' The few works
that have dealt specifically with these years have tended to emphasize
dance’s relationship to the Cold War.^ Although Cold War politics fig¬
ure prominently in the coming pages, they are viewed in relationship to
modern dance’s position in the dance field and the struggles that went
on there in the 1940s and ’50s. As for the topic of modernism, it also has
been viewed predominantly from the standpoint of American modern
dance as a whole, as in Julia Foulkes’s recent Modem Bodies: Dance and
American Modernism from Martha Graham to Alvin Ailey (2002).^ A
Game for Dancers instead draws attention to the problems surrounding
the body of modernist ideas and practices that postwar dancers and
commentators adhered to, manipulated, and exploited, but never overtly
challenged.

In attempting to come to grips with modernism in American postwar


dance, several sources have oriented my study.''' The first is John Mar¬
tin’s theory concerning the aims and methods of modern dance as a
modernist genre, which he formulated during the 1930s and which
serves as a base from which to speak about dance modernism. His work
will be examined in detail in chapter 3. Briefly, however, it called for a
dance that communicated abstracted emotion through movement forms
developed individually by each dancer. This “authentic” dance was
available to everyone because all humans shared physiological mecha¬
nisms that allowed them to intuitively assimilate meaning in movement.
Such a dance was characterized by new, independent, functional forms
that were developed to deal with the ever changing demands of moder¬
nity (Martin [1939] 1965: 42-55, 121-126). Martin’s ideas of the pro¬
cesses and goals of modern dance were generally shared by the field at
large and here serve as a baseline for speaking of modern dance.
The second element to set the course of my study are the debates
that have taken place over the last three decades in art history and crit¬
icism. Art critics and historians have emphasized the relationship be¬
tween art and capitalist ideology after the war, which is crucial to any
analysis of late modernism.^ Among the key questions they have asked
is whether modernism was simply an enforcer of the status quo in the
disguise of a revolution of form or whether it was able to work, at least
to some degree, as social critique. In addressing this issue I have given
Introduction XV

considerable thought to T. J. Clark’s description of modernism as “a


kind of skepticism, or at least unsureness, as to the nature of represen¬
tation in art” (Clark 1984: 10).In modern dance this doubt was mani¬
fested in an imperative to embody rather than represent meaning, a
subject I take up in chapter i. The notion of embodied meaning con¬
tributed to dance’s sense of its own autonomy and, significantly, set it
apart from painting and literamre.
As Clark points out, the need for autonomy, as a distinctive feature
of modernism, reached its apotheosis in Clement Greenberg’s i960
essay, “Modernist Painting,” in which he stated that an art that was not
independent was liable to be assimilated into entertainment or therapy
(Greenberg [i960] 1993b: 86). However, Greenberg had developed
his ideas on autonomy much earlier than i960. He initially laid them
out in his two much-quoted essays, “Avant-Garde and Kitsch” ([1939]
1986a) and “Towards a Newer Laocoon” ([1940] 1986b). Here Green¬
berg traced modernism’s beginnings to nineteenth-cenmry France when
artists were forced into the marketplace as a result of the rise of bour¬
geois power and shifts away from aristocratic and church patronage
after the revolution. Some artists, as a means of protest and to gain their
freedom from outside political and economic control, rejected market
institutions and methods. In this way they defined themselves as sepa¬
rate from, and in opposition to, the dominant class ([1939I 1986a: 7).
According to Greenberg, artists then turned their backs on politics to
focus their attention on preserving art, for art’s task had become to guard
the highest achievements of Western civilization from the onslaughts of
the ruling class. Artists preserved art by making it its own justification
([1940] 1986b: 28). That is to say, they intended art to be seen for what
it was—a painting, a drawing, a piece of sculpmre—^rather than an im¬
itation of some aspect of the external world. In order to give art this
independence, they engaged in self-criticism, in which they eliminated
from their medium all elements that were foreign to it, including anec¬
dotes, messages, and ideas. “Content is to be dissolved so completely
into form that the work of art or literature cannot be reduced in whole
or in part to anything not itself” ([1939] 1986a: 8). Art thus took a defen¬
sive position against the dominant order by retreating into a reductive
formal purity.
Most of Greenberg’s theory lay within the realm of generally ac¬
cepted modernist claims. However, he took to extremes the artist’s
independence from society and the limitations of artistic concerns. An
xvi Introduction

important aspect of the recent art historical debates has had to do with
the degree of separation from life called for in Greenberg’s theory and
how it related to postwar art and politics. The place of autonomy in
postwar modern dance is also a central point of inquiry in A Game for
Dancers. I will suggest that dance’s embodiment played a crucial role in
how dancers and theorists conceived of autonomy and consequently
of how that affected their ideas of dance’s relationship to society.
In addition to looking to art historical debates to help unravel post¬
war dance modernism, I have turned to elements of Pierre Bourdieu’s
theory of practice. Bourdieu posits that social agents (be they individ¬
uals, groups, or institutions) tend to think, perceive, and act according
to webs of durable, transposable, and generative dispositions that he calls
“habitus” (1990b: 52-65, see also 1990a, 1993b, Bourdieu and Wac-
quant 1992). Habitus is inscribed in bodies and constituted in practice.
Bourdieu sometimes explains habitus as “a feel for the game.” Agents
occupy positions within fields that they defend or contest. Bourdieu
conceptualizes fields as social microcosms that have their own struc¬
tures and laws (1993a: 181). At the same time that fields are governed
by objective relationships, they are also governed by their own history,
which agents both assimilate and respond to.
Bourdieu put his theory to work in numerous empirical studies, in¬
cluding that of the seminal period of nineteenth-century French mod¬
ernism (1993a, 1996). While concurring with historians on the general
outline of events that led to modernism, as a sociologist Bourdieu em¬
phasizes the developing structures and processes of modernism rather
than artists’ careers or works, or the evolution of forms and styles. He
particularly calls attention to the strategy artists developed to gain in¬
dependence from the market, referring to it as “turning the economic
world upside down.” Instead of producing art to meet demand, artists
claimed a disinterest in the market. They attempted to create an art that
for them was more truthful because it was not corrupted by market con¬
cerns. At the same time, however, as Bourdieu points out, the artists were
themselves bourgeois in the sense that they came from the bourgeois or
petty bourgeois class. Bourdieu calls them a dominated fraction of the
dominant class in that they had little power within their own milieu.
However, as members of the dominant class their position toward that
class was ambivalent. That is to say, even as they ridiculed it and battled
against it, they could not completely sever their ties from it. Some part
of them still yearned for acceptance. And some did, in fact, garner a cer-
Introduction xvii

tain recognition by standing against society and in time became known


and successful because of the status they had earned through their op¬
position. As Bourdieu points out, their disinterest became understood
and even respected, then became expected and eventually rewarded
(1996: 61). As consecration occurred, a new generation of artists ap¬
peared to take the place of their elders and to renew their oppositional
aims. This rejection of the past by young artists and their gradual as¬
similation into the mainstream became a hallmark of modernism. It
ensured an opposition to the art establishment and a market-oriented
society while at the same time providing an opportunity for success
within that world. Thus, the very processes of modernism demonstrated
an ambivalence that was similar to the artists’ own class attachments.
Although one might argue that artists’ work was simply co-opted by
the ruling order, such a case becomes more difficult to make when at
least some artists were rewarded with recognition and/or money. Wealth,
of course, is relative. Few dancers who did not participate in the com¬
mercial theater or films became wealthy in the postwar period. Nonethe¬
less, dancers such as Martha Graham, Alwin Nikolais, Paul Taylor, and
Merce Cunningham did in time become consecrated and were rewarded
with grants, honors, and other benefits of legitimation. Surrendering
to commercial interests may have been reprehensible to vanguard
artists, but few rejected the seduction of consecration, of becoming ei¬
ther “a modern classic” or, in today’s advertising jargon, “a cutting-edge
must-see.”
Although artists attempted to free themselves from the market, it is
important to note that, following Bourdieu, fields never gain complete
autonomy from larger social forces; rather they are semiautonomous,
having their own histories through which they refract outside influ¬
ences (1993a: 176-183). This means that art cannot be viewed either as
fully independent from, or as an unmediated reflection of, society at
large. This idea is particularly important with respect to the Cold War.
Although Cold War pressures were certainly felt in the dance field, how
it responded was dependent on its own history and internal needs. I will
argue that history and need were bound to modernism and its demand
for a renewed vanguard.
Although Bourdieu’s smdy of modernism deals only with nineteenth-
century France, it is clear that modernist structures and practices en¬
dured into mid-twentieth-century America, where dancers assumed that
a modernist vanguard existed and that it functioned much as outlined
xviii Introduction

here. The enduring nature of modernist structures and processes helps


explain why dancers did not question the rules they played by, and why
they did not understand the co-opting of modernism itself. To this ex¬
tent, what had begun as a response to a particular situation had become
an unquestioned set of rules. Yet at the same time, core aspects of that
predicament remained. Artists still sought independence from the mar¬
ket and from capitalist values while at the same time being inextricably
tied to the social order they opposed. Walter Benjamin described this
opposition from within as being like a secret agent in one’s own camp.
Speaking of Baudelaire he said, “Baudelaire was a secret agent, an agent
of the secret discontent of his class with its own rule” (quoted in Mar¬
cuse 1978: 20). A key question in this study is whether it was possible to
continue to carry out this subversive activity.
Modernist ambivalence showed itself in numerous ways during the
postwar years, none more so than in the debates that circulated around
issues of freedom. The rules of modernism required independence from
the market, and modern dancers were obsessed with the dangers both
of co-option and selling out. Yet they also argued endlessly over tire need
to enlarge audiences and gain a more secure financial footing. While
they worried over economic freedom, dancers were also concerned
with political freedom. Historian Eric Foner, in his study of American
freedom (1998) points out how concepts of freedom shifted between
the 1930S and the postwar years. During the ’30s, he writes, freedom
was reconceived from the old notion of freedom from government con¬
trol to freedom from the worst vagaries of the market. This idea was
manifested in the New Deal, which provided a larger role for govern¬
ment in guaranteeing the rights of labor and providing a social safety
net in innovations ranging from old-age pensions to unemployment
insurance. During the postwar years. Congress blocked expansion of
New Deal policies and pushed through legislation such as the Taft-
Hartley Act that weakened earlier gains made by labor. Historian James
Patterson notes that “the 1940s, a time of significant expansion of gov¬
ernmental social welfare in many western European countries, in fact
solidified the privatization of social welfare in the United States” (1996:
46). Paralleling shifts in attitudes toward social welfare were shifts in
ideas of what constituted freedom. Once again liberation became de¬
fined as freedom from government control. The irony was that at the
same time, federal institutions gained increasing power, particularly
those that relied on sxirveillance, such as the FBI, the newly created CIA,
Introduction xix

and investigative arms of Congress—notably the House Un-American


Activities Committee (HUAC). The dance field’s responses to these
contradictory pressures are a continuing theme throughout the book.
Yet to see modern dance only in terms of larger social issues would
be to lose sight of its internal struggles. These centered on the need to
renew itself in order to retain its vanguard position in the field. Until
the 1950S, modern dance defined itself as a genre that sought to com-
mumcate emotional essences. To use John Martin’s term, modern dance
was “expressional” ([1939] 1965). Its techniques and point of view were
based on those founded by the pioneers of the 1920s and ’30s. For
modern dance to achieve a renewal, somehow those original assump¬
tions had to be challenged without also destroying modern dance itself.
It is no secret that by the end of the 1950s “objectivist” modern dance
was declared the new vanguard.^ One of the questions I explore is why
objectivism succeeded over competing strategies. Where did it go for
support, and why was it embraced? Most important, was it able to ful¬
fill any kind of oppositional role at that late stage of modernism? Was
opposition still possible, or was it a kind of residual ghost that could be
invoked but no longer manifested?

In considering the problems I have outlined on postwar dance, I have


been particularly interested in the relationship between dancing and
writing, in how discourse and practice were intertwined in a variety of
ways and how that connection helped define what dance was at that
moment. To begin with, many dancers in the 1940s and ’50s explained
themselves in print. They declared their positions in manifestos, they
wrote at length of their aims, and they voiced their concerns. These
declarations, made despite the fact that dance was considered transpar¬
ent to any open minded spectator, demonstrates to what extent dancers
sought to shape the reception of their work. Another manifestation of
this intent was the ubiquitous lecture-demonstration, an integral part
of modern dance practice. Most often dancers expressed their views in
Dance Observer, which regularly ran their articles. Dancers’ supporters
also wrote articles. John Cage, Remy Charlip, and Carolyn Brown de¬
scribed Merce Cunningham’s work processes; Murray Louis explained
Alwin Nikolais’s philosophy; David Vaughan defended James Waring’s
choreography. Dancers who could muster this kind of support had their
work more widely discussed than those who could not, which aided
XX Introduction

their exposure, always an important aspect of the modernist process.


This discourse was an essential part of definition that accompanied the
dances themselves and established a way of seeing those dances.
Then there were the critics, the authoritative voices of interpretation
and legitimation. It is striking how close the connections were between
critics and dancers and how small the number of critics was, despite
more extensive dance coverage compared with the 1930s.® This made
the power relations between dancers and critics particularly complex.
Few critics held the view, common today, that it was a conflict of inter¬
est to be personally involved with artists. Walter Terry of the New York
Herald Tribune studied ballet and modern dance before becoming a
journalist and was a personal friend of many of the most celebrated
dancers. John Martin of the New York Times, although he professed a
determination to keep his distance, had given lecmre series at Benning¬
ton College in the 1930s, where he had close contact with many of the
first-generation dancers and which in turn helped him develop his own
dance theory. Edwin Denby danced in Europe before he began to write
dance criticism for Modem Music and, as a replacement for Walter
Terry, at the New York Herald Tribune during the war. More important,
there is evidence that he helped write George Balanchine’s articles for
Anatole Chujoy’s Dance Encyclopedia and perhaps others, as well (Mac-
Kay 1986: 31).^ Considering that Balanchine’s English at the time was
little more than functional, it wouldn’t be surprising that he (or, more
likely, Lincoln Kirstein) sought out a sympathetic critic to help express
the choreographer’s thoughts in written form.
Certainly the most powerful postwar critics were John Martin and
Walter Terry, the only full-time dance critics for New York daily news¬
papers. Dance Observer never tired of lamenting that the dailies rarely
covered concerts beyond Broadway or theaters close by, such as the Met
and City Center. Although it is true that Martin and Terry were obliged
to focus on productions in the Broadway area, they did manage to cover
more concerts outside it than might be expected. Yet these limitations
had their effect on where modern dance was performed. When pos¬
sible, dancers tried to appear in Broadway theaters on the Srmdays when
regular Broadway shows were closed. Although a crushing expense and
one that many could not afford, it was among the few ways modern
dancers could be certain to be covered by the Times and Herald Tribune,
as well as the other dailies.
There were three American magazines of the postwar period that were
Introduction XXI

dedicated to dance: Dance Observer, Dance Magazine, and Dance News.


During the 1940s, Dance Observer was the most important, serving as an
arena for dancers and their supporters to air their views. Dance Observer,
however, was not a neutral space in which debate took place. It had been
started by Louis Horst in 1934 as a means of supporting modern dance
in general and Martha Graham in particular. At the time Horst started
the magazine he was Graham’s music director, her mentor, and her lover.
Although Horst and Graham’s relationship changed over the years.
Dance Observer continued to support modern dance as Horst and others
had defined it in the 1930s. Considering that it was a part-time opera¬
tion run out of various New York apartments, the magazine wielded a
surprising amount of power within the New York dance field. The ed¬
itors of Dance Observer were often aligned with dancers or were dancers
themselves. Gertrude Lippincott, for example, was a dancer as well as
an editor who wrote several defining articles in the 1940s. Doris Rudko,
an associate editor in the 1950s, was a dancer and assistant to Horst in
his composition classes.
Non-dancers associated with Dance Observer included Joseph Camp¬
bell, who was an editor for several years. In addition to being married
to Jean Erdman, Campbell was a close friend of Martha Graham and a
respected scholar of mythology who had considerable influence on
modern dance in the 1940s. Horton Foote was also a Dance Observer
editor in the 1940s. He arrived in New York with an acting career in
mind and, in addition to the writing he did for Dance Observer, soon
began to work with dancers who were attempting to develop dance-
drama as a new form of m^odern dance. Dance Observer critics also wrote
for other publications. Robert Sabin, a music as well as dance eritic,
wrote dance criticism for Musical America in addition to his work for
Dance Observer. His sister, Evelyn, had been a member of Martha Gra¬
ham’s early group. George W. Beiswanger, a philosopher and aestheti-
cian, was also the dance editor of Theatre Arts magazine. In addition, he
wrote a monthly column on modern dance for Dance News beginning in
1943. He was married to dancer Barbara Page Beiswanger, who served
for a time as a Dance Observer editor.
While Dance Observer dominated the dance press in the 1940s, Dance
Magazine gradually came to the fore in the 1950s. Dance Magazine did
not include the lively editorials and reader debate that characterized
Dance Observer in the 1940s and that died in that magazine, as well, in
the 1950s. Dance Magazine was particularly aimed at dance students.
xxii Introduction

and included many articles on educational activities and columns on


such topics as dancers’ health. The magazine was also directed toward
advocates of ballet as well as of modern dance. In all then, Dance Mag¬
azine was a less specialized publication than Dance Observer, and it rarely
engaged in controversy or took strong stands on issues. Its very bland¬
ness, coupled with attention to advertising and sophisticated layout
and photography, may have been major reasons it came to dominate
the dance scene in the 1950s. During the 1940s, Dance Magazine ran
few reviews, but that changed in 1952 when Doris Hering was hired as
a permanent member of the staff.^° During the 1950s, she covered an
extraordinary number of concerts each month, nearly as many as Dance
Observer with its half-dozen critics. However, this also meant that with
rare exceptions Dance Magazine had only one critical voice for most of
the postwar years. Although often more flexible than Dance Observer
critics, Hering shared a similar expressional viewpoint.
The other dance publication of those years was Dance News, founded
by Anatole Chujoy in 1942, which had a tabloid format and came out
monthly. Chujoy, a native of Latvia who had received a law degree in
St. Petersburg, was a knowledgeable devotee of ballet and particularly
dedicated to Balanchine; he wrote the first book on the New York City
Ballet when the company was barely eight years old. In 1951, Chujoy
hired P. W. Manchester, an English critic, to help him cover the dance
scene. Although Dance News always favored ballet, Manchester reviewed
a number of modern dance performances. Like most of her colleagues,
she viewed modern dance through expressional eyes. The other critic
of note during that time was Margaret Lloyd. Lloyd wrote for the Chris¬
tian Science Monitor, which was available throughout the country. But of
particular importance was the book she published in 1949 entitled The
Borzoi Book of Modem Dance. Like Sally Banes’s Terpsichore in Sneakers
several decades later, it went far in anointing certain modern dancers
and in arranging them into categories of pioneers, which she called rev¬
olutionaries, and new leaders. Martin and Terry also wrote general
dance history books that similarly accomplished the task of categorizing
and consecrating various dancers.
Although most of the dance press supported an expressional aesthetic,
toward the end of the 1950s, several critics appeared who advocated ob¬
jectivism, among them David Vaughan, Selma Jeanne Cohen, and Jill
Johnston. Denby must also be counted in this camp. Although he wrote
little during the latter part of the 1950s, he had long supported Cun-
Introduction xxiii

ningham and had even longer been an advocate of objectivism through


Balanchine’s choreography. It should be added that although the dance
press was predominantly male by a small majority, with the exception of
Doris Hering at Dance Magazine the most powerful critics were men.
Dance Observer usually had slighdy more men than women critics, but
those who wrote most often during the 1940s and ’50s—Robert Sabin,
George Beiswanger, and Nik Krevitsky—^were men, and of course Louis
Horst maintained control over the magazine as well as writing reviews.
One could find dance coverage in magazines outside the dance press,
especially in Theatre Arts and Musical America. Occasionally more gen¬
eral interest magazines covered dance, such as Time, Newsweek, and Life.
As would be expected in conservative magazines, these extolled ballet,
most often in the form of attractive girls, while generally ridiculing mod¬
em dance. This kind of opposition was necessary if modern dance was
to retain its vanguard status. To have been accepted by the popular press
would have called into question the genre’s rebelliousness and its posi¬
tion as the advanced guard of high-art dance.
As for race, the dance press (here referring to critics who special¬
ized in reviewing high-art dance) was exclusively white. The general-
readership black press rarely dealt with dance unless it was a concert by
a major performer, such as Katherine Dunham, or Broadway musicals
that featured African Americans, such as Show Boat or Cabin in the Sky.
Amsterdam News, New York Age, and People's Voice (all in New York),
the Afro-American in Baltimore, the Norfolk Journal and Guide in Vir¬
ginia, and the Chicago Defender occasionally covered high-art dance, but
in the 1940s and ’50s they focused the majority of their attention on
Dunham and Pearl Primus. These publications were most interested
in success gained and racial barriers broken by black performers rather
than having any particular commitment to art forms that showed little
interest in black venues or black audiences and in which black artists
were not widely represented.
The leftist press that had been so active in the 1930s in reviews and
discussions of dance slowly died as the Cold War set in. Most publica¬
tions were gone by the end of the 1940s. New Masses, which had pro¬
vided lively dance coverage for years, succumbed in 1948. Its last review,
which appeared in March 1948, was of an Anna Sokolow concert by its
long-standing dance critic Edna Ocko, named by Jerome Robbins as a
communist in his appearance before the House Un-American Activities
Committee in 1953.
xxiv Introduction

In all, then, dance coverage was more extensive than it had been in
the 1930s, but the group who actually did the writing was small, and the
vast majority of critics honed to an expressional aesthetic. However,
above and beyond defenses of one aesthetic or another, writers of all
stripes were notable for their willingness to invoke elements of mod¬
ernism that supported their cause while deemphasizing, eliding, or ig¬
noring aspects that did not. In addition, they drew on modernism to
support broader agendas related to race, class, and gender and to either
disrupt or fortify the status quo. Thus modernism, although constitut¬
ing what Michel Foucault would have called a discursive formation, was
flexible in the sense that it was employed for a variety of purposes, many
of which were at odds with each other.
Where, it might be asked, does this abundance of writing leave danc¬
ing? Does discourse simply write over bodily practice, or is there some
less unequal relationship between them? That is a question that is con¬
fronted throughout^ Game for Dancers. I treat dancing as a socially
significant, embodied practice, and as such its analysis plays a central
role in the book. In general I follow a tradition that is based on a notion
of the “lived body,” the idea that human beings and their consciousness
are embedded in the body (Nettleton and Watson 1998: 9). This tradi¬
tion is represented by such theories as Marcel Mauss’s techniques of the
body ([1934] 1973), Erving Coffman’s concepts of bodily presentation
and communication (1959, 1963, 1979), Bourdieu’s idea of bodily intel¬
ligence, and Susan Foster’s notion of the theorizing body. Mauss showed
how common activities like walking and swimming are learned and cul¬
turally specific, while Coffman demonstrated how bodily movement
is used to communicate meaning in everyday life. Bourdieu’s ideas re¬
late to his theory of practice, in which the body is necessarily key. He
describes what he calls bodily “hexis,” as “political mythology realized,
em-bodied, turned into a permanent disposition, a durable way of stand-
ing, speaking, walking and thereby of feeling and thinking” (1990b:
69-70). Hexis is a crucial part of Bourdieu’s concept of habitus, that is,
dispositions that seem natural but are aspects of learned behavior that,
in this case, literally embody social position and attitudes and that in¬
fluence not only the thinking and feeling of the actor but of others. This
view of bodily practice offers the possibility of dance not only ordering
thoughts and feelings through choreography but through the very way
in which bodies confi'ont the world.
Bourdieu s notion of the social as bodily practice extends to a concept
Introduction XXV

of theory as practice. He argues for there being a bodily intelligence


that lies outside the realm of conscious reason. He writes: “There is a
way of understanding which is altogether particular, and often forgot¬
ten in theories of intelligence; that which consists of understanding
with one’s body” (1990a: 166). Foster enlarges this idea when she con¬
tends that through corporeal intelligence dance cultivates a body that
is capable of generating ideas as bodily writing (1995: 15). This is espe¬
cially evident in an area of dance such as the rehearsal process, where
give-and-take between choreographer and dancers forms the basis of
dance production. In the course of this activity, movement queries are
raised and solutions offered that produce particular bodies on the stage.
That is to say, something like a physical conversation goes on between
dancers and choreographer. The choreographer shows the dancers a
movement segment, the dancers adjust the movement to their own
bodies, frequently offering movement alternatives, which the chore¬
ographer accepts or changes again. The process continues in this way
until the dance is finished. Foster notes that recent critical studies in
dance have begun to treat bodies as not only written on but as capable
of writing. Such bodies, endowed with agency, have the capacity to
theorize. However, “the theoretical, rather than a contemplative stance
achieved afterwards and at a distance, becomes embedded (embodied)
within the practical decisions that build up, through the active engage¬
ment of bodies, any specific endeavor” (1995: 16). Although Foster’s
language is Foucaultian and therefore implies a discursive body, the
process she describes is a material one of bodies informing bodies.
Bourdieu would say these practical decisions, filtered through the habi¬
tus or bodily dispositions of the dancers and the dance, embody the
social. The body generates ideas through habitus, that is, the ideas are
socially constructed and limited.
For both Bourdieu and Foster, then, the body is capable of generat¬
ing ideas or theorizing through practice rather than through conscious
thought. This is not to say that dance has no relationship to language,
that it is somehow ineffable, or that choreographers don’t engage in
rational deliberation. But it does mean that dance also functions in a
way that is different from language and conscious thought. Bourdieu
and Foster’s concept of bodily theorizing is important for a social analy¬
sis of dance because it makes it possible to see the social not only in who
sponsors the dance or views it, or what kinds of narrative themes dance
embraces, or how dance might serve a broader social fiinction, but to see
xxvi Introduction

the social as embedded in the practice of dance, in the dancers’ comport¬


ment and the steps they do, in how the dance movement is assembled
and how the dancers are arranged on the stage, and in how dancers are
trained and developed. It is therefore dance as a social practice that in¬
forms the analysis of works in this book.

A Game for Dancers is organized around a set of problems and attempted


solutions. It follows a more or less chronological path, although there
is time overlap between chapters. However, although the book is to my
knowledge the first to look closely at dance within the context of post¬
war modernism, it is in no way meant to be a survey of postwar dance.
To begin with, it defines postwar in narrow terms as the years between
1945 and i960 and does not stray far beyond those bounds. It also fo¬
cuses exclusively on New York, both because of space constraints and
because New York was the center of American dance at that time. Even
so, many important choreographers who performed in New York are
hardly mentioned, including Sybil Shearer, Katherine Litz, Erick Haw¬
kins, Mary Anthony, Louis Johnson, Daniel Nagrin, and many others.
A number of dancers who emerged in the latter part of the 1950s also
have been omitted in favor of artists whose careers were more fully
developed within the temporal framework of this study. The choreog¬
raphers whose works were chosen for analysis were pertinent to the
particular problems I outline. In some cases, too, they greatly influ¬
enced the generation of dancers who came after them, and their legacy
is still discernable today. I hope that now that the postwar period has
been opened to investigation, studies of the work of many choreographers
active during those years will follow. Some, like Shearer, Litz, Hawkins,
Johnson, Nagrin, and Anthony, are familiar names whose dances deserve
more sustained analysis within the context of postwar concerns, while
others, such as Iris Mabry and Ronne Aul, have been nearly forgotten
but also warrant attention.
A brief note on the photographs that appear in the book; wherever
possible I have used photographs of original casts and productions. How¬
ever, that has not always been possible for a variety of reasons. When I
have used photographs from later productions, I have noted it. Photo¬
graphs are at best personal and artistic interpretations of what the
photographer sees within the limitations of technical possibilities. The
1940S and ’50s were still the days of light meters and flashes. Most
Introduction xxvii

photographs were not made on the stage during performances but in


studios, and often in poses that had little to do with the choreography
of the work being portrayed. Production shots were far less frequent
than they are today, owing to cost and technical difficulties. The photo¬
graphs included here are therefore meant simply to suggest dances
rather than to be precise documents of them.
The first chapter of A Game for Dancers sets out the difficulties mod¬
em dancers faced at the conclusion of the war. It discusses their predica¬
ment within the dance field and examines their relationship to ideas
that were circulating in the broader intellectual field. It also looks at
one highly controversial solution to modern dance’s problems, which
involved what was seen as a new form of dance-drama.
Chapter 2 deals with ballet, a seemingly quixotic sidetrack in a book
focusing on modern dance. But in Ballet Society, Lincoln Kirstein and
George Balanchine challenged modern dance’s vanguard position and
sought to replace it with a modernist ballet. This was not the only
competitor from the world of ballet; Antony Tudor took on many of the
elements of modern dance, as did Agnes de Mille and Jerome Robbins
to a lesser extent. However, Tudor, de Mille, and Robbins were aligned
with expressional models, while Balanchine offered an aesthetic that
was closer to an alternative to expressional dance. It should be added,
though, that this alternative was in form alone, since ballet had no real
oppositional aims to the status quo. In the 1950s, a young generation of
modern dancers used elements of the kind of objectivist aesthetic Bal¬
anchine advocated to develop a new modern dance vanguard.
The third chapter of the book looks closely at the dance theory of
three men—John Martin, Edwin Denby, and John Cage—all of whom
influenced the dance field in the postwar years. Although they were at
odds on specific issues, I point out their shared modernism and how
they theorized an authentic dance as a solution to modernity’s techno¬
logical and bureaucratized rationalization.
Chapter 4 explores the relationship of dance modernism to issues of
community and difference in light of consensus ideology. In some in¬
stances choreographers from minority constituencies attempted to de¬
fine communities distinct from a dominant culmre. This chapter looks
at Jewish dancers who sought to essentialize their identity in the wake
of the Holocaust and the founding of the state of Israel. In trying to deal
with difference, dancers put pressure on the boundaries of modernism
and at the same time demonstrated modernism’s limits. This chapter
xxviii Introduction

also examines works by Anna Sokolow that called attention to a society


in which collective life had disintegrated. In carrying out this task she
introduced innovative approaches to form that allowed her to encom¬
pass social critique through non-dance movement associated with neu¬
rosis and madness. Continuing an examination of community and dif¬
ference, chapters 5 and 6 focus on African-American choreographers
who saw modernism’s demand for universal experience used against
them in contradictory ways, even as they gained greater access to the
modern dance genre.
The final chapter of A Game for Dancers considers objectivist mod¬
ern dance, focusing on the work of Alwin Nikolais and Merce Cunning¬
ham. I contend that at least in these two cases objectivist choreographers
attempted to tie formal innovation to elements of resistance, resistance
that went unrecognized.
At this point one might well ask why it is of interest to investigate
modern dance at the end of a period of pioneering optimism and growth,
a time when, as a genre in transition, it presented more problems than
solutions and when it was beset by doubt and struggle. The answer is
the obvious one—that the problems of the postwar years are still with
us, in fact exist with particular force today. Not only are we experienc¬
ing an oppressive conservatism and attacks on civil liberties that recall
the postwar era, we also face the ever increasing power of global capi¬
talism, raising questions of resistance and how it is expressed. Postwar
dancers faced an increasingly centralized and oppressive power in a va¬
riety of ways, and there are lessons to be learned in the complexity and
difficulty of their predicament, as well as in the solutions they offered.
A Game for dancers
r
t
- V ■^1 J
• ^

■>-*‘ • ^ ...
I * . ^ • I la 'h ":* *

4t \ • ■.•'* 'V'^'' '*•'' f -■■ ♦’*'


r'-
fT'.
, <
, ', ' 1,, 1 tfrti*'/ i
^ ’s^ ^Oi *

■ >
^5 ,. - •'*'ir “T • .• ' <' * L

t '
. « • .♦.
• .1 '*

\ y' JEt:W •-'-■■


’•t ’ III' . * -v^- '< ■?/ ■.. '
V » ►"

..1 ■ . > • __ ^ V jr^.'.ifJ'-.yjHw a i ..' ■’ Kt?

' ■- .■■• r.,.vv«,N;. 'fi


''-'.wi ^"■. . A ■ ;■ Msjy ' • ■*.- ii»!iU
/i ^.l
•. j ' !•«•. «•.*' ', *■-‘ Cl '...'r'i< !

it
•f
’ >' '^.•V.'.!'*' '..1; I *■ V V •‘,*
■ .
•“% f
• '^'0 •. ^' •.' kf* . ’siiS .‘‘A '/r'^ .*■» . .
■•.';■••-.•■
'■ ; ■j’
r. ';',,.'.v' ii*'-Iii ^•'»■<*
; ,’-s ’ ' r - ^ , * , -
J'"‘ ■ v’ •■ 'iir.Vf- "•'V • ;•.

V-.\ . y...;. '•• - *• ’ ■' •' '■

. , -^'^■
'■.
'■ / /«<
)k. f ^
... .■

. ^ .-'I /, • ,-l; . ^
f-'V
••VI /
. / . I
f }

..:.. ■

hK
.' ev
•' ' ’ 1 f ■
*
. r-
I

THE TROUBLE WITH


MODERN DANCE

H orton Foote wrote in October 1944 that modern dance had


reached a crisis in its developmentd The twenty-eight-year-old
Dance Obse'}~uer editor noted that modern dancers had begun by fighting
for aesthetic goals. Now, he said, they faced another kind of battle, one
against commercial forces that threatened their artistic independence.
“Every day on the Theatrical Page we read of one more well-known
dancer going to Broadway or Hollywood. It has become a mammoth
exodus, a giant trend” (1944: 98). Foote warned that modern dance
consorted with show business at its peril. The economic bottom line
was all that mattered in the commercial theater, killing “individualism
of taste, daring, experiment.” Quoting novelist James Farrell, he con¬
tinued: “This is a culture which does not serve men; on the contrary it
makes men its servants” (ibid.).
Certainly Foote was right in saying that the borders between mod¬
ern dance and the commercial theater were porous.^ Although dancers
had long supplemented their small earnings by performing in com¬
mercial shows and nightclubs, commercial work for high-art dancers
increased markedly after the war. This occurred primarily because a
ballet-based dance began to compete with tap as the major form on
Broadway and in films, giving modern and ballet dancers new opportu¬
nities to work in shows and Hollywood (Moulton 1957; Feuer 1993;
Steyn 1997).^ George Balanchine had used ballet in the shows he did for
Broadway and films in the 1930s and early 1940s, but it was Agnes de
Mille who was credited with sealing ballet technique for Broadway
musicals in 1943 with Oklahoma! By 1945, ballets and ballet-based dance
were part of many Broadway musicals. The majority of these dances were
not done on pointe and could be managed well by modern dancers, who
were also more available than ballet dancers because they seldom worked
2 A Game for Dancers

under extended contract obligations. Also, several major modern dance


choreographers mrned to the commercial theater during the 1940s,
most notably Hanya Holm and Helen Tamiris. Balanchine also contin¬
ued commercial work into the 1940s, and Jerome Robbins began his
Broadway career in 1944 with On the Town. Ballet and modern chore¬
ographers who were doing Broadway shows often engaged high-art
dancers. Balanchine and Katherine Dunham joined forces for Cabin in
the Sky in 1940; Agnes de Mille hired Erick Hawkins to dance in Okla¬
homa! (1943); Helen Tamiris used Daniel Nagrin in most of her shows
and gave Pearl Primus a starring dance role in Show Boat (1946), which
also included Talley Beatty and Joe Nash in the cast; and there were
many others.
Films, too, began to integrate ballet into musicals more thoroughly
than they had earlier. Gene Kelly and Fred Astaire added ballet to
their films, bringing to prominence a number of ballet-trained dancers,
including Vera Ellen, Cyd Charisse, Taina Fig, and Leslie Caron. Both
modern and ballet dancers were hired to perform in the new ballet-
oriented films that ranged from Seven Brides for Seven Brothers (1954) to
Kelly and Astaire hits such as An American in Paris (1951) and Silk Stock¬
ings (19S7). The Red Shoes, which was released in 1948, also fed the taste
for ballet in films. In addition, high-art dance was appealing to tele¬
vision, which was just beginning to emerge as a major commercial genre.
Dancers appeared primarily on variety shows to add a touch of sophis¬
tication to the various acts. Ballet dancers were far more in demand than
modern dancers since they were what most of the general public thought
of as high-art performers; they were also more glamorous and could
convincingly be billed as stars. Modern dancers, at least to judge from
announcements in the press, tended to be seen most often on refigious
programs where perhaps their seriousness of purpose and relative auster¬
ity were not considered a deterrence to audiences. It is significant, too,
that after the war the dance press routinely covered Broadway shows,
films, and television appearances by high-art dancers. Dance News, Anatole
Chujoy’s monthly, not only reviewed these events but included news
and gossip, especially about Broadway shows.
The symbiotic relationship between high-art dance and the market
was actually built into dance practice, since it enabled many dancers
to have high-art careers. If dancers did not work in the commercial the¬
ater, their only means of support outside of performances was teaching,
often sporadic and not well paid, or taking menial jobs such as waiting
The Trouble with Modern Dance 3

tables, which would not unduly interfere with rehearsals and perform¬
ances.*^ Commercial employment allowed dancers to continue to per¬
form and choreograph high-art work even as it enticed some of them
away permanently.
Considering the amount of traffic that circulated between high-art
and commercial forms, much of it advantageous to high art, one might
ask why Foote was so alarmed. There were several reasons. To begin
with, as has been pointed out, modern dance’s position vis-a-vis the
dominant order was always ambivalent, but by the mid-1940s the bal¬
ance had started to tip. As Foote noted, the lure of entertainment was
killing the spirit of experiment that was necessary for modern dance to
remain a modernist form. This in turn jeopardized the genre’s position
in the field. However, the fear that commercialism threatened high art
was also part of a larger phenomenon that obsessed artists and intellec¬
tuals in the postwar years. Commercialism was an element of ration¬
alization, a central issue of modernity. Early in the twentieth century.
Max Weber had predicted that the use of rational calculation by busi¬
ness, the state, science, and technology, would increasingly infiltrate
all aspects of life (Weber [1904-5] 1999; 1946; 1978). Rationalizing
processes were spurred by capitalism and by state, military, and corpo¬
rate bureaucracies. Although art was not Weber’s central concern, his
theory posited that rationalizing processes penetrated everywhere, so
art, too, was at risk. Weber himself wrote an essay, published shortly
after his death in 1920, showing how the development of the piano, its
music, and practice were directly related to market forces and bourgeois
domestic demand (1978: 378-382).
In the 1940S, artists and intellectuals became particularly concerned
with the market in terms of the relationship of high art to mass culture.
Among the early indications of this was Greenberg’s “Avant Garde and
Kitsch,” published in 1939 in Partisan Review ([1939] 1986a: 5-22). In
his essay Greenberg not only defined the goals of what he called the
avant-garde, he examined its opposite, kitsch. He defined kitsch as the
mass-produced imitation of high art consumed by mass society. Con¬
trolled by market demand and produced for the many wage-earners
needed for the capitalist system, kitsch was meant to provide easily di¬
gestible distraction. Where genuine art demanded a viewer’s reflection
to make its effects, kitsch had effects built into it, a message, as it were,
ready-made and controlled by capitalist interests. Greenberg’s idea of
mass culture would soon receive more thorough analysis in Theodor
4 A Game for Dancers

Adorno and Max Horkheimer’s “The Culture Industry” ([1944] i979)>


written during the war while the two memhers of the Frankfurt School
were living in N^ew York. For Adorno and Horkheimer the culture
industry controlled the mass production of products, including enter¬
tainment, and it in turn was controlled by those who had the greatest
economic hold on society (ibid.: 121). As the power of the culture in¬
dustry grew, it ever more fully produced consumers’ needs, controlling
and disciplining them (ibid.: 144).
Although for Adorno and Horkheimer there was little way to es¬
cape the grip of the culture industry, Greenberg was somewhat less pes¬
simistic. Opposing kitsch or mass culture was the avant-garde, the seg¬
ment of the art field least subject to the market. As indicated in my
introduction, Greenberg saw the task of the avant-garde as one of pro¬
tecting genuine culture in the face of overwhelming rationalization
and capitalist decline. However, Greenberg argued that once the avant-
garde had defined itself in opposition to bourgeois culture, it aban¬
doned politics for the rarefied atmosphere of pure art. Significantly, he
felt the primary function of the avant-garde “was not to ‘experiment,’
but to find a path along which it would be possible to keep culture mov¬
ing in the midst of ideological confusion and violence” ([1939] 1986a: 8).
This would be art’s position until the triumph of socialism.
Foote, in his Dance Observer article, echoed Greenberg’s position on
high art’s relationship to mass culture. Represented by Broadway and
Hollywood, mass culture was a threat to the authentic art of a modern
dance, which Foote characterized as individual, daring, and experimen¬
tal. Foote felt the need to reiterate modernist principles at a moment
when high-art dance was particularly vulnerable to commercial inter¬
ests. However, he did not advocate that artists stop experimenting, turn
their backs on society, and assume the defensive position Greenberg
called for. On the contrary, it was experiment that needed protection
and the answer was not to go into hiding but to fight back. He called for
artists from all fields—he mentions Tennessee Wilfiams, Leonard Bern¬
stein, and Katherine Anne Porter, among them—to band together to
demand they be given the means to do their work. “Together we must
force our society to recognize our intrinsic worth as builders of a better
and more healthy future as assets of the national life just as the scientist,
the university or the scholar” (1944: 98). Foote did not specify how
society would compensate the artist. However, he did say that if it was
not possible in the present system, “then our energies must go to rid
The Trouble with Modern Dance 5

ourselves of this system” (ibid.: 99). Seven months after his initial article,
Foote wrote another in which he clarified his thinking. He now referred
to a “Society of Artists,” an association that would not remove the artist
further from society, he said, “but would more securely make him a part
of it” (1945b: 56). This stake in society would come in the form of gov¬
ernment subsidies to artists.
Foote’s two articles are a bridge between thinking derived from
New Deal principles of the 1930s and what would become dominant
postwar ideas. His articles were written in 1944 and 1945, as the war
turned definitively in favor of the allies and the United States prepared
for world leadership once peace was achieved. New Deal thinking was
still common currency, although in fact Congress had stalled much leg¬
islation after Roosevelt’s reelection in 1936. Funding for the Federal
Theater Project ceased in 1939, and the Works Progress Administra¬
tion as a whole was formally disbanded in 1942. Although President
Truman proposed numerous social welfare projects in his first budget,
he was able to carry out few of them in the face of determined congres¬
sional opposition. Arts subsidies were not even among his proposals.
The idea of banding together in a community to change society, so
much a part of 1930s strategies, would seem increasingly impossible as
the postwar period took hold. Dancers became distrustful of any idea of
government support, seeing it as compromising their independence.
Even in 1944, Foote’s ideas of communal action and government sup¬
port were being consigned to the past. However, his concern with the
problem of a rationalized art and the need for a modernist vanguard to
stand against mass culture were sure signs of the future.
As early as 1946, an article appeared in Dance Observer advocating an
approach different from Foote’s to the problem of modern dance’s re¬
lationship to commercialism. It was written by another editor, Gertrude
Lippincott, who was a dancer and educator. Lippincott’s article, titled
“Pilgrim’s Way” (1946: 84-85), typified the thinking that would come
to dominate during the postwar period, although the article was contro¬
versial at the time.
Beginning, like Foote, by defining modern dance’s predicament, she
compared the artist to a pilgrim who is constantly tempted by com¬
mercialism to win success. She said that Broadway choreographers had
incorporated modern and ballet techniques into their dances, and in
order to attract audiences, modern dancers had increasingly studied bal¬
let and added its virtuosic elements to their own techniques. Modern
6 A Game for Dancers

dancers were substituting virtuosity for experimentation and inven¬


tion in their dances, and they were making their works less complex and
easier to understand in order to appeal to the uninitiated. She reminded
readers that the aim of modern dance was not to entertain.
She then went on to say, in opposition to Foote, that politics had no
place in modern dance because when it did, dance ceased to be art and
became propaganda. “The arts have a life of their own, and although
they are sometimes influenced by existing social and political happen¬
ings, more often they are not” (1946: 84). Art deals with timeless and
universal themes, not local, transient ones, she said. Lippincott ended
by stating that the vanguard artist “cannot expect easy popularity unless
he is willing to lower his standards by becoming ‘chi-chi,’ Broadway-
theatrical, or downright cheap” (1946: 85). Then, reiterating modernist
rules, she stressed that the artist must accept the fact that audiences
would remain limited because only a small number of people were able
to understand “so strenuous an artistic experience” (ibid.).
Much of Lippincott’s article dealt with the threat of mass culture as
a corrupting influence on authentic art. Broadway had stolen modern
and ballet techniques, and modern dancers had been lured into adding
entertainment elements to their work in order to profit from larger audi¬
ences. As such, she echoed Greenberg’s warnings on kitsch and Adorno
and Horkheimer’s views on the insidious power of the culture industry.
She also came down on the side of Greenberg when she asserted that
art must distance itself from politics in order to retain its independence
from the field of power. In her view, political subject matter compro¬
mised the autonomy of modern dance because it placed the dancer
under the control of external ideological forces. For Lippincott, politics
and commercialism both threatened the artist’s freedom to work ac¬
cording to her own conscience. Artists could not afford to turn their art
into propaganda anymore than they could afford to bow to the whims
of current fashion. She concluded, “The way of the artist, in our west¬
ern society, is a lonely one. He must remain primarily an individual cre¬
ator, although he works under the shadow of current social events and
artistic trends” (1946: 85).
Lippincott’s modern dancer was very different from Foote’s. Gone was
Foote’s community and in its place was the lone artist, who separated
herself from direct social influence in order to remain independent and
true to herself. Like Foote, she did not imagine the artist giving up ex¬
periment, but she restricted it to a timeless world of abstraction.
The Trouble with Modern Dance 7

After the war, few dancers took on Foote’s idea of a community of


artists demanding government subsidies. It was clear that government
support was unlikely and that the kind of political action Foote advo¬
cated would fall on deaf ears. But in 1946, some dancers still argued that
art could and should have a political purpose, at least implying that it
was possible to change institutions and policies. Lippincott’s article
received responses in this vein from Mary Phelps (1946b: i lo-i 11) and
Elizabeth Wiggins (1946: 125). Phelps, who was also an editor for Dance
Obsei'ver, argued that art and politics had long been at home together,
and listed a number of artists from Shakespeare to Beethoven whose
works were political in nature. The artist who shunned politics, she
argued, gave up the chance of influencing society: “To expose falsity
and meanness on whatever ground; to confront the terror; to vindicate
a truth and see it grow; to make fun; to call for hope and pay with joy
that comes to no one alone—these are in the power of art” (1946b: 110).
Although some, like Phelps and Wiggins, argued for a political dance,
dancers and their supporters increasingly tended to center the debate
on issues of accessibility versus vanguardism rather than a political ver¬
sus apolitical binary. The question they most often argued was, should
dancers try to make their work more accessible to a mass audience or
preserve their modernism by producing experimental work for them¬
selves and their peers? When the notion of politics appeared, it was
pushed onto the side of accessibility. For example, dancer Joseph Gif¬
ford, in answering an article written by another dancer, Juana de Laban,
characterized socially conscious dance in this way. In her original article,
Laban had deplored modern dance’s emphasis on “the intellectual and
the abstract” and called for a more dramatic form that would draw on
American subject matter and rhythms and thereby appeal to wider au¬
diences (Laban 1945: 55-56). In his excoriating response, Gifford said
he found in statements such as hers “an almost frightening emphasis
on such things as audience acceptance” (Gifford 1945: 66). The real
problem with modern dance, he said, was not lack of accessibility but
that youngsters were copying the innovations of their elders, whose
work had already been absorbed into the mainstream. Shaping a dance
to conform to audience expectations, he said,

created the almost ludicrous examples of “folksy” art that is neither of the
folk nor art. It created the bastard “blues-y” art, wherein the dancer de¬
velops one superficial aspect of the people called, The Negro, and feels
8 A Game for Dancers

he or she is prancing in the democratic rhythm of our times. And, more


recently, it has created the amusing examples of Modern Dance being
interpolated into Broadway musical comedy—its note up high but one
eye always looking down to make sure that its feet stay clear of the pro¬
ducer’s toes. (1945: 66)

For Gifford, trying to identify with or defend “the people” was not a
political act but merely one more example of attempting to be acces¬
sible, which he equated with commerciahsm. In addition, Gifford said,
modern dance’s acceptance by “liberal” political groups meant that it
was now capable of being used as a facile propaganda tool. Although
Gifford didn’t define what he meant by liberal, one assumes it was a eu¬
phemism for leftist groups and the Communist Party in particular.
Gifford argued that serious modern dancers must reject all outside in¬
fluence and look to themselves when making dances. Gifford saw the
artist, as would Lippincott, as a lone soldier battling against institution¬
alized forces. However, it should be mentioned that unlike Greenberg,
neither Gifford nor Lippincott suggested that the task of modern art
was simply to guard the fortress of civilization against the philistines.
Experiment, as they both stated, was necessary to guarantee art’s in¬
dependence. The kind of defensive position advanced by Greenberg
would merely reinforce the status quo, and as such signify defeat.
Another important theme that Gifford touched on was that of na-
tionahsm versus internationalism. Gifford was contemptuous of Laban’s
suggestion that dancers draw on American sources, which he viewed as
a form of nationalism. In the 1930s, high-art dancers had turned to
American subject matter to attempt to establish an American artistic
identity. Humphrey’s The Shakers (1931), Graham’s Frontier (1935)
and American Document (1938), Tamiris’s Negro Spirituals (1928-1944),
and numerous other works exemplified dancers’ interest in American
sources. During the war such subject matter reinforced American pa¬
triotism and continued to be popular. However, once the war was over,
this view changed markedly in most quarters. The notable exceptions
were the remaining leftist dancers, many of whom were associated with
the New Dance Group and were the ones who came under attack from
Gifford.
The shift toward internationalism was at first linked to antifascism.
During the 1930s and war years, American newspapers had run nu¬
merous articles and cartoons that depicted the Nazis as destroyers of
The Trouble with Modern Dance 9

Western art and culture. This was particularly the case in the face of the
“degenerate” art exhibition of 1937 (which toured Germany and Aus¬
tria for four years) and the influx of European vanguard artists into the
United States (Hoelterhoff 1975; Hinz 1979; Barron 1991). However,
as Serge Guilbaut has pointed out, the fascists were not destroying all
Western art, just modernist art, and so as an enemy of fascism the
United States was also obliged to defend modernism. In addition, Guil¬
baut asserts that the United States, now a world leader, needed an art
that took a comparable leadership position (Guilbaut 1983: 54-55; see
also Saimders 1999; Caute 2003). As such, the old national themes and
motifs that had been so important in defining an American art were no
longer acceptable. Not only were they too provincial, they were con¬
sidered too close to the kind of nationalism that had characterized Nazi
Germany. Now an international viewpoint was called for that could
play on a world stage. Typical of this postwar stance was an article writ¬
ten by art historian H. W. Janson in the Magazine of Art, which attacked
the American regionalist artists, who during the Depression had gained
fame painting scenes of American life in a social realist manner.^ Janson
accused these painters of being dangerous nationalists, then went on
to say:

The great artists of the past and present, no less than the great philoso¬
phers and scientists, have always conceived of themselves as servants of
mankind, rather than of particular nations or groups. Their work has
been based upon the acceptance of ethical and religious values embrac¬
ing all members of the human family, regardless of origin or nationality.
Nationalism cannot possibly take the place of this allegiance to hu¬
manity as a whole. Wherever such a substitution is attempted, whether
by force or by guile, it will inevitably produce the same sad results.
(1946:186)

In Janson’s article, regionalism, with its focus on American themes, be¬


came associated with nationalism and fascism and by extension with
propaganda and the manipulation of mass society.
Although Gifford did not tie Americana dance works to fascism, he
did speak of their easy use as a propaganda tool. In this case the villain
was the political left, which would soon enough be subject to wide¬
spread attack. During the early postwar years it was common to con¬
flate fascist and communist aims (Adler and Paterson 1970). As Janson
lO A Game for Dancers

connected regionalist art with nationalism and easily manipulated mass


society, so Gifford made similar connections among Americana dance,
nationalism, leftist political propaganda, and the culture industry. For
Gifford, it was the dancer’s task to keep art independent of both politi¬
cal and commercial appropriation.
Another dance commentator, Margaret Lloyd, also linked inter¬
nationalism and modern dance in a revealing fashion when she wrote
that American modern dance “becomes more international, with no
loss of patriotism, as it becomes more liberal, as it leaves the confines
of early modernism, narrow national and class consciousness, for the
timeless, universal realm of dance, and the one world of people” (1949:
186). Lloyd’s remarks, published in 1949, stressed the idea that modern
dance had been narrow in focusing on national and class identity. She
included in this narrow category the socially conscious work of dancers
such as Sophie Maslow and Jane Dudley. In place of these interests,
Lloyd called for a dance that would be timeless and universal. She at¬
tached the word “liberal” to such a dance, presumably because it re¬
flected “the one world of people.” Thus, she linked internationalism,
modernism, and liberalisni. She also equated nation and class with the
specific, and internationalism with the universal, which accorded with
American postwar aspirations.

During 1945 and 1946, President Truman had not yet solidified his for¬
eign policy; in that time, as has been seen in the responses to Lippin-
cott’s article, at least some dancers were still able to conceive of an art
that had a direct relationship to politics. But by mid-1947, the Truman
administration had put a policy of communist containment into place,
and the Cold War had hardened into reality (Lacey 1990; Hamby 1976;
Whitfield 1991; Chafe 1999). It put the United States once again on a
wartime track, and with it came an increased centralization of power.
The FBI greatly expanded its sphere of influence in the immediate
postwar years. Truman feared J. Edgar Hoover and was unable to do
much to curb him. The Republicans took both houses of Congress in
the 1946 elections, bringing Richard Nixon and Joseph McCarthy to
Washington and helping to set the stage for the anticommunist witch
hunts that would soon follow. By mid-1946, a number of leftists had
joined the anticommunist ranks, including prominent union leaders
and such intellectuals as Irving Howe and Arthur Schlesinger Jr. Early
The Trouble with Modern Dance II

1947 brought Executive Order 9,835, intended to keep the civil service
free of “disloyal” elements. It oudawed, among other things, subversive
organizations or those associating with them, without specifying which
organizations fell into this category.
Several liberal publications protested vigorously: an unsigned edito¬
rial in the Nation stated that Truman’s executive order “reflects a fear of
totalitarian communism that is driving us to use totalitarian methods”
(“The Shape of Things” 1947: 346). Further stories in the Nation bore
such titles as “Liberals Beware!” (5 April 1947), “Washington Witch-
Hunt” (5 April 1947), and “Warning All Scientists” (2 August 1947),
which told of the dangerous turn taken in Washington and the victims
of various purges. The cover of the ii August 1947 issue of the New
Republic had subversive! stamped across such names as Thomas Jeffer¬
son, Abraham Lincoln, and Oliver Wendell Holmes. The cover head¬
lined a report on the state of civil liberties in the country. Numerous
articles followed, among them “Cry Shame . .. !” by Martha Gellhorn,
which chronicled the goings-on at HU AC (1947), and “Guilt by Gos¬
sip” by Daniel S. Gillmor (1948), which deplored the witch hunting and
atmosphere of paranoia that had swept far beyond the borders of Wash¬
ington. Considering the political climate in the late 1940s, it is not sur¬
prising that dancers were less optimistic than they had been even a year
or two earher about art’s ability to bring about social and political change,
at least through confrontation.
Another event that influenced dancers’ views of a political art came
in early 1948 when word reached the United States that officials of the
Soviet Music Union had denounced Russia’s most respected composers,
charging them with formalist tendencies. Among those named were
Dimitri Shostakovich, Aram Khatchaturian, and Sergei Prokofiev.
Alexander Werth, reporting in the Nation, quoted a Music Union offi¬
cial who had defined formalism as “the lack of ideas, lack of content,
emptiness of form, without reference to reality” (1948b: 209). In this
report and an earlier one on 10 April (1948a: 393), Werth noted that the
Soviets wished to have an art that was connected to real life, that was
clearly understandable to the masses, and that was in no way “pure” or
abstract, quahties that appealed only to aesthetic elites (ibid.). The answer
was socialist realism, which would be the only sanctioned approach to art
behind the Iron Curtain. This attack on formalism was, by extension,
an attack on the United States, where modernism (construed in Soviet
terms as formalism) had been defended since the rise of fascism and
12 A Game for Dancers

where New York was in the process of becoming the capital of the mod¬
ernist advanced guard.
The politicizing of modernism in this manner was another element
of the escalating Cold War, and a number of artists quickly rose to
defend the American Way. Gertrude Lippincott wrote an editorial for
Dance Observer entitled “Freedom and the Arts” (1948: 44-45) that en¬
capsulated this viewpoint. She began by saying that dancers had long
worked to promote the idea of arts subsidies and had looked with envy
on other countries that provided them. Many were stunned, she said,
when the fascists had taken over the arts in Germany and Italy and used
artistic organizations and activities for propaganda purposes. Ameri¬
cans had hoped for more from the Soviet Union, but that hope had
been shaken in 1936 when Shostakovitch had been censured. Now all
remaining hope had vanished with news of the newest denunciations.
Lippincott disingenuously found it difficult to understand how polit¬
ical characteristics could be ascribed to art in the first place, when true
art could not be political because “it transcends national and political
boundaries and appeals to all mankind” (ibid.: 45). Lippincott was most
shocked by the Soviet composers’ public confessions and their recant¬
ing of modernism. Since they were not allowed to leave Russia, they
were, she said, virtually artist-slaves who functioned at the will of tlieir
masters. Then she argued that although American artists were ignored,
they were free to produce what they wished. Their neglect “may be a
blessing in disguise, because no artist can ever serve any master other
than himself” (ibid.).
Lippincott’s editorial accomplished several things. It summarized the
modernist viewpoint that authentic art dealt with essences and univer-
sals that appealed to all humanity rather than being tied to specific ideas
and situations; it presented the official American position that Western
art was free and Soviet art was not; and it made the case that American
neglect of the arts was actually positive. American dance, then, should
be nonpolitical in content to show that it was free, and it should not
demand political action to ensure its place in the social sphere because
neglect also kept it free. That a dance that did not deal directly with
political subject matter could in itself indicate a political position, that
it supported Washington’s desire for a critique-free art and had the
added bonus of asking nothing of Washington in return, escaped her
notice. In their concern to be free from control, dancers in the postwar
years rarely asked what they were free to do.
The Trouble with Modern Dance 13

The result of all these pressures on dance from both the domestic
and international arenas was that the political element of the debates
surrounding commerce and artistic freedom almost disappeared by
mid-1948. However, while external conditions contributed to dancers’
sense of powerlessness to effect social change through overt con¬
frontation, and an acceleration of the Cold War further froze discussion
of a political dance, the internal problem of modern dance’s instim-
tionalization remained an immediate, crucial issue, and the argument
over how best to ensure modern dance’s future continued even as talk
of the pros and cons of a political dance faded.

One of the major points of Gifford’s 1945 article was that in order to
be more accessible, young modern dancers were not revolting against
the older generation but rather were imitating it. Gifford differentiated
between what I call the intergeneration of dancers, such as Maslow,
Sokolow, and Dudley, who began their careers in the 1930s and contin¬
ued to do socially conscious work (some of these dancers will be dis¬
cussed in chapter 4), and a second generation of artists who came of age
in the mid-1940s. The evidence suggests that Gifford was correct in his
concerns about this younger generation and that much of the imitation
he perceived in their work was related to the direction in which Martha
Graham was moving during the 1940s. Graham, the most consecrated
of the first-generation pioneers, was turning toward an ever more the¬
atricalized dance. Her concerts now included sophisticated lighting,
costumes, and props, as well as orchestral music rather than the piano and
percussion scores of her early career. As for her dances, they had begun
to resemble ballets in their length and complexity. This trend stood in
marked contrast to her early dances, which had been notable for their
austerity. Now Graham even arranged her programs like ballet perform¬
ances, usually with three works to a program separated by two inter¬
missions. In addition, she made her technique more complex and virm-
osic by adding balletic elements to it. She also began to tell stories in
some of her dances. This trend led to the descriptive title “the theater
of Martha Graham,” implying her work was less about dance than it had
been in the past and more about an integrated theatrical experience.
Although Graham’s move into theater was the result of a number of
factors, it is worth noting that she, like so many other modern dancers,
wrestled with the problem of vanguardism and accessibility. An imsigned
14 A Game for Dancers

“The Theater of Martha Graham,” a photograph taken in the early 1960s of Martha Graham
as Jocasta, Bertram Ross as Oedipus, and company in Graham’s Night Journey (1947). Photo:
© Martha Swope.

Newsweek article published on 17 January 1944 under the headline


“Doom Eager Dance” mentioned that three years earlier Graham had
promised a new kind of work that would differ from the more austere
dance she had offered in the 1930s. The magazine quoted her as saying,
“We must win back our audiences. We have alienated them through
grimness of theme and a non-theatrical approach.. .. We must prove
that our theatre pieces have color, warmth, and entertainment value”
(“Doom Eager Dance” 1944: 85). Yet while Graham was taking a stand
for greater accessibility, she continued to draw on her status as a van¬
guard rebel, a status she had earned from her years of struggle and that
had led to her consecration. Eor example, she was quoted in a Life mag¬
azine article of 17 March 1947 as saying: “Some people like me, some
don’t. Many consider me a menace” (“She Is Priestess of Intellectual
Ballet” 1947: loi). Although audiences were likely to find Graham far
less a threat to society than they had a decade earlier, a hint of danger
could be scintillating.
The Trouble with Modern Dance 15

Among the changes that Graham helped bring to modern dance in


the 1940s was a new attitude toward ballet technique. Ballet had hith¬
erto been anathema to modern dance since it was assumed to be at odds
with modern dance’s mission of creating innovative forms that could
confront the issues of modernity. When Graham began to develop her
modern dance in the 1920s, she, like the other first-generation dancers,
conceived her own technique in opposition to ballet. But in 1938, when
Erick Hawkins joined her company, he arrived as a ballet dancer, hav¬
ing come from Lincoln Kirstein’s Ballet Caravan. In the early 1940s
he taught ballet classes at the Graham studio, and in time Graham’s
dancers noticed that her technique was becoming more balletic. Jean
Erdman spoke of the addition of new movements that used ballet mrn-
out and an immobile torso, particularly in turns. These, she said, pre¬
sented problems of balancing and centering without impulse from the
center, which were difficult for the dancers (Horosko 1991: 78-79; see
also Helpern 1991: 22). When Merce Cunningham joined the company
in 1939 as the second male, Graham suggested he take ballet classes at
the School of American Ballet, and Cunningham did. Other dancers also
took ballet classes outside the Graham school on their own.
In the 1940s, Graham also began to hire dancers with bodies that
were lighter and more balletic than the dancers she had employed in
the 1930s, such as Sophie Maslow, Anna Sokolow, and Jane Dudley.
Mark Ryder noted the difference these new bodies made in the dance
movement when he remrned from the war to rejoin the company. “What
had been slow and powerful became lightning fast, like quicksilver,” he
said (Horosko 1991: 81). Critic Margaret Lloyd agreed. She contrasted
the 1936 performances of Primitive Mysteries to the 1947 revival. In the
’30s, she said, dancers like Dudley, Maslow, Sokolow, May O’Donnell,
and Gertrude Shurr “imparted the virility, or ‘guttiness,’ the work de¬
mands” (1949: 53). Now, she said, the piece was not nearly so well per¬
formed, in part because the girls were “thin.” “They are young and
attractive; they are so light they could be mistaken for ballet dancers”
(ibid.). In Lloyd’s view, they lacked “the full-blooded vigor of the con¬
cert group of the mid-thirties” (ibid.). These new dancers not only gave
the company a lighter, more balletic appearance, their increased range
of movement made it possible for Graham to add more complexity
and technical virtuosity to the dances. Graham may well have seen this
as a way of enriching her choreography, but at the same time it also
allayed some of the grimness of modern dance she had spoken of in the
i6 A Game for Dancers

Newsweek article, and of adding “entertainment value.” Although Gra¬


ham was not alone in absorbing ballet elements into her technique, she
was a leader in that trend.
The majority of young dancers who came of age in the mid-1940s
did not question Graham’s move toward ballet or any of the other ele¬
ments of her new theatricalism. Young dancers generally had no access
to major theaters, and they had no funding to moimt complicated pro¬
ductions. They did not have companies, but only small, unpaid groups
that met occasionally to put together a concert. Yet instead of turning
these apparent drawbacks to advantages by casting themselves as purists
and the older generation as sell-outs to “bourgeois” demands for en¬
tertainment, something like the opposite occurred; they attempted to
compete with the new theatricalism.
First, they expanded technique by insisting on the right to study
and use any technical language that suited a particular dance, including
ballet. Jean Erdman, who danced in the Graham company from 1938
to 1942 before becoming a choreographer in her own right, reflected
the general trend among second-generation dancers when she wrote
in 1948 that, as opposed to their elders, young choreographers were
less interested in forming personal vocabularies and styles. Nor did
they want to devote themselves to only one modern dance technique.
Rather, they preferred to study a variety of techniques, including ballet
and ethnic forms, and to draw on whatever sort of movement was ap¬
propriate for each individual dance (Erdman 1948: 40-41).*^ By using
varied techniques, the young modern dancer sought to depersonalize
and objectify what for the first generation had been a highly personal
movement language. This was a significant shift in goals from that of
prewar dancers.
Critics contributed to this shift by attacking dancers for transferring
their personal styles to movement for groups. This transfer occurred
despite the fact that the dancers themselves were attempting to deper¬
sonalize their dance vocabularies. Nik Krevitsky’s review of Iris Mabry’s
Counterpoint was typical: “It is nothing more than an extension of Miss
Mabry’s personal style, which doesn’t fit a group of sixteen dancers.
Movements and gestures which become her, just don’t suit others”
(1948: 57). And Mary Phelps ended her review of a Valerie Bettis con¬
cert with: “So with Valerie Bettis’ problem of extending her talent from
the solo to the group field we have about the status quo ante. Group
work cannot be gone at as multiplication of the soloist’s impulse. It has
The Trouble with Modern Dance 17

to be done by true addition: new substance of movement when more


figures are concerned” (1948: 19). These reviews illustrate the confu¬
sion of goals that was evident within modern dance. Modern dance de¬
manded new forms that in the past had been developed through per¬
sonal vocabularies and styles, but now dancers were criticized for not
being able to depersonalize movement to fit more ambitious group
works. At the same time, dancers were also attempting to objectify their
movement.
In addition, many of the second-generation dancers studied compo¬
sition with Louis Horst or Doris Humphrey. Although the courses gave
the dancers structural principles with which to work that the pioneer¬
ing generation of the 1930s had needed to develop for themselves, they
also tended to stifle experimentation. No longer did the student have
to ask what an appropriate subject or form might be; she was taught it.
Horst’s classes in particular stressed specific kinds of dances. Many of
the solos that were performed at concerts resembled dances composed
for Horst’s courses in preclassic and modern forms—galliards and min¬
uets, primitive and medieval studies. Solos in this vein were often first
cousins of classroom exercises, a situation that critics increasingly com¬
plained of as the postwar years wore on. One has only to look at Trans-
fotmations of Medusa (1942), an early dance by Jean Erdman, to under¬
stand the degree to which academic forms found their way onto the
stage. Erdman has said that the three-part solo was inspired by her
classes with Horst. This dance has the static quality of a well-executed
exercise in Horst’s archaic form.
In a review of the field in 1953, John Martin wrote of the results of
the dangerous trend toward accessibility that Gifford had warned of ten
years earlier. Martin noted that modern dance “has forgotten its reason
for being in its efforts to take on the forms of theatrical effectiveness”
(1953: 3; see also 1955b). He went on to say that because young dancers
merely executed movement rather than experiencing it, “the result is a
kind of ersatz ballet, pieced together from ‘Graham technique,’ ‘Holm
technique,’ ‘Limon technique’ (all of them utterly without significance
when separated from their context) and lacking both the impersonality
and the traditionalism which make the genuine ballet classic.” His list
continued with laments about the quality of the dances presented. “The
study of composition is admirable and necessary, but it does not auto¬
matically produce works of art. How many of the young dancers are
simply presenting us with their classroom exercises in composition!”
i8 A Game for Dancers

(ibid.). As a solution to these problems, he called for a return to the


basic principles on which modern dance had been built.
Certainly by the mid-1940s, many second-generation dancers were
working at cross-purposes with the aims that had put modern dance in
a vanguard position in the interwar period. They attempted to exchange
personal languages for impersonal ones, which undermined the mod¬
ernist goal of developing new forms for a new age and which also served
to differentiate individual dancers from their rivals. Yet at the same time
they learned to create movement on their own bodies, which hindered
them from successfully choreographing group dances. Then, through
the institutionalizing of composition, they learned to make dances in a
way that hindered experimentation, the means through which challenges
to the old guard were issued. And that wasn’t all. Among the most dam¬
aging trends in 1940s modern dance was an increasing use of narrative
(in the sense of plot lines or the detailing of specific sequences of events)
at the expense of plotless form.

Narrative was problematic for several reasons and as such became a


highly contentious issue for dancers and critics as the 1940s unfolded.
First, it brought modern dance closer to traditional forms of ballet and
drama, in theory making it more understandable and thus accessible to
larger audiences. In this sense it suggested commercialism. Narrative
also countered important modernist principles that left modern dance
open to attack on a number of fronts. As noted in my introduction, T. J.
Clark has described modernism as centering on an uneasiness or doubt
about the nature of representation in art (Clark 1984; 10). In a mod¬
ernist dance this doubt was most clearly manifested in a demand to em¬
body rather than represent meaning. This is the view expressed in
Martha Graham’s well-known statement: “Why should an arm try to be
corn; why should a hand try to be rain? Think of what a wonderful thing
the hand is, and what vast potentialities of movement it has as a hand and
not as a poor imitation of something else” (Graham [1937] 1985: 107),
A modernist dance was authentic to the degree that it embodied rather
than imitated or represented experience. Narrative, whether in the form
of plot or pantomime, was representation rather than embodiment.
Modern dance’s original goals had espoused an autonomous art that
placed embodied movement, as dance’s essential element, at its center.
Communication was crucial, but it was communication through ab-
The Trouble with Modern Dance 19

stracted corporeal movement, not through literary, that is, representa¬


tional means. John Martin wrote that it was in movement that dance’s
power lay: “With this discovery the dance became for the first time
an independent art—an absolute art, as they like to say in Germany—
completely self-contained, related directly to life, subject to infinite
variety” ([1933] 1972: 6). Although movement may seem the obvious
basis of dance, Martin argued that ballet made movement incidental,
favoring instead the poses between movement. Ballet did not treat
movement as a unified entity but rather as a series of distinct elements
that were assembled like a language to create meaning. Modern dance
brought attention back to dance’s essential component and in doing so
once again unified movement and meaning (ibid.: 6-7).
The drive to keep literary elements at bay in modern dance was em¬
phasized repeatedly in the 1930s. Two of Graham’s most quoted re¬
marks were: “If you can write the story of our dance, it is a literary thing
but not dancing,” and “My dancing is just dancing. It is not an attempt
to interpret life in a literary sense” ([1937] 1985: 102). While modern
dance’s avoidance of language reflected its desire to eliminate depen¬
dence on other art forms, which was necessary to its modernist goals,
there were even deeper motives for distrust. Modernists also perceived
language as dangerous to art because it could control through its power
to define. Pierre Bourdieu notes in his analysis of nineteenth-cenmry
French modernist painting that artists fought to make art “irreducible
to any gloss or exegesis” so it could not be appropriated by writers or
writing. “It is a strategy which consists of denouncing and methodically
thwarting—in the conception and the very strucmre of the work, but
also in an anticipated metadiscourse (the obscure and disconcerting
title) or in a retrospective commentary—any attempt at annexation of
the work by discourse” (1996:137). Martin reflected this viewpoint when
he pointed out that the task of the artist was to express irrationally what
others would later rationalize in writing. So strong was his sense of the
need to keep dance free from the dominating power of language that he
added: “It is safe to say that when any art form has got itself to the point
where it can be translated into words, it is dead as an art form” ([1933]
1972: ii). Modernists generally distrusted language and its power to
define meaning, largely because they associated language with ration¬
alized thought, which, in turn, was linked to the market and thus to
ruling-order ideology. Postwar vanguard dancers were consumed with
the dangers that language posed to the genre’s independence, not just
20 A Game for Dancers

from the other arts, but also from the field of power with its weapons to
control and consume. Yet as the 1940s advanced, modern dance works
increasingly told stories, and Graham, as the leading exponent of mod¬
ern dance, was in part responsible for it.
In the interwar years, Graham’s works had generally conformed to
the rules of modernism she espoused. For example, American Provincials
of 1934 consisted of two parts, a solo for Graham, “Act of Piety, and a
group dance, “Act of Judgment,” which pitted Graham against the
group. Although critics saw the work as an attack on the forces of Pu¬
ritanism, the piece included no narrative story line nor program notes
to explain it (McDonagh 1973: 102). Other works were treated similarly:
Perspectives (1935) consisted of two independent studies. Perspective
No. i: Frontier (Graham’s famous solo) and Perspective No. 2: Marching
Song. Panorama (1935) was made up of dances entitled “Theme of Ded¬
ication,” “Imperial Theme,” and “Popular Theme.” In all these dances
subject matter was kept generalized.
However, in the 1940s, Graham added to her repertory a number of
works with much more specific topics. By 1945 she had composed such
ballets (as she herself called them) as Letter to the World (1940), based on
the life and poetry of Emily Dickinson, and Deaths and Entrances (1943),
inspired by the lives of the Bronte sisters. Between 1946 and 1948,
Graham produced some of her best-known works, several of which dealt
with Greek mythological themes. These included Cave of the Heart
(1946), based on the story of Medea, and Night Journey (1947), which
concerned the Oedipus myth. When she first presented Cave of the
Heart she called the protaganist One Like Medea, then changed it to
the more generalized Sorceress. Night Journey retained the names of the
characters in Sophocles’ play. During the 1940s, Graham did not aban¬
don plotless subject matter. Dark Meadow (1946), for example, depicted
a ritualized event that was not made specific, and Diversion of Angels
(1948), which she made for the company and in which she did not ap¬
pear, was entirely without a story line. Nor did Graham stop dealing
in metaphor and ambiguity in her narrative pieces; these elements re¬
mained crucial in all her work. However, as the 1940s advanced and
Graham became more involved in dealing with literary themes, she
made dances that incorporated far more mimetic gesture than she had
in the past, in order to support the stories she was telling.
A comparison of a solo from the 1930s, Frontier (1935), and one
from the 1940S, the opening dance in Night Journey (1947), provide ex-
The Trouble with Modern Dance 21

amples of the shifts that took place in Graham’s work. Frontier is plot¬
less, and Graham did not provide program notes that told viewers what
she was trying to do or what she wanted them to think about the dance.
The movement for Frontier is not specific. It consists primarily of ges¬
tures in which the dancer stretches her arms in great arcs and of steps
in which she sweeps one leg high to the side as she moves forward, or
executes big skips that take her forward and back from the railing up¬
stage center that forms a locus for her movement. At one moment she
bends her arms and joins her hands in front of her while moving her
arms from side to side. This could indicate any number of things, from
the rocking of a baby to the motion of a wagon heading west, or it could
be simply a gesture that for Graham communicated something about
the way one moved in, or was moved by, the West. Graham’s alert posture
and the largeness and openness of her movement may convey attitudes
about the frontier, but these are movement metaphors that suggest
without becoming specific. Throughout the dance Graham kept mean¬
ing open, full of multiple possibilities that the spectator resolved or al¬
lowed to stand as he or she wished.
The opening dance of Night Jomyiey is considerably different. Here
Graham, as Jocasta, is seen resisting a confrontation with her past. We
know this is the situation because her written program notes have told
us and because Graham executes certain pantomimed gestures in the
dance that reinforce the written text. For example, Jocasta approaches
the wedding bed and recoils in horror. She repeatedly touches her head,
a gesture both in daily life and in the theater that indicates thought. She
repeatedly touches her breasts and pubic area to indicate the sexuality
that is the source of her problem (also revealed in the program notes).
This is not to say that the dance includes no abstract movement. There
are steps that would qualify as “pure” movement as well as others that
are suggestive rather than specific. Certainly, Graham did not give up
metaphoric movement in the 1940s, but she did make meaning more
explicit in some of her works both through movement and through the
narratives that she chose to relate.
The fact that Graham’s dances dealt with specific characters and nar¬
ratives in a way they had not done before the war was certainly evident
to contemporary critics, who differentiated between Graham’s former
“pure dance” and the new “theater of Martha Graham.” Cecil Smith,
writing in Theatre Arts, commented on the changes in Graham’s work
in a 1947 season at the Ziegfeld Theatre: “Not a single dance composed
Martha Graham in Frontier (1935). Photo: Edivm'd Moeller. Courtesy of Jerome Robbins Dance Division,
The Netv York Public Libraty for the Peifonning Arts, Astor, Lenox and Tilden Foundations.
Martha Graham as Jocasta in Night Journey (1947). Photo: Arnold Eagle. Courtesy of Jerome Robbins
Dance Division, The New York Public Library for the Performing Arts, Astor, Lenox and Tilden Foundations.
A Game for Dancers

between 1931 and 1939, the date of Evefy Soul is u Civeus, was given a
place in her schedule. Clearly she now regards herself as a theatre artist,
and does not feel that the early pieces—many of which were magnifi¬
cent examples of ‘pure dance’—are suited to her present following
(1947: 29). Smith, however, did not use this shift toward theater to at¬
tack Graham; rather, he defended her theater work as a new develop¬
ment in vanguard dance. He wrote that although Graham’s dances were
narrative, they did not deal with outer reality but with the “veiled and
perplexed world of the inner consciousness” (ibid.; 30). She conveyed
this world, he said, with pure movement, an evocative prop or two, and
costumes, and in doing so transfigured the turmoil of personal emo¬
tions. Thus, according to Smith, Graham accomplished the task of a
modern art.
Critic Margaret Lloyd was even more intent on stressing the van¬
guard nature of Graham’s new dance, which, she said, put pure move¬
ment at its center. She wrote that in Graham’s theater, “movement is
still paramount, in spite of remote literary, mythological, or pantomimic
references” (Lloyd 1949: 64). However, as much as supporters defended
Graham’s new theatricalism, they nonetheless perceived it as moving
away from abstraction rather than toward it, a step that put a vanguard
genre at risk. It was on this point—the use of narrative as an element
foreign to dance—that several generations of dancers would attack Gra¬
ham in the 1950s and ’60s. However, in the 1940s, Graham’s dances
were still seen as vanguard productions that in their concern with myth
and the unconscious revealed universal aspects of human existence.
Where Graham led, others followed, and many did not have Gra¬
ham’s nuanced approach to literature or the complexity of her vision.
For example, Valerie Bettis choreographed a work entitled Daisy Lee
(1944) in which she danced out a story of “the frantic remorse of a
woman who destroyed not only her husband’s but also her sister’s life
through her own pride, possessiveness and emotional dishonesty” (Sabin
1945a: 70). While Bettis performed the central role, voices offstage rep¬
resented the other characters. When the work premiered in 1944, it was
listed as a dance-play with a libretto by Horton Foote. In another dance
Bettis drew on sections of William Faulkner’s novel As I Lay Dying (1948)
in which she interspersed portions of spoken text with movement to
create portraits of two southern women. The characters in the piece in¬
cluded nearly a dozen members of the Bundren family and their friends
and acquaintances. Bettis performed some of her works under the aus-
The Trouble with Modem Dance 25

pices of the Choreographers’ Workshop, which billed itself as “inter¬


ested in presenting not only pure dance, but works of theatre that in¬
clude a dance viewpoint and works of dance that include a theatre view¬
point.”^ There were many other examples of dance dramas at this time,
from William Bales’s (1949), described as a “theatre piece based
on the story of Judith and Holofernes” (Krevitsky 1949: 98) to Jean
Erdman’s The Fair Eccentric, or The Temporary Belle of Hangtown (1951),
which, according to one critic, approximated the oleos of vaudeville and
included extended sections of pantomime (Krevitsky 1951b: 42). In
other instances, dancers simply quoted a poem or story in a program
note, and then danced it.
Since modernism required artists to produce new forms, it is not
surprising that a number of second-generation choreographers wished
to develop dance and drama into a new amalgam. However, they had
difficulty reconcihng the antiliteral aspects of modernism with narra¬
tive. Valerie Bettis attempted to address the problem by dividing what
she called theater dance into two types, one based on plot, which was
developed narratively, the other on an emotional idea, which was de¬
veloped thematically. “It must be realized that only in the former [nar¬
rative dance] can theatrical devices such as literal costuming, special
props, definitive sets, and stylized acting or mime be used freely, if in¬
telligently” (1945: 82). In addition to theater dance, with its narrative
and thematic forms, there was “pure” dance. Bettis continued: “Free
dance movement must be studied and understood; and it is the dancer’s
obligation to distinguish for his audience between the theatre dance, in
which the idea is developed thematically, and the ‘pure’ dance, both
of which may employ this kind of movement” (ibid.). This not alto¬
gether lucid explanation suggests that Bettis had difficulties defining
how the various forms of pure, thematic, and narrative dance related to
each other.
By 1948, Nina Fonaroff was better able to argue the case for dance
drama (1948: 52). She began by stating that pure dance was theatrical
because acting and pure movement came from the same root, that is,
an emotional source. There was a difference, she said, between sensory
movement and explanatory movement. Pantomime was explanatory
movement and so was ballet, which, at least in Fonaroff’s view, explained
itself through costumes, scenery, and mimetic gesture. After mention¬
ing several actors who incorporated pure movement into their work,
such as Jean-Louis Barrault and Faurence Olivier, she went on to say
Valerie Bettis and Duncan Noble in Bettis’s Daisy Lee (1944). Photo: Tripp atid Delknback. Courtesy
of Jerome Robbins Dance Division, The New York Public Library far the Petfarming Arts, Astor, Lenox and
Tilden Foundations.
The Trouble with Modern Dance 27

that actors who used pure movement distorted drama beyond the literal
to reveal its essence: “When the intention demands extremities; when
the drama is the epitome of what it is about—then its working material
can, by distortion rise to that epitome and fulfill the furthest possibil¬
ities of the mediums, movement approaching dance, and speech, po¬
etry. But if the intention of the play stops at the naturalistic, the repre¬
sentational, there is no call for distortion” (1948: 52). Pure movement
was what made it possible to reveal essential elements of drama. When
a choreographer used words, she had to be sure the dance did not merely
illustrate the text. Rather, she had to create a structure that needed both
while at the same time keeping them separate. Fonaroff advocated
using a common rhythmic structure that would allow each medium to
remain independent. Conceding that the vast majority of dance theater
conformed to word illustration, Fonaroff nonetheless said she would
devote herself to developing a form that would allow for the autonomy
of both media.
A year later, Katherine Litz made a similar argument regarding the
common emotional source shared by dance and drama. She suggested
that dancers study acting because “every movement, no matter how
arbitrary, can be given dramatic quality through an understanding of
emotional values” (1949: 67). The dancer who studied acting came to a
“deeper understanding of the dramatic basis of movement” (ibid.). In
this statement Litz, like Fonaroff, argued that dance had a dramatic
base; accordingly dance was not opposed to drama but related to it.
However, as much as dancers attempted to combine two existing
forms into a new one, both on paper and on the stage, they were ulti¬
mately working against modernist prescriptions, and as such they put
the vanguard status of modern dance at risk. This held true despite the
fact that in some cases they succeeded in producing works that merged
drama and dance with marked success. One of the best examples was
Jose Limon’s The Moores Pavane (1949), which became a signamre work
for the Limon company. Limon had been a leading dancer with the
Humphrey-Weidman group since the early 1930s. Humphrey-Weidman
disbanded in 1945 while Limon was serving in the army. After his dis¬
charge he decided to form his own company, and Humphrey agreed to
become artistic director. Although Limon’s earlier choreography had
not received a great deal of praise, once under Humphrey’s tutelage, his
work became stronger, and he soon emerged as the great hope of the
postwar generation. Critics from John Martin to Doris Hering to the
28 A Game for Dancers

Dance Observer writers looked to Limon as the best chance of carrying


on the work begun by the first-generation pioneers.
The Moor’s Pavane, based on Shakespeare’s Othello and set to music by
Purcell, premiered in 1949.^ Limon reduced the number of characters
in the plot to four, telling in succinct fashion the story of Othello’s jeal¬
ous love and the tragedy to which his unbridled emotions led.^ Like
Graham, Limon had to use pantomime to tell his complicated story, but
he also enclosed the drama within a framework of courtly dances, which
controlled narrative and constantly led the viewer’s attention back to
the dance itself. The work opens with the four characters in a circle, their
arms held high, hands joined. As the dance progresses they circle, bow,
advance and retreat, in the formal patterns of the dance. These pat¬
terns, in addition to being “pure” dance movement, also suggest the
characters’ interconnected relationships. At the same time, the dance
also incorporates within its overall structure many of the gestures and
movements that will typify the characters’ relationships in more specific
ways; Othello and lago lock their arms together in silent struggle; Des-
demona holds out her handkerchief; Emilia and lago hover closely
and give each other meaningful glances as they watch Desdemona and
Othello embrace. The next dance deals with the relationship between
lago and Othello. lago clings to and wraps himself around Othello,
while Othello wards him off with large, impetuous gestures. The dance
ends with lago executing a series of low, cringing bows. And so the work
progresses, through dances for lago and Emilia, Othello and Desde¬
mona, lago and Othello, until the final dance in which Desdemona’s
murder is hidden behind Emilia’s voluminous skirt. The piece ends with
a tableau of the four dancers, lago and Emilia framing Othello, who is
seen stretched across Desdemona’s body.
The Moor’s Pavane interweaves pantomime, metaphoric movement,
and the formal designs of court dances with unusual finesse. Nonethe¬
less, Limon’s choreography depended on mimetic gesture to a degree
seldom seen in the pioneering modern dance of the 1920s and ’30s. As
Marcia Siegel has noted, Limon’s gift was “to explain, to dramatize, to
become a hero for a while and play out that hero’s role” (1979: 168).
The desire to specify and explain, successful as it may have been in a work
like The Moor’s Pavane, was a problem for modern dance, a problem that
caused a clash that reverberated through the dances and debates of the
period.
Joseph Campbell, a respected scholar of mythology, sounded the
From left, Lucas Hoving, Jose Limon, and Pauline Koner in Limon’s The Moor’s Pavane
(1949). Photo: Gerda Peterich. Courtesy of the Jerome Robbins Dance Division, The New York Public Library for
the Peifarming Ans, Astor, Lenox and Tilden Foundations.
30 A Game for Dancers

alarm in articles in Dance Observer in 1944 and 1945. Campbell de¬


plored the trend toward literary means in modern dance, writing that
“ ... unhappily, the entire Modern Movement is going over to little
plays, and to a horrendous bastard-art of poem-dancing” (ip/pfa: 31).
Dance was not a discursive art but an embodied one, he argued;

The dancer, that is to say, is not a semaphorist, but a work of art in the
flesh; her function is not the [hr] flash messages back and forth from brain
to brain (that is the role of the discoursive {sic\ paragraph), but to embody
Significant Form. And what is Significant Form? It is the rhythm of life
projected in design; the invisible pattern of the psyche reflected in time
and space; a profoundly inspired disposition of feeling-charged materials
stemming from, and addressed to, that creative center where human
consciousness and the unconscious fruitfully touch. (1944b: 66)

In making this argument, Campbell, like Martin a decade earlier (Mar¬


tin [1933] 1972: lo-ii), was pointing out that modern dance’s purpose
was to embody those hidden realities that language could not convey.
The ability to embody the human experience that words could not
touch was dance’s unique strength, and modern dance was in the pro¬
cess of giving it away.
Campbell blamed the literary trend on Graham’s dance-dramas, but
explained that the influence was ironic, as Graham had actually devel¬
oped a new dance form. Her works were legitimate because she did not
attempt to illustrate stories but to inhabit “the timeless realm of dream”
that was linked to the unconscious. In addition, Campbell said, Graham
depicted her characters as archet3q)es, characteristic of certain psycho¬
logical states and situations, which in Campbell’s view gave them uni¬
versal significance. Both Campbell and Graham were followers of Carl
Jung, and as such accepted archetypes as the content of the collective
unconscious (Jung 1971: 59-60). Campbell, like other critics, was loathe
to hold Graham in any degree accountable for the direction that mod¬
ern dance was taking. It must be remembered, too, that he and Graham
were close friends, and he was writing in a magazine that wholly sup¬
ported her work.
The need to commrmicate was a liability for modern dance because
it was always in danger of illustrating or imitating rather than embody¬
ing authentic states of mind through movement alone. Campbell began
another article the next year by making this point and attempting to
The Trouble with Modern Dance 31

provide an answer through myth, which the surrealists, those certified


avant-gardists, had also used as a path to the unconscious. Myth and
legend, Campbell wrote, provided archetypal images that were univer¬
sal and that were keys to the unconscious (1945: 52-53). Citing Freud,
Jtmg, and the mythologist Stith Thompson, he asserted that the episodes
and personages of myth, legend, fairytale, and fable were timeless and
essential. For every aesthetic element there was a psychological one
identical with a psychological value of a mythological element. If the
dancer could fully understand the implications of the images and ren¬
der them with appropriate aspects of her craft, she could successfully
fuse form and content. What was necessary was to find the correspon¬
dences between the psychological effects of the aesthetic form being
used and those of the archetypes of myth and symbol. Campbell had no
advice on how the dancer was to locate those links; that apparendy was
her task. However, in his argument he attempted to provide an answer
to modern dance’s predicament through means that actively sought to
maintain art’s independence from causal logic by plumbing the uncon¬
scious and bringing authentic experience to the surface.
Campbell’s attempts to offer at least one way out of trouble for
modern dance only served to point up the problem, and arguments
continued well into the 1950s over whether or not dancers should em¬
ploy literary means. These debates had much to do with dancers’ am¬
bivalence about the field’s relationship to the market and hence the
culture industry. Narrative made dance more accessible to spectators,
which meant, for some, larger audiences and greater success. As such,
narrative became an element in modern dance’s assimilation into the
mainstream, because, as Gertrude Lippincott and others argued, van¬
guard art was aimed not at a broad public but at a small group of peers
who chose to ignore the market.
Writers’ views on the use of literary devices in modern dance were as
complicated as dancers’. Some commentators were outright supporters,
as were Bernice J. Wolfson (1947: 88-89) Winthrop Palmer (1949:
112), who wrote articles tracing a history of literary means used in
dance, and which were intended to legitimize the dual form. On the
other hand, there were also numerous articles that supported the pure
dance viewpoint, and critics as a whole subscribed to the idea that de¬
tailed storytelling was not good for modern dance. One of the ways they
manifested their antinarrative stance was to brand works they felt illus¬
trated a story as literal. For example, Nik Krevitsky wrote that Patricia
32 A Game for Dancers

Newman’s Green Mansions was “a lengthy, literal interpretation of


Hudson’s beautiful novel. . . and when it ended with Abel’s anguish,
there was a feeling of rejoicing in that it was over” (1948: 57). Yet, in
fact, critical response to narrative was largely arbitrary because once the
idea of a theater-dance had been accepted as a new form, critics then
had to distinguish between dramas that appeared literal and those that
did not. Trying to decide at what point a theme left the realm of the
universal and became specific depended on the personal opinion of
the critic. Consequently, while some writers praised a work like Bettis’s
As I Lay Dying, with its biographical sketches of two southern women,
others found it dangerously close to illustration.
The debates and the confusion that tore at modern dance in the
1940S were indicative of a genre in transition, but they also suggested
the ambivalent position artists held in relation to the dominant order.
Most wanted to save modern dance’s floundering vanguardism, but they
couldn’t agree on how to do it. And hovering over the arguments and
proposed solutions was the specter of the market with its seductive pos¬
sibilities for larger audiences and greater success. In addition, other major
pressures from outside the arena of dance came to bear on the field in
the immediate postwar years, further tangling dancers’ and critics’ mo¬
tives for support of one position or another. Foremost among these
were the rise of the Cold War and the repressive atmosphere that ac¬
companied it. Accusations of literalism were thus used, not just as a way
to reinforce modern dance’s position in the field, but as a means of con¬
demning social and political protest. For example, Joan Brodie found
Patsi Birsh’s In the Mines of Avondale, which dealt with a mining disas¬
ter, “a little too literal” (1948: 44), while Doris Hering warned in Dance
Magazine that Jose Limon’s El Grito was perilously close to a polemic in
its defense of a Mexican Indian uprising (1953a: 54).
In other instances, dancers were less overt in their politics, veiling
their critique in various ways that needed decoding. For example, as the
anticommunist hysteria of the late 1940s gained momentum, dancers
choreographed a number of works that dealt with the New England
witch hunts of the seventeenth century. These included Mary Anthony’s
The Devil in Massachusetts (1952), Miriam Pandor’s Salem Witch Hunt
(1952), James Dalgleish’s Salem Witchhunt (1952), and Marion Scott’s
The Afflicted Children (1953). (Arthur Miller’s The Crucible appeared on
Broadway in 1953.) Jose Limon’s The Traitor (1954), which dealt with
Judas’s betrayal of Christ, also was choreographed in these years. Yet
The Trouble with Modern Dance 33

critics did not decode such works. They either ignored any connection
to contemporary life in the dances, or if that proved impossible, they
labeled them literal. The critical response to protest narrative, whether
through silence or an accusation of literalness, was used as a means of
censoring material that might bring unwanted attention from the field
of power; as such it became complicit with that power.

Although the anticommunist witch hunts of the postwar years centered


on government employees, they also invaded the arts and entertainment
fields, the academy, unions, business, and many other areas. A number
of dancers had been involved with left-wing causes in the 1930s and suf¬
fered for it after the war. The most notorious case was that of Jerome
Robbins, who was called before the House Un-American Activities
Committee and named names, among them Edna Ocko, critic for sev¬
eral left-wing publications, including New Masses and the Daily Worker
(Bentley 1971; Jowitt 2004). Others who came under scrutiny included
Pearl Primus, who was called before the committee and whose passport
was revoked for a time; Anna Sokolow, who was interviewed without
being called; and Bella Lewitzky, who was kept under FBI surveillance
for a number of years (Barber 1992: 10; Warren 1998: 96; Craig 2001).
Such direct encounters with official repression were rare among high-
art dancers, but the threat of them sent a chill through the field.
In addition, dancers had another reason to fear outside social forces:
that was homophobia, which was rampant in the postwar period. It is
difficult to know exactly what percentage of men in the dance field
were gay, as such information did not enter written discourse. As John
D’Emilio has pointed out in his study on the emergence of the gay lib¬
eration movement in the 1960s (D’Emilio 1998), homosexuality in
the 1940S and ’50s was generally thought to be a personal, individual
issue. There was as yet little concept of a unified homosexual minority;
this idea developed only gradually over the course of the next several
decades. In the postwar period, heterosexual norms were enforced on the
stage, and heterosexuality was assumed offstage in written accounts. Yet if
homosexuality did not enter dance’s written record, it is generally agreed
that the percentages of gay men working in the dance field—including
dancers, choreographers, designers, composers, administrative person¬
nel, and critics—have been high since early in the twentieth century.
During the postwar years, for example, a large number of male chore-
34 A Game for Dancers

ographers were gay or bisexual, including Robbins, Tudor, Weidman,


Limon, Cunningham, Nikolais, Taylor, and Alley (Jowitt 2004; Chazin-
Bennahum 1994; Siegel 1987; Burt 1995; Louis 1973; Taylor 1987;
Dunning 1996; DeFrantz 2004). Important dance critics were also gay
or bisexual, as were a number of company directors, administrators, and
supporting personnel. Both women and gay men found a haven in the
dance field where they could garner support from like-minded individ¬
uals and where, too, they could gain power. However, being dominated
by women and gay men gave dance less stature than the other arts. Read
through feminist theory, dance was a feminized site in a society of
heterosexual male domination. Within the field there was much anxiety
about dance’s effeminacy, as witnessed in articles that appeared in the
dance press. These generally ran along two lines: a case would be made
for the masculinity of dance by comparing its physical demands to those
of other physically taxing activities, or a case would be made for why par¬
ents should allow their sons to study dance. In an interview in the
1950S, George Balanchine combined both points (Balanchine 1954:
504). To the question of whether or not hoys should be allowed to take
ballet classes Balanchine replied that he understood that parents were
hesitant to send their sons to ballet school because they were afraid they
would become “sissified” or would not develop the strong muscles that
other activities would give them. However, he argued, ballet builds
strong, flexible bodies that have a great deal of endurance. Then, ap¬
pealing to manly patriotism, he added, “this is the reason many of our
best dancers were good soldiers during the war” (ibid.). Balanchine was
a convincing spokesman for the virility of male dancing, since his nu¬
merous marriages to beautiful ballerinas apparently demonstrated his
heterosexuality. However, public perceptions, as Balanchine noted,
were that dance was for “sissies,” and this was attested to by the low
numbers of boys compared with girls taking dance classes and by the
advanced age at which boys entered the field compared with girls
(Kirstein [1953] 1983; 97-106; [1959] 1983: 398-400). The low eco¬
nomic rewards and short careers of dancers also must have made dance
unattractive to all but the most determined men.
Homophobia was tied to anticommunist rhetoric to an extraordinary
degree in the postwar period. As historian Stephen Whitfield has noted,
“If only through innuendo, nothing in the political arena in the 1950s
was more convenient than establishing some sort of connection between
political and sexual ‘perversion’” (Whitfield 1991: 43). The casualness
The Trouble with Modern Dance — 35

with which communism and homosexuality were linked is remarkable.


For example, in 1946 President Truman fumed privately that “All the
‘Artists’ with a Capital A, the parlor pinks and the soprano voiced men
are banded together.... I am afraid they are a sabotage front for Uncle
Joe Stalin” (quoted in Griffith 1981: 298), while Arthur Schlesinger Jr.
noted in The Vital Center that communism “perverts politics into some¬
thing secret, sweaty and furtive like nothing so much, in the phrase of one
wise observer of modern Russia, as homosexuality in a boys’ school”
([1949] 1998: 151). Communists were routinely equated with effemi¬
nacy and homosexuality on a metaphoric level, while acmal homosexu¬
als were viewed as a threat to national security and the moral fiber of the
country. According to a Senate report commissioned in 1950 to inves¬
tigate homosexuality in the federal civil service, the stigma of their life¬
style left homosexuals open to blackmail and therefore made them dan¬
gerous if they held jobs in the government or military. On the subject
of morality the Senate report continued: “These perverts will frequendy
attempt to entice normal individuals to engage in perverted practices.
This is particularly true in the case of young and impressionable people
who might come under the influence of a pervert” (quoted in D’Emilio
and Freedman 1997: 293).
Homophobic hysteria may well have been aided by the Kinsey re¬
port on male sexuality, which appeared in 1948 and revealed that homo¬
sexuality was far more widespread than previously realized. Kinsey re¬
ported that at least 37 percent of the male population had engaged in
some overt homosexual activity between the onset of adolescence and
old age. Among those unmarried to age thirty-five, 50 percent had
some homosexual experience, as had 10 percent of married males be¬
tween the ages of sixteen and twenty-five. The report also countered
ideas of stable sexuality, showing that sexuality was often flexible and
changing throughout a person’s life. Kinsey said that his research found
that homosexuality existed in every age group, social level, and occupa¬
tion, and in both cities and rural areas. He added that his statistics were,
if anything, understated because of the stigma of homosexuality (1948:
168-172, 610-656).
Another factor in the growth of homophobia during the postwar
years was the religious revival that swept the country. Between 1940 and
1959, church-going in the United States rose from 49 to 69 percent,
while men like Bishop Fulton Sheen and the Reverend Billy Graham
became household words. In the 1950s “one nation under God” was
36 A Game for Dancers

added to the Pledge of Allegiance, recited each morning by school-


children throughout the country, and “In God We Trust” was added to
the U.S. currency. Godless communism and perverted homosexuality
were anathema to an organized religion that grew ever stronger during
the1940S and ’50s (Oakley 1986: 319-327; Silk 1988; 15-107; Whitfield
1991: 82-84; O’Neill 1986: 212-215). These various sources of homo¬
phobia had real consequences as gays and lesbians were increasingly
barred from the civil service and the military, academia, and other fields
such as businesses with government contracts. In addition, local police
routinely harassed homosexuals by raiding gay bars and meeting places
(D’Emilio and Freedman 1997: 293-294). Newspapers often published
the names of homosexuals detained by the police, and should a gay man
be exposed, his working life, even as a dancer, would likely be over.
Considering this political and social climate, the last thing the dance
field needed was close scrutiny from government agencies and insti¬
tutions that linked social protest with communism, communism with
homosexuality, and homosexuality with perversion and treason.
Modern dance, then, faced troubles on several major fronts in the
postwar years. These included most importantly its own institutional¬
ization, which put its vanguard position in jeopardy, and increasing
pressures from inside and outside the field alike, which sought to elim¬
inate or suppress any element of critique. The rules of modernism de¬
manded solutions to both sets of problems, yet modernism itself was
problematic as elements of it were co-opted for Cold War use. In the
meantime, another challenger appeared from outside the genre in the
unlikely guise of ballet. Although modern dancers had long worried
about the power of ballet, the assumption generally had been that bal¬
let would overwhelm and assimilate modern dance. George Beiswanger’s
view was typical. Writing in 1942, he stated: “It is quite possible that the
final and only lasting service of the modern dance will be the rejuvena¬
tion of ballet. Let us face the facts. Ballet has the money, the organiza¬
tion, the repertory and the audience. And, like all established institutions,
it has a flexible conscience” (1942: 116). Having devoured its own past,
Beiswanger said, ballet looked for new sustenance: “Whenever it begins
to see something fresh and exciting in the new impulses of its own day,
it takes them over, babbling all the time about the eternal verities of the
five positions and the mrn-out of the thigh” (ibid.). Although Beiswanger
had cause for concern, as ballet choreographers such as Antony Tudor
and Agnes de Mille appropriated aspects of modern dance, this did not
The Trouble with Modern Dance 37

prove to be the major threat. Rather, it arrived in the guise of Lincoln


Kirstein, and through him George Balanchine, who did not so much
absorb modern dance as attempt to compete with it. Kirstein’s auda¬
cious plan was to wrest from modern dance its vanguard place within
high-art dance. The extent of this challenge will be taken up in the next
chapter.
2

BALLET'S CHALLENGE

I n the fall of 1946, Lincoln Kirstein issued a manifesto. The occasion


was the inauguration of Ballet Society, the subscription series he had
undertaken and funded with his own money. It stated, in part:

The Ballet Society will present a completely new repertory consisting of


ballets, ballet-opera and other lyric forms. Each will have the planned
collaboration of independent easel-painters, progressive choreographers
and musicians, employing the full use of advance-guard ideas, methods
and materials. . . . Since ballet in the U.S. is relatively new, our interest
has been primarily in the revival of productions already famous, or the
creation of works based on national themes. Now, with the close of a
second world war, broader directions are possible and desirable.^

Although Kirstein’s remarks would seem to be aimed at other ballet


companies, his intentions went beyond challenges to the Ballet Russe
de Monte Carlo and the six-year-old Ballet Theatre, the major Ameri¬
can companies at that time.^ Elis intention in using advance-guard ideas,
methods, and materials was also to provoke modern dance. ICirstein
planned to carry out his conquest of the vanguard through the chore¬
ography of George Balanchine, whose work ICirstein viewed as a refor¬
mation of the classical danse d’ecole. Kirstein’s notion of “advance-guard
ideas” was primarily aesthetic. He was not interested in opposing insti¬
tutionalized art but in building large, knowledgeable audiences. Never¬
theless, the fact remains that in the 1940s ballet briefly attempted to
compete with modern dance for position in the field, and this competi¬
tion came at a time when modern dance was particularly vulnerable. The
real danger in Kirstein’s challenge was not that ballet would become a
true modernist vanguard, but that it would expose the fault lines in mod-
Lincoln Kirstein and George Balanchine in the 1950s. Photo: TanaquilLeClnxq. Courtesy of the
Tanaquil LeClercq Collection, New York City Ballet Archives.

ern dance and so further weaken, if not destroy, the genre’s already vul¬
nerable status and position. For through Balanchine, ballet offered an
aesthetic based on an antiliteral dance that was more stringent than much
of what modern dance was offering by the mid-1940s. The Kirstein-
Balanchine enterprise is examined here not only because it set itself
up as a rival to modern dance, but because Balanchine’s neoclassicism
helped encourage a young generation of modern dancers to devise an
obj’ectivist dance of their own that, in the 1950s, was declared a new
modernist vanguard.
40 A Game for Dancers

Kirstein’s antipathy to modern dance was long-standing; he had begun


to write polemics against it in the 1930s and remained opposed to it for
the rest of his life (Reynolds 1983: xvi-xvii).^ His argument, laid out in
several early articles, was that ballet had a better chance of fulfilling the
requisites of a modernist vanguard than modern dance. Modern dance’s
failure, he said, was that it was a full-scale revolt from the accepted form
of high-art dance, and as such was a dead-end because it had no way
to remake itself Kirstein compared ballet’s position to that of cubism,
the innovations of which had not killed the tradition of painting but re¬
newed it ([1937] 1983: 253). Cubism was a method of self-criticism that
reconsidered the tradition of painting. Ballet, too, built on and renewed
tradition rather than dismantling it."^ Modern dance, on the other hand,
constimted a complete break with the past; as such it had no basis for
further development. Compounding the problem in Kirstein’s view was
the fact that modern dance was not founded on an impersonal vocabulary
that could be expanded or refreshed, but on a dancer-choreographer’s
personality. The subject of Graham’s dances was not dance, but Gra¬
ham herself. For this reason, each modern dance choreographer was
forced to reinvent the genre. Kirstein saw little future for modern dance
under such conditions ([1934b] 1983: 38-41).
One might well ask why Kirstein wanted to challenge the existing
vanguard, especially when the two most prominent American ballet com¬
panies saw no need to do so. The obvious answer was to differentiate
Ballet Society from other ballet companies, and it is likely this was one
of his reasons. However, additional answers come to mind.
The first has to do with Kirstein’s own background and dispositions,
beginning with his Harvard education and contacts (Weber 1992; Kir¬
stein 1994; Jenkins 1998). During his smdent days in the 1920s, he had
been attached to several important modernist projects. These included
Hound and Horn, a literary magazine he helped found, which published
some of the most advanced modernist writing of its day. Authors in¬
cluded James Joyce, Gertrude Stein, T. S. Eliot, William Carlos Wil¬
liams, e.e. Cummings, and Ezra Pound. He also started the Harvard
Society for Contemporary Art with his friend Edward Warburg of the
banking family. The gallery showed the work of a host of modern artists
ranging from Americans such as Georgia O’Keeffe, Arthur Dove, and
Edward Hopper, to School of Paris painters such as Matisse, Braque,
and Brancusi, and a number of surrealists. The society was an acknowl¬
edged precursor of the Museum of Modern Art. Among Kirstein’s uni-
Ballet’s Challenge 41

versity friends were Alfred H. Barr, who would become the first direc¬
tor of the Museum of Modern Art, and Philip Johnson, the modernist
architect who in the 1960s designed an addition to the Museum of
Modern Art and who also designed the State Theatre at Lincoln Cen¬
ter, the eventual home of New York City Ballet. In the 1930s, Kirstein
also had been on the junior advisory committee for the Museum of
Modern Art and established a dance archive there. He had close ties
to many of the surrealists, neoromantics, and other vanguard European
emigre artists who lived in New York during the war and who designed
sets and costumes for his projects. This world of modernist movers and
shakers in which Kirstein circulated may well have made him feel that
his ballet company had to fit a modernist model if it were to garner the
kind of prestige he sought.
By the end of the war, Kirstein, with his connections and well-honed
feel for the game, also may have realized that taking an advanced-guard
position could reap more than the limited, if rarefied, status of the
prewar years. As powerful elites within the United States came to view
modernism as a weapon, first against fascism and then communism,
there were larger prizes to be had. Granted, modernism in 1946 was
controversial, but Kirstein had influential friends, such as Nelson Rock¬
efeller, who were backing it.^ Certainly he understood that the Ameri¬
cana subject matter he had favored in his concert group. Ballet Caravan,
in the late 1930s was no longer viable. That small concert group, which
Kirstein had assembled two years before his more ambitious American
Ballet expired in 1938, had been devoted to works that dealt primarily
with America’s past or with a vernacular scene of the present. Pocahon¬
tas (1936), Yankee Clipper (1937), Billy the Kid (1938), and Filling Station
(1938) were just some of the works that Ballet Caravan presented in its
five-year existence.*^ These ballets paralleled the paintings of regional
artists and the works of many modern dancers of the same period. By
1946, Kirstein realized it was necessary to take a different path. The
“broader directions” he spoke of in his manifesto would be toward the
internationalism that was emerging at the time, and the vehicle would
be Ballet Society, led by Balanchine.
Kirstein had attempted to cast Balanchine in several guises since he
had brought him to the United States in 1933, each designed to bene¬
fit from current trends. An article for the leftist periodical New Theatre,
written the year after Balanchine’s arrival, depicted him as a revolu¬
tionary with communist sympathies (Kirstein 1934a: 12-14). Kirstein
42 A Game for Dancers

mentioned that Balanchine’s father, a musician, had recently been hon¬


ored by a jubilee in the Soviet Union while neglecting to mention that
Balanchine had left his native country in 1924 with no intention of
going back. In the article, Kirstein did not emphasize abstract classi¬
cism, which was to become the basis of Balanchine’s postwar style, but
rather his interest in renewing the relationship between dance and
drama. In support of this, he pointed to an early Balanchine piece in the
USSR set to a poem by Alexander Blok. He went on to say that Balan¬
chine had been interested in dance-drama at the Ballets Russes, perhaps
^■nkingo^Prodigal Son (1929), and in 1933 had set The Seven Deadly Sins
by Bertolt Brecht, who Kirstein described as a “superb young commu¬
nist artist.” Balanchine, Kirstein wrote, “knows ballet as ‘ballet’ is
dead.” Instead he had reinstated another term, “choreodrama,” and ac¬
cording to Kirstein, planned a revolutionary kind of theater around it.
The form would be that of the old ballet—^Kirstein argued that revolu¬
tions preserve the best elements of the past, which included balletic
form—^with new content delivered in new ways. “The workers need
a demonstrated subject matter,” he wrote, “a dramatized, legible spec¬
tacle, far more than they need a new form of expression” (1934a: 14).
Choreodrama, and the accompanying image of Balanchine as a
communist-inspired revolutionary, soon disappeared from Kirstein’s
writing, but he still did not emphasize classicism or abstraction. Rather,
he now stressed the neoromantic aspects of Balanchine’s choreography.
Neoromanticism, which like surrealism dealt with the irrational—^with
dreams, visions, hallucinations, and the unconscious—^was associated
with artists like Pavel Tchelitchev and Eugene Berman, both of whom de¬
signed interwar Balanchine ballets. In Blast at Ballet, Kirstein described
Balanchine in neoromantic terms:

His choreography is not ever literally narrative; it seldom tells a consec¬


utive story. It weaves, however, a dominating but eventually a climactic
and profound atmosphere which is frequently romantic, but just as often
so odd in its romance as to seem a genre of romanticism the like of which
has seldom been seen before. It is the poetic romanticism not of a
Rousseau or Chateauhriand, but of the cruel tales of E. T. A. Hoffmann,
Grimm or Andersen; the somber verses of Lenau, Heine and Holderlin,
the cymcal self-contempt of Byron and Poushkin. It is a romantic attitude
divested of sentimentality. It is an atmosphere at once nervous, individual,
spoiled, tender and tragic. ([1937] 1983: 178)
Ballet’s Challenge 43

Kirstein, himself, was partial to neoromanticism, as well as surrealism.


For many years Tchelitchev, in particular, was a friend and consultant to
both Kirstein and Balanchine (Taper 1984: 163; Buckle 1988; Kirstein
i973> 1994)- Ballet Society choreographers also used some of the sub¬
ject matter that concerned the surrealists and neoromantics, namely
myth and legend. John Taras created a ballet entitled The Minotaur
(1947), and Balanchine produced The Triumph of Bacchus and Ariadne
(1948) and Ofpheus (1948).
However, by 1947 Kirstein was no longer promoting Balanchine as
a neoromantic, even though neoromantic and surrealist-inspired works
made up part of Ballet Society’s repertory. Kirstein now pronounced
Balanchine a neoclassicist who had developed an American style of ballet
through a nonnarrative, or abstract, dance. In two articles written for
Theatre Arts, Kirstein traced a history of Balanchine’s work that saw him
first as inheriting the mantle of Marius Petipa, the father of classical
ballet (Kirstein 1947a: 37-41; 1947b: 37-41). Balanchine received this
inheritance through his training and association with the Imperial Bal¬
let of St. Petersburg and his dedication to the danse dTcole. Next, Balan¬
chine had advanced his classical heritage into pure-dance ballets that he
had evolved in the United States through such works as Serenade (1935),
Concerto Barocco (1941), Ballet Imperial (1941), and The Four Temperaments
(1946). Kirstein’s linking of plotless ballet to America suggested that
neoclassical abstraction was an American form.
Yet, however eager Kirstein was to have Balanchine seen as the cre¬
ator of a new American form, the opinion of most critics from the time
of Balanchine’s arrival in the United States had been that his work was
part of a decaying Franco-Russian ballet. Few denied Balanchine’s chore¬
ographic facility, but they considered him glib and superficial. Walter
Terry (1941: 10) of the New York Herald Tribune compared Balanchine
unfavorably with Martha Graham, while John Martin spoke of Balan¬
chine’s “anemic aesthetics” ([1939] 1965: 207). In a 1937 review of the
American Ballet, Martin wrote:

When dealing with Balanchine’s compositions, it is necessary always to


accept the fact that for him the dance, as he has expressly stated, is purely
spectacular and should strive only for the attainment of sensuous plea¬
sure. That he believes such pleasure is best attained by tenuous rather
than robust means is apparentiy characteristic of his personal approach.
It should not be surprising, then, to discover that tbe three ballets he
44 A Game for Dancers

has presented, two of them newly created, should be without substance


and trivial. (1937: 7)

Of the same American Ballet program Gervase Butler wrote in Dance


Observer: “His [Balanchine’s] chief error lies in the choice of movement
and concept so foreign to the American idiom. In a world wherein we
find ourselves the youngest and most vital nation, the European cachet
has long lost its potency on the arts it dominated before the war. In other
words, we have not yet attained American Ballet” (1937= ^i)-
As late as the mid-1940s, Dance Observer critics were still accusing
Balanchine of superficiality and foreign glamour. Robert Sabin wrote:
“Mr. Balanchine suffers from his own cleverness. He is such an artist
to his finger tips, so witty and ingenious that he can say practically noth¬
ing with an air of having said a great deal. Dances Concertantes is a case
in point, for it contains few passages of memorable choreography, yet
one cannot resist its glittering facility” (1944: loi). Ayear later, Sabin
spoke of Balanchine’s Concerto Barocco as “neo-classic chi-chi” (1945b:
97). It was clear that if Balanchine were going to be seen as the inven¬
tor of a new American style, some minds would have to be changed.
And change, they did. Between 1945 and 1955 Balanchine was trans¬
formed from a foreign interloper out of touch with American dance
into the creative genius of a native American ballet. This is not to say
that Balanchine did not have his supporters early on. Anatole Chujoy,
editor of Dance News, wrote intelligently of Balanchine’s work and
published a book on the New York City Ballet in 1953. Even more im¬
portant was Edwin Denby, who almost single-handedly defined what
would become Balanchine’s American style. But it may have been John
Martin’s capitulation more than anything else that shifted the tide of
opinion. In the late 1940s, Martin, the great defender of modern dance,
began to see in Balanchine many of the qualities to which modern
dance had hitherto held sole claim. These included experimentation,
an American style, and a dance that put movement at its center. As such
this ballet challenged modern dance’s position on several fronts.
What was Balanchine doing in the mid-1940s that encouraged crit¬
ics to change their opinions, and why did his ballet come to be so
appealing to postwar observers? It is time to look at Balanchine’s work
of those years, in this case his most radical ballet and one that set the
style for other “experimental” endeavors. Balanchine chose to choreo¬
graph The Four Temperaments for the first Ballet Society program on 20
Ballet’s Challenge 45

November 1946. The ballet represented only one of several styles in


which Balanchine worked simultaneously, and it must be emphasized
that it remained only one such style. Throughout his career Balanchine
made narrative works, and for many years he continued to produce neo¬
romantic ballets. However, The Four Temperaments v^oxAdi come to epit¬
omize what for critics was Balanchine’s American style.

Balanchine had conunissioned the score for The Four Temperaments


from Paul Hindemith in 1940 with money from his work in films and
on Broadway' Hindemith supplied a chamber piece entitled Theme with
Four Variations (according to The Four Temperaments). The temperaments
were associated with the four humours of medieval physiology and the
corresponding temperaments they were supposed to produce in humans:
black bile (melancholic), blood (sanguinic), phlegm (phlegmatic), and
yellow bile (choleric). The score for The Four Temperaments consists of
a three-part theme and four variations, the latter entitled Melancholic,
Sanguinic, Phlegmatic, and Choleric. In Balanchine's New Complete Sto¬
ries of the Great Ballets the work is described as follows:

Subtitled “A Dance Ballet without Plot,” The Four Temperaments is an


expression in dance and music of the ancient notion that the human or¬
ganism is made up of four different humours, or temperaments. Each one
of us possesses these four humours, but in different degrees, and it is from
the dominance of one of them that the four physical and psychological
types—melancholic, sanguinic, phlegmatic, and choleric—were derived.
Greek medicine associated the four humours and temperaments with the
four elements—earth, water, fire, and air—which to them composed the
human body as well as the world. (Balanchine 1968: 171)

The author adds that “neither the music nor the ballet itself make spe¬
cific or literal interpretation of the idea” (ibid.).
The Four Temperaments is divided according to the score with three
duets stating the three-part theme.® The duets are followed by the four
variations. The first. Melancholic, features a male soloist and corps de
hallet; the second, Sanguinic, is a duet with corps; the third. Phleg¬
matic, is for male soloist and corps; and the fourth. Choleric, is for a fe¬
male soloist and corps. Balanchine also generally followed the score
in aspects such as tempo and pulse drive, and he took his cue from the
46 A Game for Dancers

The Phlegmatic variation with Todd Bolender and corps de ballet in the original Ballet
Society production of George Balanchine’s The Four Temperaments (1946), © The George
Balanchine Trust. Photo: Peter Campbell. Courtesy of Jerome Robbins Dance Division, The New York Public
Library for the Performing Arts, Astor, Lenox and Tilden Foundations.

composer in the plotlessness of the ballet, using the notion of tempera¬


ments simply as a point from which to begin. Balanchine’s close follow¬
ing of the score is not surprising since he was an expert pianist who had
studied to be a composer as well as a dancer. For him dance was always
inextricably tied to music.^
Ballet movement, derived from the vocabulary and comportment
of the classical dance or danse d^kole, makes its effects through such el¬
ements as pattern and dynamics, but also through what are commonly
called movement images. Sally Banes has defined such images as “sug¬
gested characterization which appears, disappears, and reappears through
the articulation of posture and gesture” (Banes 1995: 350). The most
prosaic version of such movement images is pantomime, an imitation of
movement from everyday life, such as recoiling in fear. A step removed
from pantomime is what theater director William Ball calls connotative
Ballet’s Challenge 47

movement, those conventionalized gestures and movements that are


used on the Western theater stage and that are understood to have gen¬
eral meaning (Ball 1984; I lo-i 12). A person kneeling before another to
suggest supplication is one example, as is a couple holding hands to sug¬
gest romance. At a less mimetic level, dance imagery may draw on other
encoded systems to suggest associations, such as baroque sculpture or
military drill. Or dance imagery may derive from movements and ges¬
tures that have no specific meaning in themselves but that suggest
meaning through formal processes, including such devices as repetition,
symmetry, and recontextualization.^®
Looking at The Four Temperaments, it is clear that despite the moods
that are indicated in the various section titles, little of the movement
makes clear reference to them. Occasionally a dancer performs an ac¬
tion that derives from daily life and is immediately recognizable. So,
in the Melancholic variation, the male soloist circles down into a tight
heap, his head lowered in what is a virtual cliche for dejection. But such
images are rare. Far more common are movements that are suggestive,
but less specific. The Melancholic soloist pushes through the “obstruct¬
ing” arms of the female supporting dancers. In the Sanguinic duet the
man lifts the woman in a series of low split lifts that encircle the stage
and in which she seems to be pushing through the air in slow flight. In
other instances movement derives from encoded systems within paint¬
ing and sculpture that refer to medieval imagery. The Phlegmatic fig¬
ure at one point resembles an unhappy jester, his knees turned in, his
body drooping, while at another, he executes a courtly bow. Lincoln
Kirstein described Todd Bolender, the original dancer of the Phleg¬
matic variation as “a fluidly sluggish acrobatic mendicant” (1973: 82).
The women in the ballet recall medieval painting and sculpture, as well
as twentieth-cenmry showgirls and fashion models, when they stand
with their hips thrust forward and their body weight back.
There are many dance images in the ballet that could suggest mul¬
tiple meanings, such as the women’s pose just mentioned. Another ex¬
ample occurs in the Choleric variation, which centers on a female
soloist. The dancer enters from upstage left and moves horizontally
across the space in a series of fast, tight turns. At center stage she whips
to a kneeling position with her head and arms stretched downward as
if she were diving toward the floor. Although this speed might be in¬
terpreted as anger, one of the dominant traits of the choleric tempera¬
ment, or associated with fire (the Greek counterpart of the medieval
48 A Game for Dancers

The Melancholic variation from George Balanchine’s The Four Temperaments, © The
George Balanchine Trust. New York City Ballet production, 1969. Photo: © Martha Swope.

humour), it could just as easily be seen simply as speed, since there is little
else in the dance that attaches itself to clear cues of anger or flames in
everyday movement. The opening music, however, is in a fast tempo
that suggests speed, and Balanchine, faithful to the music, has followed
its suggestion. Music, therefore, also plays a role in accounting for
dance images.
Added to these various movement images are others that seem sim¬
ply mysterious. In the second duet of the theme (repeated in the finale)
the man and woman exit across the stage in sharply pointed walking
steps while their heads and torsos face front like Egyptian bas-reliefs. As
they step, they flip their arms, bent at sharp right angles, up and down
in opposition to each other as if engaging in an untranslatable kind of
semaphore. At other moments the women make cat’s paw-like gestures
with their hands curled softly in front of them, or execute stiff arm move¬
ments out to the side as if demonstrating the kick of the Australian crawl.
Ballet’s Challenge 49

The Four Temperaments, then, offers an array of movement images


to which a viewer’s eye can attach and make fleeting connections yet can
never organize into a narrative or even into a unified mood. As critic
Walter Terry noted in an early review, the various images emerge and
then are immediately dissolved into the larger framework of abstract
classicism (1951b: 17). However, while classicism ultimately unifies the
disparate dance images, the danse dTcole itself is constantly disrupted by
anticlassical elements. These include the use of a flexed as well as the
pointed foot of the danse d'kole, turning the leg in as well as out, shift¬
ing the weight down as well as up, and allowing the torso to collapse as
well as assume the held-up position that is a basic part of ballet com¬
portment. Most important among these disruptive elements is a recur¬
ring decentering of the body, which forms a leitmotif through the ballet.
So, for example, in the many grands battements that are included, the
body is not centered and straight, but is pushed front, back, or sideways.
In another repeated pose, the body is cantilevered back over the sup¬
porting leg, which is bent in a demi-plie, while the working leg points
forward. The Melancholic variation is filled with backbends, giving the
impression that the soloist is on the edge of gravity, that if he were to
move another inch he would collapse—and these are only a few examples
of decenteredness that could be mentioned.
Dancer Merrill Ashley has spoken of having difficulties with a step
in the Sanguinic variation in which, unsupported, she was supposed to
look as if she were falling backward, although the fall was in fact con¬
trolled. She explained: “In The Four Temperaments the body is frequendy
at an oblique angle. It’s a very unusual type of movement which takes
some getting used to. I think all of the dancers in ‘Four T’s’ have expe¬
rienced some difficulty in learning to move in such a unique way” (Ash¬
ley et al. 1987: 27; see also Ashley 1982: 384-386). Patricia Wilde agreed:
“I found Sanguinic difficult. Very difficult. It’s that strange, off-balance
thrusting and then moving back again to another off-balance position
and then having to right yourself and go into the many multiple turns
and things. That does create problems” (Wilde et al. 1987: 28). The
steps may have been difficult because they occurred neither in the danse
dTcole nor in usual human comportment. Being off-balance was there¬
fore unusual and uncomfortable. However, although the use of such
anticlassical elements as turn-in, flexed feet, a collapsed torso, and de¬
centeredness disrupt and fragment the unity of the danse dTcole, it should
be emphasized that they are not allowed to dominate the dance, which
A Game for Dancers
50

Mary Ellen Moylan and Fred Daniel! in the Sanguinic variation from the original Ballet
Society production of George Balanchine’s The Fotir Temperaments (1946), © The George
Balanchine Trust. Courtesy of Ballet Society Collection, New York City Ballet Archives.

continually returns to the ballet vocabulary and comportment. The sig¬


nificance of the dominance of classicism will be discussed shortly. First,
however, I want to examine the anticlassical aspects of the ballet in
more detail.
At least one commentator in the postwar years saw influences of
modern dance in The Four Temperaments. Walter Terry wrote in 1951
that “the total effect is quite modern-dance in quality, for fluid torsos,
archaic arm patterns and breath rhythms (as opposed to merely musical
rhythms) are apparent. . .” (1951b: 17). Terry may well have been cor¬
rect, since by the mid-1940s Balanchine certainly had been exposed to
modern dance. I believe, though, that movement of another kind is
more central to the work. Brenda Dixon Gottschild (1996) and Sally
Banes (1994: 53-69) have made a case for the use of what Gottschild calls
an “Africanist presence” in Balanchine’s work. According to Gottschild,
this presence is represented most strongly in references to African and
African American-derived dance, and in an African-derived aesthetic.
Ballet’s Challenge 51

which favors such elements as speed and sharp attack, polyrhythm and
polycentrism (more than one body part being emphasized at once),
omission of transitions, syncopation, and a cool presentation (Gottschild
1996: 11-19). It is important to note that “Africanist” must be thought
of here in its broadest terms, since by Balanchine’s time such elements
often were filtered through the work of white artists and performers to
enter a general lexicon of entertainment media as well as white social
dance. High-art forms, particularly music, painting and sculpture, had
also absorbed Africanist rhythms and imagery. If considered within this
broad definition, Gottschild and Banes make compelling arguments
for Africanist sources in Balanchine’s work. In the 1950s, Edwin Denby,
too, noted these sources ([1953] 1986: 438).
Both Gottschild and Banes state that Balanchine had links to African¬
ist art and culture even before leaving Russia in the 1920s and that he
used Africanist elements in his work at least as early as Apollon Musagete
(1928). Afore important, Balanchine had worked extensively in the com¬
mercial theater and films during the 1930s and ’40s, often with black
artists such as Josephine Baker, Katherine Dunham, Herbie Harper,
and Buddy Bradley. Dunham and other black dancers noted how much
Balanchine appreciated their dance, whether it was tap, jazz, or the
Haitian-inspired work of Dunham herself. Gottschild includes a photo
of Balanchine with Dunham and other dancers in which Balanchine is
executing a black social dance move, his left leg lifted high, his right
finger pointing skyward. The picture was taken when Balanchine was
working with Dunham and her troupe on the Broadway musical. Cabin
in the Sky (1940).
Both Gottschild and Banes have analyzed Africanist sources specifi¬
cally in The Four Temperaments. They note the focus on Alficanist ele¬
ments such as angular arms, foot flexes, jutting hips, turn-in, and bent
knees. They also mention a leaned-back “cakewalk” pose in which the
torso is thrown back while the legs execute high-stepping prances, and
the little pawlike gestures that were seen in postwar black social dance
and in chorus lines. In addition, there is a crouch that the female
dancers assume from time to time, which Banes and Gottschild label
the “get-down” posture common in Africanist dance. Gottschild points
out a lift that is seen frequently in the ballet in which the man lifts his
partner from behind as she executes a wide side-split. Gottschild says
that this was a popular move in the Lindy Hop and illustrates her argu¬
ment with a photo. She also notes several other steps that are balletic
52 A Game for Dancers

versions of Lindy moves, including a moment in the second theme duet


where the man holds his partner from behind and pushes her hips for¬
ward and back as he turns her. Most notable, however, according to
Gottschild, is Balanchine’s use of decenteredness, which in the duets
“deconstruct and defy the ballet’s canon of verticality and male sup¬
port of female centeredness, essentials in the classic pas de deux. . . .
The males in these duets push, thrust, or manipulate their partners off
center. They seem to play with letting the female fall” (1996: 73).
Gottschild sees this as “a ballroom dance risk-taking that Balanchine
has crafted to meet the needs of the ballet aesthetic and the concert
stage” (ibid.).
A number of dancers have commented on the amount of male ma¬
nipulation of female bodies that goes on in The Four Temperaments, as
well as the work’s emphasis on speed and attack (Reynolds 1977: 73-74;
LeClercq 1982: 153)^2 The manipulation of the female body often has
to do with holding women so that they can maneuver far off-balance.
But in some instances, as in the Choleric variation, the men actually
throw or toss the woman into the air, something that they also did in
postwar African-American social dances.
Ballet choreographers have traditionally used elements from outside
the danse d’e'cole as a means of decoration or foreign color. However,
Balanchine did not insert African-American steps into The Four Tem¬
peraments as references or identifying marks of the foreign. Rather he
did just the opposite, decontextualizing and abstracting them as he did
other dance images. As Banes and Gottschild point out, Balanchine also
integrated Africanist aesthetic elements such as speed, attack, and angu¬
larity into the ballet medium.
Along with Africanist elements, Balanchine also used methods simi¬
lar to those of the surrealists to disrupt any kind of sustained narrative
or thematic logic in The Four Temperaments. Balanchine had connections
with surrealism going back to his time with Diaghilev’s Ballets Russes,
and he had worked witli surrealist designers later in Europe and the
United States. Important, too, was Kirstein’s enthusiasm for surrealism
and his understanding of its place in the art world. A number of the
surrealists had found their way to New York during the war, including
Kurt Seligmann, Andre Breton, Max Ernst, Man Ray, and Andre Mas¬
son. Surreahsm represented the current international avant-garde, and
during the 1940s, Americans proceeded to absorb its lessons. At the
same time, because the United States was now the “leader of the free
Ballet’s Challenge 53

world,” it needed an art that not only stood in opposition to the Soviet
Union’s, it also needed one that could compete favorably with that of
Europe. Surrealism could help place American art within that inter¬
national milieu. Significantly, when it came time to select a designer for
The Four Temperaments, Kurt Seligmann was enlisted.
Although Balanchine did not use the kind of subject matter in the
ballet usually associated with surrealism, particularly myth and dream,
he did employ surrealist devices. This is not to suggest that Balanchine
shared the surrealist vision of art’s relationship to society, a relationship
that called into question the very existence of the institution of art
and that Balanchine’s dedication to ballet classicism would hardly have
allowed. Rather, his assumption of a number of surrealism’s means im¬
bued The Four Temperaments with an ambiguity and enough disruption
of tradition to put it on the side of formal experiment while not chal¬
lenging the ruling order in any substantial way.^"^
Among the surrealist devices Balanchine used in The Four Tempera¬
ments were dislocation, contradiction, and incongruity. In surrealism
these were often represented in the form of collage, an arrangement of
disparate elements that in visual art usually includes cut-outs from mass-
produced publications that, in combination, communicate no specific
theme or narrative. Balanchine’s use of vernacular movement and pop¬
ular dance steps resembled collage in that they were disparate and un¬
attached to any sort of narrative or thematic framework. Instead, they
were decontextualized and arranged formally through visual and tem¬
poral elements such as shape and texture, tempo, and dynamics. They
were also dissolved into more abstract movements so they lost their
specificity or were themselves melded into balletic comportment and
style. So, for example, the Melancholic soloist crumples into what may
appear to be a pose connoting dejection, but the image is nearly free-
floating because it is not reinforced or referred back to narratively by
other movement surrounding it. This includes both the movement the
soloist does temporally before and after the pose and the movement
done simultaneously by the corps dancers surrounding him. However,
it must be emphasized that although Balanchine broke up narrative
logic, he did not interrupt formal logic. As mentioned above, he related
movement through such elements as pattern, texture, and dynamics.
He was, of course, also highly sensitive to the musical structure, which
served as a strong organizing force in the work.
Manipulation and transformation of the (usually female) body were
54 A Game for Dancers

common in surrealism, especially in photography. In Primacy of Matter


over Thought (1929), Man Ray depicted a woman melting into a pool of
liquid, and Brassai, in Woman-Amphora (1935), showed a female figure
cropped to resemble a ceramic vessel. Balanchine also used transforma¬
tion in The Four Temperaments. The best-known example occurs in the
third theme duet when the man turns the woman in a manner reminis¬
cent of the way in which jazz musicians used to spin the bass viol as
part of a virtuosic display. When Balanchine spun a female dancer like
a bass fiddle he, too, was making a comment on the transformative pos¬
sibilities of the human body.
In surrealism, manipulation of the body most often occurred in the
artist’s manipulation of the image rather than in the subject matter of
the work, that is, in one figure depicted transforming or distorting an¬
other. Sometimes this manipulation was violent and was especially aimed
at women. Man Ray, Andre Kertesz, and other photographers distorted
female body parts, as did Hans Bellmer, who twisted his grotesque dolls
into tortured, erotic poses. Balanchine’s use of men to handle women
never approaches this level of violence, but the repeated placing of fe¬
male dancers off-balance or lifting them in ways that look strained or
painful creates a sense of tension. For example, in the first theme duet,
the man holds the woman under the arms, and she hangs in a split, her
points touching the ground while the man drags her off stage. In an¬
other instance, this time in the third theme duet, the woman clings to
the man’s back as he drags her after him. Such poses and movements
produce a sense of unease that critics have noticed in various ways since
the ballet’s premiere. The men’s handling of the women, although per¬
haps no more than a means of exploring the choreographic potential of
decentering the body, was seen as violent by postwar critics.
On first viewing, Edwin Denby made comments that reflected The
Four Temperament's closeness to surrealist concerns in its emphasis on
fantasy and violence. After noting the denseness and power of the work’s
dance images, he spoke of the unpredictability and fantastic quality of
its sequences, which “crowd close the most extreme contrasts of motion
possible” ([1946] 1986:415). A few months later he expanded on his ini¬
tial reaction; “It appears to have the dispassionate ferocity of a vital pro¬
cess; its subject is the ‘four temperaments’ (or humours) of medieval
endocrinology and it suggests the grandiose impersonal drama of or¬
ganic energies. It is an impersonal drama that appears to be witty, cruel,
desperate, and unconsoling, like that of our time. Yet all that actually
Ballet’s Challenge 55

Jason Fowler and Jennifer Tinsley in the third theme from George Balanchine’s The Four
Temperaments, © The George Balanchine Trust. New York City Ballet production, 2004.
Photo: © Paul Kolnik. Courtesy of New York City Ballet.

happens onstage is rapid exact ballet dancing in classic sequences that


are like none you could ever imagine.” ([1947] 1986: 516).
Doris Hering of Dance Magazine said of the dances: “They have an
almost ruthless quality as though the bodies used in their fabrication
should be discarded for new ones after each performance” (1949: 34).
Several years later she noted: ‘Toz/r Temperaments [sic] is steely and capri¬
cious. Balanchine sends his dancers into violent front extensions that
end in buttery back falls. He twirls his girls off-center like crooked tops.
He pits tension against rag-like relaxation with endlessly exciting re¬
sults” (1953b: 10). The sense of violence, then, in The Four Temperaments
could be accounted for through such anticlassical elements as the ex¬
treme decenteredness of the movement, which causes a sense of tension
in the viewer; the harsh manipulation of the dancers (twirling the girls
like crooked tops); the extreme juxtaposition of movement (violent front
extensions, buttery back falls; rag-like relaxation and tension); and trans¬
formation and distortion of the body (crooked tops).
56 A Game for Dancers

Aiding the sense of violence and dislocation was the fact that the bal¬
let did not reinforce the moods indicated in the temperaments. Several
critics saw this as a failing of the ballet, including Walter Terry, who said
that although The Four Temperaments was rewarding for its movement
design, it did not “mirror the descriptive qualities listed on the pro¬
gram” (1947a; 4). This lack of discernible mood images may well have
made The Four Temperaments seem impersonal and cold, as Denby com¬
mented. However, there were other elements, as well, that contributed
to the work’s sense of violence—these were the ballet’s designs and its
finale.
Seligmann’s costumes were peculiar assemblages that sometimes
looked as if bandages had been added to tights and tutus, at others like
ragged medieval clothing with horned or mushroomlike headdresses.
According to Kirstein, “Seligmann swathed our dancers in cerements,
bandages, tubes, wraps, and tourniquets” (1973: 82). The set, too, sug¬
gested twisted bandages painted on a drop that resembled Seligmann’s
paintings. Altogether, the sets and costumes gave the stage a roiled, un¬
easy appearance. Then there was the ballet’s finale. Kirstein described
it as “a volcanic fountain or atomic eruption of bodies” (ibid.). As shown
in a 1946 film of a Ballet Society rehearsal, the dancers are grouped in
a tight circle. Beginning low to the floor, a wavelike movement sweeps
over them, rippling around the circle higher and higher until die dancers
are standing with arms raised. In the final moments, a male figure is
repeatedly thrown upward from the center in what seems slow motion.
If to these elements are added Denby’s description of the ballet as “an
impersonal drama that appears to be witty, cruel, desperate, and un¬
consoling, like that of our time,” the ballet begins to resonate with the
immediate postwar era’s concern with the impersonal horror of atomic
catastrophe. This aspect of the ballet, choreographed little more than
a year after the bombing of Hiroshima and Nagasaki, disappeared in
1951, and with it any trace of timely reference. By that year. Ballet So¬
ciety had become New York City Ballet and was firmly ensconced in
City Center. Balanchine replaced Seligmann’s sets and costumes with
practice clothes and a plain blue cyclorama, which served to make the
ballet less specific. He also changed the finale essentially to the one it
has today. In the present version, the female soloists are seen in a series
of split lifts in which they soar above the other dancers who form what
Arlene Croce called a “runway” for their flight (1977: 190). Croce re¬
flects the general opinion of today’s commentators in finding this end-
Ballet’s Challenge 57

ing stirring and hopeful, a very different reaction from the dark vision
critics saw in the ballet’s original form.

Balanchine’s use of surrealist elements in The Four Temperaments, from


collage to corporeal dislocation and the sense of violence it engen¬
dered, put his dance on the side of experiment for many viewers; in ad¬
dition, during the postwar years his ballets came to be seen as uniquely
American.
Modern dance had long laid claim to being the only American form
of high-art dance. This was in contrast to ballet, which was consistently
portrayed as foreign. Martha Graham made pronouncements through¬
out the 193os on what an American dance would be, and she assumed
that the dance she was describing would be modern dance. This new
dance, she said, would not be represented by one technique but by
many and would be characterized not by a series of steps but by a unique
rhythm and quality of movement, “a characteristic time beat, a differ¬
ent speed, an accent, sharp, clear, staccato” ([1937] 1985: 105). She de¬
scribed the American temperament as young, fresh, exuberant, with an
overabundance of youth and vigor. American dancing, she predicted,
would be woven from the threads of many old cultures, but “the whole
clotli will be entirely indigenous” (ibid.: too).
Denby attributed many of the traits Graham enumerated to Balan¬
chine’s dancers and choreography while at the same time being careful
to emphasize the classical form in which they were contained. In 1948,
for example, he wrote that Balanchine’s American balletic style was
strictly classical and supremely musical. His dancers were notable for
their long limbs, simple carriage, strength, speed, clarity, animation,
and technical and musical exactness (Denby [1948] 1986: 521-522).
Denby pointed out that similar traits were seen in “young Americans in
a ballroom or in a dance hall” (ibid.: 521). By 1952, he was more explicit
about Balanchine’s sources. He noted Balanchine’s use of “Negro and
show steps” and spoke of his “extraordinary absence of prejudice as to
what is proper in classicism” ([1953] 1986: 438). Other critics were less
specific about Balanchine’s Africanist sources, simply ascribing ele¬
ments that had long been appropriated as “American” to his work and
dancers. In 1948, Reed Severin wrote in Dance Magazine: “Balanchine
has exploited such typical American traits as speed, accuracy, a good ear
for rhythm, straightforwardness, athleticism” (1948: 43), while in 1950,
58 A Game for Dancers

Doris Hering said; “Mr. Balanchine has carefully and intelligently


exploited the native speed and high energy of his American dancers”
(1950: 12).
Balanchine’s use of Africanist elements also introduced a democratic
element into the danse d’e'cole that may have struck spectators as par¬
ticularly American. Denby spoke of Balanchine’s lack of prejudice in
transforming show and ballroom dance elements into classical ballet.
As noted earlier, the high art of ballet had always incorporated popular
and social dance but had usually separated high and low. Peasant dances,
buffoon dances, folk and national dances—all appeared on the stage but
were not allowed to mix with the danse d^ecole except as an occasional
decorative detail (a hand gesture, a step or two) to remind viewers of
the ballet’s theme or location. Popular and social dances had their place,
but they were not done on pointe and did not incorporate the ballet
vocabulary, although they might be balleticized through elements such
as comportment and virtuosity. Balanchine mixed low art with high,
changing not just steps but the comportment of his dancers. It is here,
in the carriage and presentation of the body, that Pierre Bourdieu finds
“political mythology realized, em-bodied'’’ (1990b: 69-70). Balanchine
let torsos bend and sag, hips jut and sway, feet move flat on the floor. At
the same time, however, these postures and movements, long associated
in Western culture with the “people” or the “folk,” were subsumed into
a high-art choreography that continued to reinforce the upright, the
symmetrical, the balanced, and the light. This movement, democratic
but aristocratic, was well suited for America’s ascent to world power. It
suggested dedication to equality and the common man while embed¬
ding these symbols in a courtliness associated with those who rule. At
the same time, the ballet’s disruption yet extension of tradition made it
appear an appropriate heir to European artistic domination. Here was
a dance that had not only absorbed all that Europe had to offer but had
then reformed the art, giving it new shape and purifying it of contami¬
nating elements, such as plot and pantomime.
Graham had said that American dance would have a sharp, staccato
beat. Stephanie Jordan has commented on the fact that pulse was an or¬
ganizing principle for Balanchine and that pulse drive was reflected in
his movement style. The dancers tended to move, in Denby’s words,
“on top of the beat,” with a rhythmic thrust that was quick and exact.
Jordan connects this “discipline of pulse” with modernist musical think¬
ing, which played down emotional qualities for control, regularity, and
Ballet’s Challenge 59

precision (2000; 112-120). Certainly critics commented on the emo¬


tional coolness of Balanchine’s dances. As late as 1958, the subject was
still an issue. Dance Magazine published an article by Eugene Palatsky
that year in which he wrote that New York City Ballet was controver¬
sial in part because of its perceived lack of emotion. In interviews with
a number of the company’s dancers, Palatsky returned repeatedly to this
issue. Balanchine, himself, considered detachment a definitively Amer¬
ican characteristic. In an article for Dance in 1937 he wrote:

Superficial Europeans are accustomed to say that American artists have


no “soul,” no personal or national style. This is wrong. America has its
own spirit—cold, crystalline, luminous, hard as light. American dancers
have a genuine quality or “soul,” but, unfortunately, they rarely possess a
technique capable of transmitting it.
It is perhaps the most remarkable soul of all because it is capable of
giving an exposition of any material without identifying or diluting it by
subjective identification, either national or personal.
Good American dancers can express clean emotion in a manner which
might be almost termed “angelic.” By angelic I mean the quality sup¬
posedly enjoyed by angels which, when they relate a tragic situation, do
not themselves suffer.
To be the most perfect instrument without drawing attention to one’s
personal notion of the chosen composition is not a meager task, but a
heroic service, however anonymous. (1937: 13)

Cool detachment was not something that was ordinarily attributed to


the American character, and this quality, which Balanchine assigned
to the American dance “soul,” may actually have been due to more
mundane reasons. The early New York City Ballet dancers were young
and, except for the principals, had little stage experience. Many were
hardly more than students. Tanaquil LeClercq, a sixteen-year-old who
had been plucked from the School of American Ballet, spoke about the
impossibility of adding interpretation to the Choleric variation while
meeting the demanding speed of the steps (LeClercq 1982: 153). The
young dancers’ inability to charm with stage personality, and perhaps
to do little more than master the technical demands being made on
them, was transformed by Denby and others into American honesty
and straightforwardness, and then into a national style.
Although critics saw elements such as musicality, athleticism, youth-
6o A Game for Dancers

fulness, and energy in Balanchine’s dance and saw tdiem as uniquely


American, he had, in fact, long been interested in youth and virtuosity,
as well as Africanist rhythm and movement. As Banes and Gottschild
pointed out, Balanchine had inserted Africanist elements into his work
while still serving as a choreographer for Diaghilev in the 1920s. As for
youth and virtuosity, in 1925 Balanchine had cast a fourteen-year-old
Alicia Markova as the Nightingale in Le Chant du Rossignol for Diaghi¬
lev. When he joined Les Ballets Russes de Monte Carlo in 1932, he had
astonished Europe with his virtuosic “baby ballerinas”—Irina Baronova,
Tamara Tbvmianova, and Tatiana Riabouchinska—^who were in their early
teens and who were said to be able to accomplish amazing technical
feats (Kochno 1954: 293; Taper 1984: 136-137; Buckle 1988: 58-59).
Thus, Balanchine did not become interested in youth as a uniquely
American attribute; he had long favored young dancers.
One of the reasons for his interest in youth was that he was con¬
cerned with molding and shaping dancers, and the younger they were,
the more likely he was to succeed. On coming to the United States, Bal¬
anchine clearly realized that American dancers needed better training if
they were to perform his choreography. His insistence on having a school
before he began to turn his attention to a company supports the argu¬
ment for his focus on technique. In the article from Dance News quoted
above, he mentioned that American dancers for the most part were in¬
capable of transmitting their own native qualities. As both Tanaquil
LeClercq and Maria Tallchief have testified (LeClercq 1982; Tallchief
1997) Balanchine spent many long hours working with his young Ameri¬
can dancers to improve their technique. In the early postwar years, he
was particularly concerned with bringing them to a high degree of
technical proficiency quickly in order to be able to compete on an inter¬
national level. Balanchine nearly always taught the company class each
day but, according to LeClercq, before New York City Ballet’s first tour
to England in 1950, he additionally worked with his leading dancers
each night to improve their technique (LeClercq 1982: 167).
In training dancers, Balanchine continually sought greater speed
and attack as well as higher extensions (for virtuosity), more precision
and greater turn-out (for clarity of body image). Again, these were not
interests that originated in the United States but were given impetus
in America because he had greater control over his dancers’ training
and, after 1948, a stable company with which to work. These elements,
which Balanchine emphasized as part of his own interests and disposi-
Ballet’s Challenge 6i

tions, reinforced aspects of what were already considered American


attributes. This is not to say that Balanchine also did not view such at¬
tributes as typically American, but that they existed in his work long
before he entered the United States.
As Balanchine’s dance was increasingly seen as American, Balanchine
himself became more American in the written discourse of the period.
During the 1930s and the early 1940s, he had been treated as a mem¬
ber of the Franco-Russian ballet world, one of several such choreogra¬
phers who were working in America. His foreignness was emphasized
in the popular press through crass imitations of his accent, such as the
following quote from an article in Colliers magazine; “His name grew,
as he puts it, ‘more and more famouser’” (Davis and Cleveland 1940:
22). However, as Balanchine came to be accepted as the “creator” of an
American ballet, his foreignness was blurred. America’s tradition of
immigration and assimilation made Balanchine an American through
citizenship. It is notable that many of the articles written about him in
the postwar period referred to the notion that he had become Ameri¬
can through naturalization. For example, Allen Churchill wrote in The¬
atre Ans in 1949; “When America’s contribution to ballet is some day
added up, the development of ballet as pure dance may be designated
as American ballet, just as the contributions of Fokine are labeled Russ¬
ian. Balanchine, in this coimtry sixteen years, has been an aggressively
American citizen for nine” (1949: 35). Churchill tied Balanchine and
abstraction to a definitive American dance, and he made Balanchine
American. In an even more astonishing sleight of hand, he compared
this American choreographer (and the American dance he had devel¬
oped) to Fokine, a Russian. Not surprisingly, Kirstein, too, promoted
the notion of Balanchine’s Americanness. In an article written in 1947,
he noted Balanchine’s American citizenship and said that he had re¬
cently mounted a work for the Paris Opera Ballet, the first person from
“his country,” that is, the United States, to do so (1947a: 37).
Balanchine himself aided the notion of his Americanness in various
ways. In 1946, he married ballerina Maria Tallchief, whose father was a
Native American. Tallchief speaks in her autobiography about Balan¬
chine’s interest in her family and the Oklahoma reservation where she
spent part of her childhood. “He claimed that by marrying me he finally
felt he was a real American, and he compared us to John Smith and Po¬
cahontas” (1997: 119)-
In the late 1940s, some critics saw in Balanchine’s ballets a modernism
62 A Game for Dancers

they felt was being compromised in modern dance. However, although


Balanchine’s work took on some of the attributes of vanguard art, it did
not share vanguard goals. For as much as Kirstein spoke out against
commercialism and complained of New York City Ballet’s poverty, he
well understood how necessary it was to find support among establish¬
ment institutions in order for a ballet company to succeed. He had
spent his life among people of influence, and he knew how to function
within that social framework. In 1948, when Ballet Society became New
York City Ballet under the aegis of New York City Center of Music
and Drama, Kirstein’s intentions became clearer. In the 1950s, New
York City Ballet proved to be an exemplary representative of an elitist,
cultural arm of institutional power. Balanchine’s choreography em¬
bodied America’s new position of world leadership; his choreography
built on an imperial past that had been democratized and given youthful
energy. At the same time, his emphasis on the lack of specific meaning
in his work guaranteed its freedom from the taint of political critique.
Balanchine, himself, was the embodiment of the immigrant American
who found success on the shores of the land of opportunity. And added
to that bit of American mythology was the fact that he had turned his
back on Soviet repression for the freedom of the West, part of Ameri¬
can Cold War ideology.
Balanchine’s dances were exploited in the postwar years as a sign of
America’s superiority to its European allies in the West and its Soviet
enemy in the East. In 1952, New York City Ballet appeared in the Paris
festival, “Masterpieces of the Twentieth Century,” sponsored by the
Congress for Cultural Ereedom, a powerful anticommunist organiza¬
tion secretly sponsored by the CIA (Saunders 1999: 113-128; Kirstein
1973: 118; Chujoy 1953: 318-319, 341-345). Among the organizers of
the exposition was Nicholas Nabokov, the composer and close friend of
Balanchine. City Ballet also was among the early recipients of federal
government largesse when, in 1954, the State Department began to
sponsor overseas tours by American “cultural ambassadors” (Prevots
1998). Kirstein was conveniently on the committee of dance experts that
made recommendations on which companies should be sent abroad.
New York City Ballet visited the Soviet Union in 1962, and during the
tour the notion that Balanchine had renounced Russia for freedom in
the West was a common theme in the American press.
Balanchine choreographed several suites of dances dedicated to Amer¬
ican themes in the postwar years, among them Western Symphony (1954)
Ballet’s Challenge 63

and Stars and Stripes (1958). Stars and Stripes shocked Dag Hammar-
skjold, secretary-general of the United Nations and Kirstein’s friend, who
considered the ballet jingoistic in its exuberant celebration of American
patriotism. Kirstein tried to pass off the work as a parody of hyper-
Americanism, “but Hammarskjold judged that the cause of peace among
nations was not served by the blatant effrontery of our imperialist ges¬
ture” (1973: 151-152). Given that New York City Ballet chose to dance
Stars and Stripes at Nelson Rockefeller’s inauguration as governor of
New York in January i960, it is unlikely that the ballet was meant as a
parody of patriotism or that spectators viewed it as such. But by that
time. New York City Ballet had long represented establishment culture,
both at home and abroad.
Nonetheless, for a time the Kirstein-Balanchine enterprise ruffled
the waters of vanguard dance, and it made an impact on the genre with
which it chose to compete. Kirstein invited Merce Cunningham to
create The Seasojis for Ballet Society and Iris Mabry to show her work
there, and it is notable that both favored an objectivist dance. Cun¬
ningham, of course, studied and taught at the School of American Bal¬
let. Other modern dancers who either taught or studied there in the
1940S and ’50s included Louis Johnson, Anna Sokolow, Janet Collins,
and Paul Taylor. Erick Hawkins danced for Balanchine in the 1930s and
taught during that time at the School of American Ballet; in the 1950s,
he turned to an objectivist dance after years of working in an expres-
sional mode. Balanchine came in contact with more modern dancers
through his musicals, including Katherine Dunham and Talley Beatty.
Most important, though, was Balanchine’s nonliteral approach to dance,
which came at an opportune moment for a young generation of mod¬
ern dancers who, by developing an objectivist dance of their own in the
1950s, created what would be declared a new modern dance vanguard.
3

MODERNIST THEORY:
JOHN MARTIN, EDWIN
DENBY, JOHN CAGE

T hree of the most influential dance theorists in the postwar years


were John Martin, Edwin Denby, and John Cage. Martin, in addi¬
tion to his role as critic for the New York Times, wrote important theo¬
retical works in the 1930s that helped define the aims of modern dance.
In the 1940s, he consolidated this theory and in die 1950s contributed
an important study on contemporary ballet. Denby was, with Lincoln
Kirstein, the first to promote neoclassicism as an American ballet style,
and Balanchine as the exemplar of that style. His writings on dance
were published in book form in 1949 and made an immediate and last¬
ing impact on the field. Cage, although best known as a vanguard com¬
poser, also was an important influence on modern dance. He wrote
only four essays on dance, but he made his ideas known through his
musical theory and also directly to dancers, for whom he composed
music. While Cage was most closely associated with Merce Cunning¬
ham, he provided scores for a wide range of dancers during this period,
including Jean Erdman, Nina Eonaroff, Valerie Bettis, Pearl Primus,
Marie Marchowsky, Yuriko, Merle Marsicano, Louise Lippold, Iris
Mabry, and Alwin Nikolais.^
These three theorists are often presented in opposition to each other:
Martin promoting a dance that communicated psychological essences,
Denby and Cage advocating a dance of so-called pure movement.
Denby also is viewed as supporting ballet, in contrast to Martin and
Cage, who favored modern dance. However, although they had their
differences, I want to suggest that all three shared a modernist vision
that differed primarily in degree, and which helps account for who they
Modernist Theory: Martin, Denby, and Cage — 65

supported and why. Their modernism led them to advocate a dance that
attempted to imdermine rationalization through a corporeal intelli¬
gence. To this end they shared a faith in the dancing body’s ability to
demonstrate freedom from systems and processes governed by means-
ends calculation.

John Martin (1893-1985) was born in Louisville, Kentucky.^ His father


was a purchasing agent for the Louisville and Nashville Railway, and his
mother was a singer. He attended the University of Louisville where his
major subject was classics, then studied violin at the Chicago Conser¬
vatory. He subsequently worked with the Chicago Little Theatre. After
World War I, in which he served in the Army Air Force Signal Corps,
Martin moved to New York, where he became editor of a trade publi¬
cation, the Dramatic Mirror. He was the executive director of the Richard
Boleslavsky Laboratory Theatre from 1924 to 1926 and occasionally
directed summer stock productions in New York and Pennsylvania
until 1934.^ From 1927 to 1962, he was the dance cxiticior the. New York
Times. In addition to the hundreds of reviews Martin wrote in the
course of his thirty-five-year career, he also published a number of in¬
fluential books.
During the 1930s, Martin had been an outspoken advocate for mod¬
ern dance, positioning it at the forefront of American high-art dance
in The Modem Dance (1933), America Dancing ( 1936), and Introduction
to the Dance (1939). Martin’s ideas were similar to those of Louis Horst,
with whom he came in close contact during the 1930s. Both taught or
lectured at Bennington College, the summer home of the modern
dance pioneers. Both sought to legitimize modern dance’s place within
modernism and to solidify its opposition to ballet, the traditional form
from which it had revolted. However, Martin’s theory was far more
developed than Horst’s and less concerned with the craft of dance
composition.
Introduction to the Dance includes the major part of Martin’s early
theory. Like much of Martin’s writing, it was aimed not at scholars but
laymen. In this instance he had two major purposes: the first was to con¬
vince readers that all they had to do to enjoy dance was look at it with¬
out obstructing preconceptions; the second was to convince them of the
superiority of modern dance to other forms. Consequently, throughout
the book he often spoke of dance in general terms when he meant an
66 A Game for Dancers

John Martin. Photo: Leo Lerman.


Courtesy of Jerome Robbins Dance
Division, The New York Public Library
for the Performing Arts, Astor, Lenox
and Tilden Foundations.

authentic dance, best characterized by modern dance or the roots and


aims of modern dance. An authentic dance had a social purpose, which
was to communicate essential truths. Martin posed this authentic dance
against another, which he called by various names but which was one so
dominated by form that it robbed dance of its social element. Formal¬
ism was the great destroyer, severing dance’s ties to society and driving
artists to an art-for-art’s-sake mentality that only reinforced art’s sepa¬
ration from its communal source ([1939] 1965: 15-17). Martin credited
formalism’s strength to capitalism with its division of labor and its re¬
duction of all relationships to economic ones. Capitalism divided art,
like religion, of which it had been a part, from a daily life now defined
in purely economic terms. It was the task of an authentic dance to re¬
connect art and life. Thus, Martin took the modernist position that an
authentic dance stood against rationalizing processes, of which formal¬
ism was one result.
Martin’s first topic in Introduction to the Dance was the nature of
movement. He asserted that dance was “the very stuff of life” (ibid.: 31)
because its medium was human movement and its instrument was the
Modernist Theory: Martin, Denby, and Cage 67

human body. Martin posited a link between movement and psychic


states, which in an earlier volume he had called “metakinesis,... this
correlation growing from the theory that the physical and psychical
are merely two aspects of a single underlying reality”(ti933] 1972: 13).
Emotions were the result of human needs, drives, and desires. They
were a “stirring up” caused by the action of the vital organs and nervous
system in response to human contact with the environment ([1939] 1965:
36-40).
Dance as art was a mode of human interaction in which the dancer
communicated feelings (manifested in movement) that the spectator
actively assimilated. The assimilation process did not take place on an
intellectual level. Rather, it occurred corporeally without recourse to
conscious thought. It was made possible by the fact that all humans
possessed what Martin identified as a movement sense. This sense was
physiologically based, working through motor and neural mechanisms.
It was through these mechanisms that, for example, the body remained
balanced as changes occurred in position and weight. But the body did
not only respond to environmental encounters in this limited way; Mar¬
tin claimed that each bodily action also produced an emotion about
it. This emotional response to movement embedded itself in a “motor
memory of emotion.”'^ “Every emotional experience tends to make what
we might call records of itself in motor patterns, setting up more of those
well-worn paths in the neuromuscular system and adding new phases to
those already set up” ([1939I 1965: 47).
Related to this storehouse of emotional memory was what Martin
called “inner mimicry.” This was a sympathetic motor response that
occurred when an individual reacted to an object or another living being’s
action in a particular way because of having experienced something simi¬
lar in the past. An individual could also extrapolate from life experience
to emotionally relate to other situations not directly experienced. So
one might unconsciously tense one’s own muscles and then feel a sense
of fatigue at the sight of a man or woman carrying a heavy burden.
Motor memory and inner mimicry were key to the spectator’s ability to
respond to meaning in movement.
Dance communication, then, was corporeal; it was carried out body-
to-body rather than through any sort of intellectual process. Commu¬
nication began in life experience. Life experience, for Martin, did not
consist of daily events; rather it was made up of the psychic states pro¬
duced by physiological impulses that occurred when human beings met
68 A Game for Dancers

their environment and that became embedded in the motor memory of


emotion. In an authentic dance, dancers drew on their own life experi¬
ence, which they then abstracted and which spectators, in turn, assimi¬
lated through their life experience. Speaking of how a young choreog¬
rapher might proceed in light of this theory, Martin pointed out that
the solution to the problem of choreography “lies in turning directly to
the fact that the body reacts to all stimuli first in terms of movement,
and that communicative movement suitable for dance can be drawn
only from what might be termed his [the dancer’s] motor memory of
emotion. He must learn how to call upon his emotional associations
and translate them into action directly from life experience” ([1939]
1965: 86). However, Martin continued, although the artist communi¬
cated feeling or emotion to spectators,

it is not the dancer’s, or any other artist’s, purpose simply to arouse us to


feel emotion in a general sense, to stir us up to no end. It is his purpose,
rather, to arouse us to feel a certain emotion about a particular object or
situation. He wants to change our feeling about something, to increase
our experience, to lead us from some habitual reaction which he has dis¬
covered to be perhaps merely inertia or otherwise limited and restrictive,
to a new reaction which has an awareness of life in it and is liberating and
beneficent. ([1939] 1965: 53)

An authentic dance challenged the status quo, thereby increasing the


viewer’s awareness of the possibilities in life in ways that were liberat¬
ing. It was the dancer’s task to introduce into the community an aware¬
ness of freedom otherwise prevented by habits, limitations, restrictions,
or inertia.
The artist accomplished this vital task through a process of abstrac¬
tion and formal organization that allowed the dance to communicate to
viewers. The assumption here was that if life experience was not or¬
ganized and abstracted through form, it was not communicable, or as
Martin termed it “intelligible.” The process of abstraction was a crucial
element in Martin’s theory because it was the means by which the dance
was given universal significance. The artist abstracted from personal
life experience, winnowing away inessential elements so that the dance
could communicate to others: “The dancer’s movements are abstract;
that is, they have abstracted the essentials from a particular life expe¬
rience, omitting all that is merely personal and without universal sig-
Modernist Theory; Martin, Denby, and Cage 69

nificance” ([1939] 1965: 89). Although the dancer abstracted specific


elements from her movement, the resulting dance was not in itself ab¬
stract in the sense of being a purely formal design. The reason for this
was, first, that the dance embodied rmiversal experience and, second,
that “the body is totally incapable of becoming an abstraction itself or
of producing movement that is abstract in the sense of divorced from
behavior” (ibid.: 63). The dance was always connected to human life
through the body. Those connections may have been obscured, twisted,
or otherwise mystified by rationalized theories and academic formulas,
but they continued to exist.
Always fearful of such intellecmal mystification, Martin contended
that the process of imposing form on experience was simply a matter of
craft. In the case of dance, the choreographer must imderstand how to
manipulate movement in terms of space, time, and dynamics. The busi¬
ness of form was to shape material so that it fiilfilled a specific purpose
([1939] 1965: 57). In this sense the artist was like a workman who knew
how to build a functional chair, except that instead of wood or leather
the choreographer’s material was movement. Although artists had to be
well versed in how to shape their materials, spectators needed to know
little about it in order to respond to art. In fact, Martin said, complex
theories only blocked communication between the artist and spectator.
Martin included sections on materials, music, and drama in Introduc¬
tion to the Dance to give his readers a clearer understanding of the place
these elements held in dance making. His discussion of the relationship
of dance and drama is of greatest significance here because of the de¬
bates on narrative in postwar modern dance. Martin treated drama as
narrative in the sense of its being a series of causally related events. He
came down strongly against drama in an authentic dance while at the
same time conceding drama’s ubiquitousness within the dance field as a
whole. He noted that although dance and drama were closely related,
“dramatic form as such can be said to exist when the dance, instead of
presenting the essence of an emotional experience, deals with a specific
sequence of events out of which such an experience grows. The more
literal its treatment, the less it has of dance about it” ([1939] 1965: 84).
Dramatic form was found in many types of dance, but since modern
dance dealt with essences, drama was not appropriate to it.^
After speaking of form and composition in dance, Martin turned to
the subject of style, paying particular attention to what he considered
the three major historical styles found on the Western stage. The earliest
70 A Game for Dancers

one of these was romantic, the source of which was the emotions. Ro¬
manticism valued feeling, spontaneity, informality; it grew out of the
people, was democratic, colloquial, sometimes chaotic. Classicism de¬
fined and eventually petrified romantic style; it was concerned with rules,
codes, and categories; it organized and defined what had gone before.
Eventually classicism became mired in academicism and weakened by
lifeless formulas. Then decay set in, or revolution, and the process began
again ([1939] 1965: 110-121).
Modernism, the third historical style, combined elements of roman¬
ticism and classicism since it was characterized by a drive toward func¬
tional form. Modernism was a product of industrial society, in that
modern technology relieved the artist of having to imitate visible real¬
ity. This allowed the artist to delve below the surfaces of life to seek out
“the subjective roots of experience” ([1939] 1965: 123). Then, abstract¬
ing from this experience the artist was able to create an autonomous
work of art, embodied life experience stripped of irrelevant elements.
This embodiment of truth circumvented an otherwise rationalized
world. “Here lay the complete answer to representationalism, the
complete defiance of the machine in art” (ibid.).
Because the modern artwork was not an imitation, the artist could
use materials in a less disguised way than in the past. Sometimes this
attention to materials remained focused on finding and exposing essen¬
tial forms. In this case it was classicist because it was not concerned with
life experience. Martin thought that although pure form could perhaps
exist in painting and music, it could not in dance because “the body can¬
not be separated from implied intent” ([1939] 1965: 125). So although
Martin’s description of an autonomous art in some ways resembled
Clement Greenberg’s, Martin, significantly, tied dance to life, from
which it could not be separated because of its corporeal instrument.
This difference suggested that dance had a more necessary relationship
to society than did painting. Martin said as much when, as noted earlier,
he stated that an authentic dance’s function was not simply to stir up feel¬
ings for no purpose, but to change the viewer’s awareness of the world
in a way that was liberating and beneficial (ibid.: 53).
In the second part of his book, entitled “Dance in Action,” Martin
took up various ways in which Western society uses dance. In his dis¬
cussion of theatrical dance he made a distinction between “spectacular”
and what he called “expressional” dance. The former was meant to be
watched objectively, while the latter was participated in vicariously. Spec-
Modernist Theory; Martin, Denby, and Cage 71

tacular dance was sensuous and intellectual in its appeal; it dealt with
surfaces and was part of the category of entertainment in that it was cre¬
ated purely for pleasure. Expressional dance stemmed from the same
sources as religion, magic, and social rimal and was meant to awaken
emotional perceptions ([1939] 1965: 173-174). Martin placed this dance
in the category of art. In the twentieth century, the binary of spectacu¬
lar and expressional dance was best represented by ballet and modern
dance, respectively.
In his chapter on spectacular dance, Martin surprisingly did not argue
that ballet, as a form dependent on surfaces, sensuous pleasure, and in¬
tellectual appreciation, was incapable of achieving authenticity. Since
he had posited that authenticity was dependent on the process of ab¬
straction, which rendered a particular kind of dance autonomous, Mar¬
tin contended that ballet could become autonomous, and thus authentic,
by eliminating its dependence on decor (painting) and narrative (litera¬
ture) and by returning to a “geometrical-aesthetic basis” of movement.
Martin said that ballet’s classicism was quite different from purely visual
decoration or lifeless mechanics because it idealized the human body.
Presumably it was this idealization that set ballet apart, for example,
from a chorus line with its mechanical synchronization. Corporeal
idealization was at the heart of ballet aesthetics and the source of its
drive. Ballet was

the presentation of its ideal essence freed from the encumbrances of a ra¬
tionalistic universe of cause and effect, a pragmatical universe of organic
drive and utilitarian function. It exemplifies the personal achievement of
abstraction, of transcendent self-containment. In its idealization of the
body it strips away all necessity for practical accomplishment, turning
certain of its conformations to use more harmonious than the functional
processes that have shaped them, and superseding where possible even
structural elements which have been bred by utilitarian demands alone.
([1939] 1965: 212)

By devoting itself to a glorification of the body for itself, rather than as


a unit of labor, ballet freed itself from canse-and-effect rationalization.
Martin’s defense of ballet is startling, considering his insistence on
the notion that an authentic dance communicated essential life experi¬
ence. Ballet, as a spectacular form, had no interest in penetrating to
experiential roots. Martin based his defense on the notion of balletic
72 A Game for Dancers

abstraction, that although ballet was concerned with form, it was es¬
sential form. Thus, the dancer’s body achieved an abstraction that was
greater or more harmonious than the rational processes that had fash¬
ioned it. Martin also maintained that dance always contained implied
meaning because its medium was the body and, consequently, it could
not be completely separated from life. This idea considerably weakened
his argument against formalism, which Martin implicitly linked to bal¬
let and which he claimed severed the ties between art and life. However,
it gave to ballet some possibility of meaning. Yet ballet, since it was ded¬
icated to form alone, made only a kind of accidental use of corporeal
meaning. It lacked the power to embody, as Joseph Campbell had said,
“the invisible pattern of the psyche reflected in time and space”
(1944b: 66). Since an autonomous ballet could only circumvent cause-
and-effect logic through its corporeal idealization, it would continue to
have little or no power to change society. This enabled Martin to main¬
tain modern dance’s superiority, as it was dedicated to functional form.
Martin’s argument for an authentic ballet may have been strained,
but in 1939 that was not of crucial importance because, in any case, he
saw no indication of ballet’s moving toward modernist autonomy. He
looked sympathetically on the work of Michel Fokine as a choreogra¬
pher who had attempted to reform ballet through a synthesis of “real¬
istic life impulses” and classical means. However, Martin felt that Fokine
had led ballet as far in the direction of synthesis as possible; the next
step would have been to abandon the classical vocabulary altogether
([1939] 1965: 210). As for Diaghilev, Martin considered his modernism
faulty because it was not based on dance but developments in painting
and music. Nor was he impressed with Balanchine, whose aesthetics, as
noted earlier, he labeled anemic and whose work he considered part of
a decadent Franco-Russian tradition.
By 1952, when he published The World Book of Modem Ballet, Martin
had radically changed his view of ballet’s likelihood of becoming an in¬
dependent form. He had also become convinced that Balanchine was
the choreographer to lead hallet toward that goal. Now he argued that
Fokine’s experiments, though admirable, had been useless. To locate the
correct trajectory for ballet, one had to go back to Marius Petipa, the
great classicist of the nineteenth century, who realized that ballet could
not tell stories and attempted to solve the problem by alternating dance
with pantomime, thus separating the two. In the mid-twentieth century
it was Balanchine who, in his development of the plotless ballet, had as-
Modernist Theory: Martin, Denby, and Cage 73

sumed the mande of Petipa. It was he who had most consistendy purged
ballet of the unnecessary elements of narrative and pantomimed illus¬
tration to become a neoclassicist and a true reformer.
As Martin’s enthusiasm for Balanchine grew, certain elements of his
interwar theory fell by the wayside. Among them were his attacks on
formalism and his frequent, thinly veiled conflation of formalism and bal¬
let. In his postwar writing Martin spoke little of formalism as a corrupter
of authentic dance. He rarely mentioned the notion that formalism pro¬
moted an elite form that robbed dance of its social function and that
robbed the public of dance by making it impossible for all but a small
group of cognoscenti to imderstand. Now he spoke instead of Balanchine
and Kirstein’s refusal to surrender to market interests (1952: 14, 125-
158). Balanchine’s work was difficult; it did not court audiences, which
indicated that it was more than easy entertainment. The difficulty of his
work also put Balanchine on the side of innovation and experiment. Ac¬
cording to Martin, Balanchine had replaced the flair of the old Russian
ballet with substance. Significantly for the United States and its post¬
war policies of cultural domination, Balanchine, according to Martin,
had also made New York the capital of the dance world (1952: 14).
It is worth noting that Martin’s defense of Balanchine was written in
the early 1950s, while Introduction to the Dance was published at the end
of the 1930s. In those dozen years not only had the political landscape
shifted as markedly as Martin’s opinions, modern dance was facing
problems that had not yet surfaced in the late 1930s. Martin’s support
of Balanchine went hand in hand with his critique of postwar modern
dance, which reflected the genre’s travails at the time. Martin continued
to support Graham and Limon, but now he depicted them as the last
standard-bearers of a true modern dance. He contrasted their authen¬
ticity with the younger generation’s lack of ability to embody emotional
essences. He also ignored the fact that Graham’s and Limon’s postwar
choreography owed far more to dramatic form and characterization
than had dances of the 1930s. Yet, this position allowed Martin to sup¬
port what he claimed to be an authentic modern dance while faulting the
younger generation of modern dancers and defending the neoclassicism
of Balanchine.

Edwin Denby (1903-1983) had none of Martin’s difficulties reconciling


ballet with social function because his ideas about the purpose of dance
74 A Game for Dancers

and how it fulfilled that purpose were quite different. At the same time,
however, Denby’s thinking was similar to Martin’s in a number of ways.
A poet as well as a critic, Denby was an eloquent spokesman for an
American ballet during the crucial years of the 1940s. His upbringing
could not have differed more from Martin’s (MacKay 1986: 11-34).
Born into a family of diplomats, Denby was raised in Asia, Europe, and
the United States. He was a brilliant student, entering Harvard at six¬
teen. Halfway through his second year he left, confused and unsure of
what he wanted to do except in some way to write. In 1923, he traveled
with a friend to Vienna where he eventually discovered the Hellerau-
Laxenburg School, devoted to the teaching of Emile Jaques-Dalcroze,
the father of eurythmics, and also a center of Ausdruckstanz. Denby
studied Kbrperbildung (physical development) at the school, graduating
three years later. He then worked as a dancer-choreographer based in
Germany and Switzerland. He also visited the Soviet Union where he
saw much of the vanguard theater that was taking place there in the
1920s, as well as the Bolshoi Ballet. And in Paris he saw Balanchine’s
short-lived Ballets 1933, which made as profoimd an impact on him as
it did on Lincoln Kirstein.
Returning to the United States in 1935, Denby settled in New York
where he soon began writing dance reviews for Modem Music. He also
rekindled old friendships with artists he had known in Europe such as
Aaron Copland, Virgil Thomson, and Paul Bowles, and he made new
friends with the abstract expressionist painters Arshile Gorky and Elaine
and Willem de Kooning, poet Frank O’Hara, actor Orson Welles, and
a host of other New York artists and intellectuals. From 1943 to 1945, he
replaced Walter Terry as dance critic for the New York Herald Tribune.
Most of the rest of his life was devoted to writing poetry, with occasional
essays and reviews on dance. His criticism was collected in Looking at the
Dance (1949) and in Dancers, Buildings and People in the Streets (1965).
Much of Denby’s theory is summarized in five essays written in the late
1940S and early 1950s: “Ballet: The American Position” (1947), “A Brief¬
ing in American Ballet” (1948), “Against Meaning in Ballet” (1949), “A
Letter on New York City’s Ballet” (1952), and “Some Thoughts about
Classicism and George Balanchine” (1953). However, he elaborated his
position in numerous reviews and in comments throughout his writing
in the 1940s and ’50s.
Denby’s focus in his writing and his standard in dance was ballet,
as Martin’s was modern dance. Denby would not have differed greatly
Modernist Theory: Martin, Denby, and Cage 75

Edwin Denby. Photo: Jerome Robbim. Courtesy of the Robbins Rights Trust and Jerome Robbins Dance Division,
The New York Public Library for the Performing Arts, Astor, Lenox and Tilden Foundations.

from Martin in his assessment of ballet as an art that emphasized form


and spectacle, nor would he have disagreed with Martin on the need for
ballet to return to its classical roots in order to become autonomous.
Classicism for Denby consisted, above all, of a dance in which move¬
ment was the driving force ([1953] 1986: 438). Classical ballets might
have stories, he said, but they were unimportant; audiences went to see
the dancing, not to wimess fascinating plots unfold. As its means, clas¬
sical ballet employed the vocabulary, comportment, and placement of
the danse d'’ecole, a virtuosic technique that allowed the body to be seen
with the greatest possible clarity. It was this orientation toward move¬
ment in combination with the danse d'ecole that characterized the classic
dance. Where Denby differed from Martin was that, for him, a ballet
stripped of associations with literamre and painting was not only enough
to give it autonomy, it was all that was needed in dance. Modern dance’s
attempt to communicate essential experience was simply adding liter¬
ary elements to an art that did not need it. Denby also decided, long
76 A Game for Dancers

before Martin, that Balanchine was the choreographer who was most
completely dedicated to returning ballet to the classical values lost after
Petipa. But Denby went farther, defining the quintessential American
balletic style as virtually synonymous with his idea of Balanchine’s neo-
classicism. Denby’s theory and position become clear when we examine
the ways in which he contrasted the old European with the new Amer¬
ican ballet, classicism with other ballet styles, and ballet with modern
dance.
Denby was writing when native ballet companies and dancers were
only recendy established. Ballet Theatre, the first major American
company, was founded in 1940. The Ballet Russe de Monte Carlo setded
in the United States during the war and gradually replaced its European
personnel with Americans, becoming in essence an American company.
Erom 1944 to 1946, Balanchine served as informal artistic director of
the Ballet Russe before taking command of New York City Ballet when
it emerged out of Ballet Society in 1948. Within this context of recendy
arrived American ballet, Denby defined an American ballet style in a
way that served to differentiate it from the older Eranco-Russian style
typified by the Ballet Russe companies of the i930s.'^ Speaking of the
Ballet Russe style at its height, Denby described it as “more hot-and-
bothered, its rhythm in dance scenes made more sweeping climaxes, its
techniques looked more casual and undefined, and its temperament was
more exotically fiery” ([1947] 1986: 510). In contrast, American danc¬
ing, he wrote, was large, clear, accurate, and unaffected. It was not very
personal or emotional. The Europeans were better at “effects of imagi¬
native impersonation, of imaginative atmosphere or stage presence”
(ibid.). In other words, the Europeans excelled at those elements that
were less dependent on movement than on acting and personality. They
were less technically accurate, more affected, perhaps less honest and
straightforward than the Americans, who depended on movement alone
to create their effects. The problem with this argument was that there
were many Europeans in American companies. Balanchine was living
and working in New York; Ballet Theatre had European stars, such as
Alicia Markova and Anton Dolin, both of whom had danced with Di-
aghilev’s company in the 1920s. Denby’s answer was that these artists
had been transformed in the United States, absorbing American char¬
acteristics. European artists who could not change, failed ([1948] 1986:
518-521).
Once Denby had placed American ballet at the center of the dance
Modernist Theory: Martin, Denby, and Cage 77

field, he bisected the field again by identifying two branches of a dis¬


tinctively American kind of ballet choreography. One was the neoclas¬
sicism of Balanchine; the other, the psychological ballet of Antony Tudor
and the American folk ballet of Agnes de Mille and Jerome Robbins
([1947] 1986: 512-513).^ Denby’s argument was that the dance of de
Mille and Robbins was dependent on literary means rather than move¬
ment for its full effect. Contrasting the two kinds of ballet, he said that
Balanchine’s work was more difficult to discuss because it was based on
“dance qualities” rather than literary ones: “It is easier to talk about
the novelty of our gay local-color Americanism in ballet or about the
gloom-steeped psychological aspects of Tudor’s gripping large-scale
dramas of frustration. These pieces make a good deal of their appeal
through their literary content, and from a literary point of view the
value of them in nationalizing our ballet or in modernizing it has been
properly stressed by many reporters” (ibid.: 512).
In other analyses, Denby pointed out that the American local-color
ballets were derived from pantomime and character-dance elements.
He found Tudor’s works more ambitious and complex than those of de
Mille and Robbins, but equally dependent on literary elements. Tudor’s
ballets were “full of passion, of originality, of dramatic strokes, of ob¬
servation, of brilliant pantomime ideas and fastidiously polished detail”
([1948] 1986: 524). Denby further commented on the weak formal as¬
pects of Tudor’s works: “Their shock value, thrilling at first, does not last;
their shaping force is discontinuous; they have a weak and fragmentary
dance impetus; they peter out at the end. They can find no repose and
no spring because balance is no element of structure in them” (ibid.).
Although Martin was more sympathetic to Tudor’s ballets than Denby,
Martin also found them disturbing when he felt they moved too far out¬
side the limitations of academic dance (Martin 1952: 71).
Denby’s analyses of the works of Tudor, de Mille, and Robbins did
not easily fit his description of the new American style that, he said, was
characterized by a dependence on movement alone to make its effects.
On the other hand, his descriptions of Balanchine’s style emphasized just
that element: “Not the story, but the dance rhythm, the surprising dance
figures, the witty solutions, the clarity, and above all the musicality of
the action seem to carry the piece” ([1947] 1986: 512). And speaking of
Balanchine’s classicism, he noted: “He has made our dancers look nat¬
ural in classicism. His pieces carry onstage when they dance them clas¬
sically clear and large, without nervousness or self-conscious glamour;
78 A Game for Dancers

he has shown how fascinating their buoyant rhythm can be in all sorts
of variations of forthright impetus” (ibid.: 511). Balanchine’s neoclas¬
sical style, which Denby stressed here as definitively American, closely
resembled what Martin had called for in an authentic ballet and that
Martin himself came to see in Balanchine by the end of the 1940s.
The case Denby made against modern dance was similar to the one
he had made against ballet that did not conform to a classical model. Of
Graham, he wrote:

I cannot omit mentioning Martha Graham, the greatest dance celebrity


in the United States. Now past fifty, she is an actress of magnificent power,
a dancer of astonishing skill; her choreographies abound in extraordinary
plastic images of great originality. They are expressionist in rhetoric,
violent, distorted, oppressive, and obscure; there is rarely a perceptible
rhythmic unit or any dance architecture. But the ardor of her imagina¬
tion, the scope of her conceptions, the intensity of her presence make her
a dance artist of the first rank. ([1948] 1986: 525-526)

As with Tudor, Denby faulted Graham’s choreography for its lack of


dance qualities. It had, he said, no rhythmic structure, nor dance archi¬
tecture. Instead, it was built on plastic images (an element of painting)
and on Graham’s personality and acting skills (elements of literature).
Graham’s dance, like Tudor’s, owed what power it had to pantomime
and personality, rather than movement. It was only in Balanchine’s
neoclassicism that dance became truly independent.

John Cage’s critique of modern dance was not based on the genre’s
fatal attraction to literature but on its lack of dance structure and an im¬
possible desire to communicate. Cage (1912-1992) was one of the best
known and most influential American composers of the twentieth cen¬
tury. The son of an inventor, he was raised primarily in Los Angeles,
where he studied with Arnold Schdnberg (Cage 1973; Revill 1992).
Cage became involved with dance in 1938, when he worked at the Cor¬
nish School in Seattle, accompanying and composing for the classes of
former Graham dancer Bonnie Bird. There he met Merce Cunning¬
ham and flve years later, when they had both settled in New York, their
lives and work became connected in an association that lasted until
Cage’s death.
Modernist Theory: Martin, Denby, and Cage 79

John Cage with prepared piano.


Courtesy of the Cunningham Dance
Foundation.

Among the occasional articles on dance that Cage wrote was a piece
from 1944 entitled “Grace and Clarity.” Published in Dance Observer, it
called for new thinking in modern dance (1944: 108-109). Like Denby
and Lincoln Kirstein, Cage complained that modern dance had been
too dependent on personalities. The first-generation modern dancers
had developed practices based on their own interests, dispositions, and
bodies. These practices were treated as sacrosanct; they could only be
altered by the dancer who had developed them. Then, Cage said, the
first-generation dancers had gradually deserted their own teachings,
often adopting ballet elements, particularly ballet vocabulary, which had
caused confusion among their students. The next generation of dancers
had either copied their elders and remained shadows of them, or had,
themselves, sought to incorporate into their own dance aspects of other
established idioms such as folk and Asian dance and ballet, or had ca¬
pitulated to Broadway and the commercial theater. Cage, then, echoed
much of the criticism that modern dancers were leveling at themselves
during the 1940s.
Cage had an answer, but it wasn’t one that most modern dancers
were considering. Modern dance. Cage contended, needed to establish
8o A Game for Dancers

stable, impersonal art practices. He cited ballet, singling out Balanchine’s


“exceptional” Danses Concertantes, as a form that achieved this objective
and that flourished despite being “devoid of interesting personalities
and certainly without the contribution of any individual’s message or
attimde toward life” (1944: 108). It was not ballet’s vocabulary, style,
patterns, or set and costume designs that was the secret. Cage declared,
but its clear rhythmic structure. What made it completely legible to a
viewer was the way in which the length of time the dance took was di¬
vided into large parts and then phrases. Whether one was speaking of a
dance, a poem, or a piece of music. Cage said, this was a time art’s life
strucmre.® The pleasure in watching dance, as in any of the time arts,
was to see the interplay between rhythmic structure and what Cage
called “grace,” “the play with and against the clarity of the rhythmic
structure.” Modern dance, with rare exceptions, had no understanding
of this concept. In order for it to become a mature art form and to serve
a useful social purpose it was necessary for it to “get itself a theory, the
common, universal one about what is beautiful in time art” (1944: 109).
Cage thus advocated the use of more impersonal methods in modern
dance, not through vocabulary, which is where many dancers were fo¬
cused, but through the rhythmic structure of dances.
In his 1944 article. Cage acknowledged that modern dance should
have contemporary relevance, although he did not specify how. It was
not, though, to be through purposeful communication, which he had
abandoned in his own music by the mid-1940s. Cage came to believe
that art could not communicate artistic intent. “I could not,” he said,
“accept the academic idea that the purpose of music was communica¬
tion” (Revill 1992: 88). Citing his own experience as evidence, he went
on to say that whenever he tried to communicate particular emotions
through music, he failed. In addition, he felt that “all artists must be
speaking a different language, and thus speaking only for themselves.
The whole musical situation struck me more and more as a Tower of
Babel” (Tomkins 1976: 97).
However, if dance was not to communicate, what was it to do? In an
article taken from notes for a talk before a joint concert with Cunning¬
ham in 1956, Cage expanded on his ideas: “We are not, in these dances
and music, saying something. We are simple-minded enough to think
that if we were saying something we would use words. We are rather
doing something. The meaning of what we do is determined by each
one who sees and hears it” (1957; 10). Cage went on to point out that
Modernist Theory: Martin, Denby, and Cage 81

there were no symbols in the dance, no stories, no psychological prob¬


lems. “There is simply an activity of movement, sound and light.” This
did not mean that the dance was not expressive. “The activity of move¬
ment, sound, and light, we believe, is expressive, but what it expresses
is determined by each one of you” (ibid.). Unlike Martin, Cage did not
believe the dancer could purposefully transmit emotional experience to
the spectator. What, then, was the function of art? Cage found it in the
spiritual. Quoting the teacher of an Indian friend, he wrote, the purpose
of music is to “sober and quiet the mind, thus rendering it susceptible
to divine influences” (Tomkins 1976: 99; Cage 1973: 158).
Part of Cage’s means of quieting the mind was to make his work
more impersonal, ridding it of what he called likes and dislikes. Cage
employed a number of strategies for eliminating preferences and judg¬
ments in his music, the best known being chance operations. Beginning
in 1951, he used the I Ching (Book of Changes) to introduce unpredict¬
ability into his work (Cage 1973: 57-61, 67-76; Tomkins 1976: 111-112;
Revill 1992: 128-139). This ancient Chinese book had been employed
to obtain oracular knowledge; Cage used its procedures, governed by
tossing coins, to take decision making out of his hands. The dada and
surrealist artists had also employed chance procedures to escape means-
ends rationality (Richter 1965; Nadeau 1989; Burger 1984: 64-66), and
Cage, with his many surrealist contacts (he had camped out at the home
of Peggy Guggenheim and Max Ernst upon arriving in New York) may
well have discovered the notion through surrealism. Cage also had by
this time discovered Zen Buddhism, which greatly influenced his the¬
ory and which fit well with anticausal ideas behind the use of chance
procedures. The specific impact of Zen will be discussed in greater de¬
tail in chapter 7 within the context of Cunningham’s choreography.
In addition to eliminating intentional communication and prefer¬
ences in dance. Cage, and through him, Cunningham, also changed the
relationship between dance and music, making music separate but equal
to the dance. The two existed in a cooperative arrangement, but inde¬
pendent of each other. In a sense. Cage’s idea of a modern dance was a
more extreme version of the dance that Denby and Martin had outlined
for an autonomous ballet. That is, it favored movement that excluded an
attempt to communicate meaning. However, it was more self-sufficient
than an abstract ballet, which remained dependent on music. Modern
dancers had long attempted to sever ties to music, usually by settling on
a score after having choreographed a work. This strategy, though, made
82 A Game for Dancers

music subservient to the dance. Cage, being a musician, sought a more


equal, but nonetheless independent relationship between the two. Al¬
though separating music and dance was an important difference be¬
tween modern dance and ballet, the crucial difference between Cage’s
conception of a new modern dance and modern dance as it heretofore
had existed was that Cage sought the genre’s basis in rhythmic structure
rather than in movement that emerged out of feeling, and in doing so
he pushed communication off the stage.

Yet as different as Cage was from Martin, and even from Denby, all
three men shared views and assumptions that kept them firmly within
the bounds of modernism. To begin with, they all attempted to shear
away from dance those elements that were dependent on other art
forms, particularly literature. A dance became authentic when it made
movement its central focus. The difference among them was their no¬
tion of when extraneous elements intruded. For Denby and Cage, mod¬
ern dance’s attempt to communicate produced a dance dependent on
acting and pantomime, while for Martin such elements appeared only
when a dance dealt with characterization and plot. Where Martin saw
embodiment, Denby and Cage saw representation.
In addition to an insistence on artistic authenticity through auton¬
omy, all three theorists took a position against rationalization. At its
most obvious this showed in a common dedication to anticommercial¬
ism. Martin did not wish to see modern dance enlarge its audience at
the expense of compromising its vanguard principles; he deplored the
use of elements such as balletic virtuosity and what he called “theatrical
effects” as part of his critique of postwar modern dance. Martin also ap¬
provingly ascribed an anticommercial stance to Balanchine and Kirstein,
and Denby likewise applauded what he considered Balanchine’s lack of
interest in the market. Cage and Cunningham’s work was so arcane that
there was no question of their being interested in commercial aims. Au¬
diences walked out of their concerts for years without any noticeable
change in their approach.
For Martin, an authentic dance defied a culture that was increasingly
defined by industry and technology. Art’s inability to represent nature
with the perfection of the machine made it clear that the artist’s strength
lay in the power to abstract from nature. Armed with this advantage, the
artist was able to give experience more value than nature itself ([1939]
Modernist Theory: Martin, Denby, and Cage 83

1965: 12 2-12 3). It was in this unique human ability to find and ex¬
plore the roots of experience and to expose essential truths that the
artist defied the power of a culture ever more oppressed by means-ends
calculation.
For Denby, classicism embodied the civilizing virtues of grace, cour¬
tesy and harmony, especially needed as a weapon against tlie violence
and destruction of the modern age (Denby, [1947] 1986: 507-508; [1948]
1986: 518; [1953] 1986: 433-440). Denby compared Balanchine’s at¬
tention to stylistic details to speaking a language with “purity of vocab¬
ulary and cleanness of accent, qualities that belong to good manners
and handsome behavior” ([1953I 1986; 435). Denby used these terms in
what Raymond Williams calls their traditional sense, as standing against
barbarism (1983: 57-60). Civility was part of art’s arsenal against the en¬
croachments of the brutality of twentieth-century war, made possible
by the rationalized processes of modern technology. For Denby, free¬
dom from market forces also came down on the side of civilization.
Commenting on Ballet Society in 1947 he wrote, “At the moment it
[lack of commercial interestl is the most effective way of keeping it
[ballet] civilized, though the method is obviously a lot of trouble. But
unless it stays civilized, ballet is no fun. Staying civilized is always every¬
body’s trouble, so why not ballet’s?” ([1947] 1986: 516-17).
Cage broke up means-ends rationality through chance procedures,
and through the 1950s he developed additional methods to guarantee
unforeseen results in his compositions and performances. He mentions
using collage, and he also experimented with other means employed by
the dada and surrealist artists, including the cadavre exquis, in which
a group of artists sequentially made part of a composition with only
the knowledge of a small fraction of the preceding artist’s effort (Revill
1992: 103). Such processes disrupted rational patterns of control that
Cage felt prevented sounds or movements from becoming themselves.
He noted that the devices he used were familiar from modern art and
architecture, and he said he shared the philosophic principles of these
modernists.
Perhaps the most important aspect of the three theorists’ anti¬
rationalization was that they all had faith in the ability of the dancing
body to mitigate the power of means-ends logic. They did this by claim¬
ing it was possible to “understand” dance, no matter how unfamiliar it
might be, if only the viewer disconnected intellectual analysis. Dance was
based on human movement, which everyone performed and understood
84 A Game for Dancers

intuitively no matter how transformed that movement might he. For


Martin, dance was assimilated through such physiological mechanisms
as muscular sympathy and itmer mimicry. When dance was not absorbed,
it was because spectators brought to it learned mental obstructions, in¬
cluding theories aimed at mystification, that inhibited their motor
responses ([1939] 1965: 54-55). Although Martin did not separate psy¬
chic from physical processes, he made a sharp distinction between con¬
scious and unconscious ones. For him responding to dance was quite a
different process from intellectual contemplation; “It is useless to ap¬
proach any work of art with the notion that it must be understood be¬
fore it can be responded to. Understanding is a process of rationaliza¬
tion after the experience; first there must be the experience or there is
nothing to rationalize about” (ibid.: 51). For Martin, the process of
communication from dancer to viewer was a transference of experience
that was different from any rational analysis that necessarily took place
after the fact.
Denby and Cage also viewed dance reception as being outside the
grasp of intellectual mediation. Denby argued that ballet, like poetry,
was first responded to in a “spontaneously sensual” way, and he advised
viewers not to think about what they were seeing but to simply enjoy it
([1949a] 1986: 530). Like Martin, he believed that intellectual endeavor
ruined a viewer’s ability to be receptive to dance. This is not to say that
ballet could not be analyzed; rather, it was that its expressive qualities
had little to do with rational thinking. Denby wrote: “Anyone who
cannot bear to contemplate human behavior except from a rationalistic
point of view had better not try to “understand” the exhilarating ex¬
citement of ballet; its finest images of our fate are no easier to face than
those of poetry itself, though they are no less beautiful” (ibid.: 531).
Denby also alluded to an idea of bodily intelligence, which he called
dancers’ intelligence. Pierre Bourdieu has described bodily intelligence
as a kind of understanding that occurs “only with our bodies, outside
conscious awareness, without being able to put our understanding into
words” (1990a: 166). According to Denby, dancers make the dance ex¬
pressive through this kind of intelligence. He remarked that, “it is an
error to suppose that dance intelligence is the same as other sorts of
intelligence which involve, on the contrary, words only and no physical
movement whatever” ([1944a] 1986: 204). Dance intelligence allowed
the dancer to heighten the viewer’s perception of her in line or silhou¬
ette and in mass:
Modernist Theory: Martin, Denby, and Cage 85

A dancer can emphasize a passage in the dance by emphasizing the shape


her body takes in the air. When she does this she does not call attention
merely to the limb that moves, she defines her presence all around in
every direction. At such moments she looks large, important, like a fig¬
ure of imagination, like an ideal human being moving through the air at
will. . . .
These are some of the physical characteristics of dance expression,
and the brilliant use of them to arouse our interest, to thrill and to satisfy
us, is proof of an artist’s exceptional dance intelligence, (ibid.: 204-205)

For Denby, the body had agency. It could, as Susan Foster has argued,
theorize in the moment through practice rather than after action in
contemplation (Foster 1995: 15-16). Foster sees this agency coming
into play especially in the rehearsal process, but as Denby noted, it also
occurs on the stage in the act of dancing. For Denby, then, intelligent
and articulate bodily practices could produce an entire world of expres¬
sion that lay outside the control of conscious thought and language.
Cage, too, believed that dance could best be understood without the
interference of any sort of intellectual contemplation. Sounding much
like Denby, he commented in 1957, “At a recent performance of ours
at Cornell College in Iowa, a student turned to a teacher and said,
‘What does it mean?’ The teacher’s reply was, ‘Relax, there are no sym¬
bols here to confuse you. Enjoy yourself!’” (1957: 10). Cage credited
this ability to respond to dance to Alartin’s concepts of inner mimicry
and muscular sympathy, which he called “kinesthetic sympathy,” and
which took place between the body of the spectator and dancer (ibid.).
Audiences had only to look and allow themselves to respond to the
movement without any intellecmal blockage, and the dance would make
its effects. What Martin, Denby, and Cage were all saying in different
ways was that an authentic dance could only be absorbed, at least ini¬
tially, on a corporeal level between bodies and free from the tyranny of
causal logic. In this sense, to come to a corporeal “understanding” of
dance was to experience freedom.
Since rationalization was considered a social problem, dance’s ability
to deal with it would have been enough to give it a social function.
However, for all three men, dance was socially relevant in other ways, as
well. To begin with, Denby and Cage would have agreed with Martin
that dance could not be separated from life because it embodied move¬
ment, which was common to everyone. For Martin, the social function
86 A Game for Dancers

of an authentic dance was to challenge the status quo by demonstrating


truth as society changed through time ([1939] 1965: 15-16). An au¬
thentic dance aimed at closing the fissure between art and society that
a modern division of labor and market economy had opened. For Denby,
dance could offer society practices for a civilized and peaceful world
after two brutal wars, while for Cage dance provided a means of bringing
a healing peace to the troubled human mind. In all, then, despite their
differences, Martin, Denby, and Cage shared a theoretical view based in
modernist tenets and assumptions. They also attempted to solve one of
the central problems of modernity, finding in the dancing body a means
of escaping the stranglehold of rationalization. But for all three, dance
was more than simply an escape; in its truth it held the possibility of a
better world, however utopian such an idea might have been.
EMBODYING
COMMUNITY

In The End of Ideology, Daniel Bell wrote of a “rough consensus” that


dominated Western society. He argued that the ideological revolu¬
tions of the years preceding the mid-twentieth century had died due to
a munber of causes. These included, on the one hand, disillusionment
after the rise of fascism in Germany and Stalinism in Russia and, on the
other, modifications in capitalism that took social welfare into consid¬
eration. Now, he said there was general agreement on political and eco¬
nomic issues: “the acceptance of a Welfare State; the desirability of
decentralized power, a system of mixed economy and of political plu¬
ralism” (Bell [i960] 1988: 402-403). The consensus that Bell invoked
with such complacency was not rmusual during the Cold War. As his¬
torian Alan Brinkley has noted, many observers saw American society
as one in which divisions were disappearing, a society that “reflected an
essential unity of interests and values widely shared by Americans of all
classes, regions, races, and creeds” (Brinkley 2001: 62). Although Cold
War rhetoric had much to do with such thinking, another factor was the
astonishing growth of the American economy during the 1950s, which
led to the conclusion that the United States could grow its way out of
any internal tensions (Chafe 1999; O’Neill 1986; Diggins 1988; Pells
1989; Brinkley 2001). Yet the “illusion of unity,” to use Brinkley’s term,
hid a good deal of dissent, much of it from disaffected minorities as dis¬
parate as artists and intellectuals, women, African Americans, and the
young. What made the illusion successful was that dissent tended to be
hidden or disguised. Bell maintained that most dissent was social rather
than political. Critique, he said, was aimed at such areas as mass culture
or an increasing sense of anxiety and alienation in American society. He
was right to a degree, but he failed to understand the necessary political
aspect of social issues. Bell also ignored the fact that during the Cold War
88 A Game for Dancers

years it wasn’t always safe to overtly criticize political institutions and


policies. If dance was any measure, social critique had political implica¬
tions that Bell and others of his generation did not see or did not wish
to acknowledge.
Although postwar high-art dance included its share of consensus
thinking, there were choreographers who engaged with social concerns
that veiled political issues. In some instances these choreographers fo¬
cused on ideas relating to community. Examples that come to mind in¬
clude the numerous works depicting the Salem witch hunts, which pre¬
sented communities rent by betrayal. Anna Sokolow, whose work will
figure prominendy in this chapter, showed how anxiety and alienation
characterized a society in which collective life no longer held.
Concepts of community formulated in the Progressive era of the late
nineteenth and early twentieth centuries posited difference as a given,
with consensus being hammered out through argument and compromise
(Wiebe 1990; Chambers 1992; Cooper 1992). Consensus ideology re¬
placed this idea with an aggregate of supposed commonly held attitudes
and beliefs. Instead of accepting difference as part of a collective whole,
consensus thinking advocated accommodation to the group through
such means as “togetherness” and “belonging” (Leuchtenburg 1973:
73-74; May 1988). Regulation was imposed through rules of normalcy
that suppressed difference. Consensus thought admitted few racial, class,
or ethnic differences to disturb the flow of postwar American life (Skol-
nick 1991: 68-69; Lears 1989; Pells 1989: 116-261). Furthermore, as
we will see, the rules of modernism could be invoked to suppress dif¬
ference through modernism’s demand for rmiversal meaning. Among
those who sought to account for difference while struggling to obey the
rules of the game they played were African Americans (whose work will
be discussed in the next two chapters) and dancers who interested
themselves in Jewish ethnicity.
In the wake of the holocaust and the emergence of the state of Israel,
a number of Jewish dancers mrned their attention to their ethnic roots.
Others, who had long dealt with Jewish themes, found their work given
new impetus by current events. Among the dancers who treated Jewish
subjects were Hadassah, Pearl Lang, Sophie Maslow, Fred Berk, and
Anna Sokolow. Naomi Jackson, in her study ofjewish dance at the 92nd
Street Y, argues that Jewish modern dancers who treated Jewish subject
matter were not trying to depict realistic representations ofjewish life
but rather sought to embody an essential Jewish identity (Jackson 2000).
Embodying Community 89

Yet one has to ask how possible it was within the bounds of modernism
to capture those sought-after essences, given that modernism demanded
“universal” essences that were inevitably defined from the point of view
of a dominant social order. One of the questions to be discussed here is
how an idea of Jewish identity, as distinct from any other, could be fit
into a modernist framework. Sophie Maslow’s The Village 1 Knew (1950)
and Anna Sokolow’s Kaddish (1945) are cases in point.
Maslow was born in 1911 on New York’s Lower East Side, the daugh¬
ter of Jewish immigrant parents from Russia. She took her first dance
classes at a socialist school, followed by studies at the Neighborhood
Playhouse with Graham and Horst. She danced with Martha Graham’s
company for twelve years beginning in 1931. At the same time, she was
involved in protest dance, often in her capacity as a member of tlie New
Dance Group. The group was the only one of the revolutionary dance
associations to survive through the postwar years. It had been founded
in 1932 as a means of educating dancers in order to participate in the
class struggle. After the war the group continued to provide a variety of
dance classes to amateurs and professionals at nominal fees, and it sup¬
ported a performing branch (Korff 1993; Graff 1997). In keeping with
its goals, the New Dance Group company specialized in socially con¬
scious works. Maslow taught classes at the school and performed and
choreographed for the company. She was also part of a trio that in¬
cluded Jane Dudley and William Bales. The trio, started in 1942, made
its debut under the aegis of Dance Observer, which was trying to help keep
modern dance afloat during the lean war years. After the war, Maslow,
Dudley, and Bales continued to appear together under the banner of
Jane Dudley, Sophie Maslow, William Bales and Company, drawing
on dancers from the New Dance Group for their supporting artists.
In the 1930s, Maslow had created such protest pieces as Two Songs
about Lenin (1935) and Ragged, Hungry Blues (1937). She had also been
interested in folk idioms, including Russian folk dance and themes. By
the end of the 1930s, she became concerned with the plight of itinerant
American farmers, which she treated in Dust Bowl Ballads (1941) and
Folksay (1942), both set to music by Woody Guthrie. The Village I Knew
was related to Maslow’s earlier work in that it dealt with folk material,
now focused on her Jewish roots. It was the most ambitious piece Maslow
created in which she tried to come to grips with the issue of Jewish
communal identity.* Based on Yiddish stories by Sholem Aleichem, it
was danced to traditional songs and music set by Gregory Tucker and
90 A Game for Dancers

Samuel Matlowsky. The work was premiered on i8 August 1950 at the


American Dance Festival in New London, Connecticut. I want to de¬
scribe The Village I Knew in some detail in order to make clear its vari¬
ous dance elements, as well as the degree of mimetic gesture found in
it compared with less literal movement.
Most of the seven sections deal with the small pleasures and dramas
of shtetl life in nineteenth-century Eastern Europe. However, the last
dance, “Exodus,” radically changes the mood of the work by depict¬
ing a pogrom. The first dance, “Sabbath,” has men on one side of the
stage, women on the other, each going about Sabbath prayers. The
women form a circle, then break into two groups, three women hold¬
ing candles, the other three praying, alternately holding their hands out
to the candles and covering their eyes in the traditional Sabbath bless¬
ing. Meanwhile, the men place prayer shawls over their heads and
shoulders, swaying in what is also part of orthodox prayer. The praying
gestures and positions are integrated into repeated dance patterns that
serve to emphasize the ritual aspect of their movement. The women
and men remain separate on each side of the stage, as they woiild in an
orthodox setting. The women dance in closely knit circles and parallel
lines, while the men spread out more, their movement and gestures
larger and freer as they arc their arms up and out in prayer, their bod¬
ies boldly arching as they sway.
“It’s Good to Be an Orphan” is essentially a pantomime. A girl skips,
stretches, and kicks her heels in happiness as she inspects two apples she
has apparently stolen. An older woman enters, scolds her, and extracts
one of the apples from the girl’s pocket. The woman then softens,
shrugs, and hugs the girl. Einally she takes off her boots and gives them
to the child. The next dance, “A Point of Doctrine,” is also a panto¬
mime vignette in which a housewife drives a rabbi to distraction with
her incessant talking and complaints.
“Eestival” consists of a group dance that is less mimetic and plot-
driven than the previous sequences. It was clearly inspired by folk dances,
especially evident in the line that snakes through the scene from time to
time like a leitmotif. However, the movement also includes numerous
modern dance steps, such as sweeping extensions and floor work. And
although folk elements like flexed feet and stamping runs are used, the
dancers’ comportment tends to be upright and straight in the modern
dance style rather than less formally held-up, as it is in many folk forms.
“The Eiddler” section again includes a great deal of mime to convey
Embodying Community 91

Jane Dudley, Sophie Maslow, and Ronne Aul in “The Fiddler,” from Maslow’s The Village
I Kneiv (1950). Photo: Walter E. Owen. Courtesy of Jerome Robbins Dance Division, The New York Public
Library for the Performing Arts, Astor, Lenox and Tilden Foundations.

a mother’s argument with her daughter who wishes to marry a poor


violinist rather than, presumably, someone with more financial security.
“Why Is It Thus?” follows. It is a dance for three men, who are called
students in the program, and who appear to be meditating on a Talmu¬
dic question. A bench is used as a prop for sitting and reclining as the
men take up various positions of meditation and prayer. As in the first
scene, there are the sways and sweeping gestures of the arms outstretched
in prayer.
Finally, in “Exodus,” Maslow shifted the mood of the work from one
92 A Game for Dancers

of affectionate humor to fear and panic. Chaos reigns as the characters


we have just seen in previous sections rush across the stage waving their
arms in terror, embracing each other, and running off. Women in stiff,
angular poses of pain are carried off amid the melee. Meanwhile, a group
of dancers acts as a chorus, commenting on the action through their
movement. They cross in a line at the back of the stage, their bodies bent
as if bearing a heavy load. The work ends with this scene of destruction.
As Naomi Jackson notes, Maslow did not attempt to replicate Jew¬
ish religious practices or shtetl life in The Village I Knew, rather she
reimagined them through mimetic gesture and dance. Her view of the
past, a past she had never experienced, was painted in primitivist forms
and colors, like the illustrations in a children’s book. The movement
throughout the work is made up of simple ballet steps like pas de basques
and pas de bourrees along widi modern dance movement, including the
Graham contraction (the latter particularly present in “Exodus”). These
are combined with folk elements and that amalgam, in turn, is attached
to pantomime that tells the individual stories of each vignette.
The Village I Knew became one of Maslow’s most successful works.
John Martin called it “warm and tender and funny and distinctly a credit
to Miss Maslow” (1950: 8). Walter Terry commented, “Its glimpses
(sharply focused and knowingly selective) of Jewish community life in
czarist Russia are amusing, touching, colorful and intensely human”
(1951a: 7). The only complaint concerned the last scene, which nearly
every critic, including Martin and Terry, commented upon. Nik Kre-
vitsky’s objection was typical. He wrote in Dance Observer:

Only one point, which we have made previously, and which still applies,
seems to detract from the perfection which it is possible for this dance to
achieve. That is its sad, though realistic, ending. The Exodus, though rel¬
evant to the type of community which this work portrays, is alien in spirit
to the attitude of life which Aleichem emphasizes; the notion of hope is
gone and with it much of the zestful dancing that precedes this sad clos¬
ing. (1951c: 43)

Krevitsky was correct in saying that Maslow’s conclusion was not in


the spirit of Aleichem, but from a modernist standpoint that was the
least of the work’s problems. The modernist issue lay in Maslow’s speci¬
ficity of time, place, and character and in her extensive use of panto¬
mime, which constituted, in modernist terms, a representation of Jew-
Embodying Community 93

ish identity rather than its embodiment. Yet none of the critics objected
to the work on those grounds. What they did object to was the intru¬
sion of realistic tragedy into the dance, implying they viewed the rest of
the piece as fantasy. It would appear that as long as Maslow’s vision of
Jewish identity lay in the realm of the picturesque, of quaint villages and
amusing peasants, it wasn’t necessary to come to grips with issues of
abstraction and essential truth. The Village I Knew could be specific be¬
cause it was not really modern, and commentators could therefore
excuse Maslow’s work from the rigors of an authentic dance. Massacres
were something else, though. “Exodus” was a jarring reminder of life
experience, which modern dance was supposed to essentialize. It im¬
posed harsh reality on fantasy, and it remained alarmingly specific. As
such, it had no place in the exotic folktale world Maslow had depicted.
Another element of The Village I Knew that obliquely touched on its
modernist problem had to do with Maslow’s inclusion of African Amer¬
icans in the cast. Ronne Aul and Donald McKayle were members of the
New Dance Group, which Maslow drew on for dancers.^ According
to McKayle, John Martin did not understand why there were, in Mc-
Kayle’s words, “black boys” in a Russian Jewish village (McKayle 1993;
1966; 73).^ McKayle said he wrote to Martin advising him to ask him¬
self why the fact that there were no Russian Jews in the ballet didn’t
bother him. Martin replied that by including black dancers the work
lacked theatrical verisimilitude and that it was as incomprehensible as
Shirley Temple playing Hamlet or John Barrymore playing Juliet. It is
notable that Martin’s comparisons were to drama, not to dance, and his
belief that characters who were supposed to be of a specific race or gen¬
der could not cross those boundaries. As Martin made clear in his own
theory, drama was far more literal than dance, and the more drama a
dance included, the less it was concerned with essential life experi¬
ence. Martin’s comments, leaving aside the issue of racism, indirectly
addressed the problem of modernism in The Village I Knew; that is, the
work’s specificity of time, place, and character kept it from coming to
grips with abstraction. Maslow’s effort, although a dance work that
pleased many, was unable to bridge the gulf between communities that
sought identity outside the dominant order and the rules of modernism.
In her solo, Kaddish, Anna Sokolow tried a more strictly modernist
approach to Jewish identity than did Maslow in The Village 1 Knew. How¬
ever, this only served to demonstrate the limits of modernism in other
ways. Sokolow (1912-2000), like Maslow, had been involved in social
94 A Game for Dancers

protest dance in the 1930s. Also like Maslow, she was the daughter of
Jewish immigrants from Eastern Europe who became interested in her
own Jewish roots. In 1939, she choreographed The Exile, in which she
contrasted an idyllic Jewish culture before the war (“I had a garden...”)
with the horror of Nazism (“The Beast is in the garden ...”). The solo
was set to a poem by Sol Eunaroff and had music by Alex North. In 1943,
she created Revelations made up of The Exile and dances devoted to the
biblical women Ruth, Naomi, Miriam, and Deborah. As was her habit,
she combined older pieces with new ones, sometimes redding them.
Songs of a Semite of 1943 was essentially with the addidon of
a dance entitled “March of the Semite Women.” Sokolow premiered
Kaddish in Mexico in 1945, then performed it for the first dme in New
York on 12 May 1946 as part of a solo concert at the 92 nd Street Y. In
that concert she also offered two other works with Jewish themes: Im¬
ages from the Old Testament (which included the “Miriam” solo) and The
Bride.
In the mid-1940s, Sokolow was working primarily in Mexico and so
performed infrequendy in New York. In general, her dances of these
years were not considered to be her strongest, but The Exile and Kad¬
dish were exceptions. Although The Exile has disappeared, Kaddish has
survived and exists in a video danced by Deborah Zall. I want to com¬
pare it here to Graham’s solo Lamentation (1930), long accepted as a
model of modern dance.^ Kaddish, listed in the program as a “prayer for
the dead,” was inspired by a Jewish mourning prayer. It is set to Ravel’s
meditation on the Kaddish. Eor the dance, Sokolow wound a tefillin, an
Orthodox prayer box, around her head and arm, something only men
do in the Orthodox faith and which they use for daily prayers, not for the
Kaddish. Deborah Zall eliminated the prayer box in her reconstruction.
The solo begins with the dancer standing erect on a darkened stage,
her hands folded into fists at her breast. It is this tight, narrow vertical
that is the central position in the dance, the one repeatedly returned to.
The dancer slowly unfolds her hands, palms up, and looks up as if ques¬
tioning heaven. She moves forward haltingly, her movement tense and
bound as if constrained by pain. She places her fist against her oppos¬
ing shoulder, bends forward, her hand to her head, and stands swaying.
She kneels and does a back bend, then falls forward. Gradually her move¬
ment enlarges. She swings her arms outward as she turns from side to
side, then runs with her arms outstretched. She plunges to the ground,
rises, plunges again. As she rises a second time, her movement once more
Embodying Community 95

Anna Sokolow in Kaddish (1945). Photo courtesy of the Sokolov) Dance Foundation.

becomes constrained. The dance ends as she haltingly steps backward,


lightly tapping her breast.
If one compares this dance to Graham’s solo of grief, the differences
and similarities are instructive. Graham’s celebrated solo is done pri¬
marily while seated on a bench, and she is wrapped in a flexible jersey
tube. Her movement is literally contained, consisting mostly of sways
forward and back and from side to side accompanied by arm gestures
that sweep to the side or plunge downward between her legs, the latter
open in a wide second position. At one moment she touches each side
of her face in a ritualized wiping of a tear.
Martha Graham in Lamentation (1930). Soichi Sunami, photographer. Courtesy of the Snnami estate.
Embodying Community 97

Both Sokolow and Graham’s solos have a sense of containment, but


Sokolow’s movement breaks out of the vertical in periodic explosions
that carry her to the side or down onto the floor. Whereas Graham’s
gestures have no clear mimetic associations except the ritualized touch¬
ing of her face, Sokolow included gestures that could be associated with
Judaism but that are also part of a general lexicon of lamentation, like
the beating of the breast and the questioning gesture of the arms. The
tide of the dance and Sokolow’s use of the prayer box would have tied
her solo to Judaism, but the movement itself is, as a whole, generalized.
Yet despite their similarities, the two dances differ markedly in mood.
Graham’s containment is so complete it is tempting to call it puritan¬
ical while Sokolow’s is broken, the emotion more exposed and more
antique in its gestural associations, however abstracted they may be. In
this sense the dance’s references can be viewed as biblical and hence by
extension Jewish. Yet both dances are highly abstracted treatments of
the pain of grief, and the less specifically “Jewish” Sokolow’s dance was,
the more it spoke of a generalized pain and the less of something es¬
sentially Jewish. Kaddish therefore points up the problem of encom¬
passing difference within the rules of modernism. Created to confront
a dominant culture from within, modernism was not a flexible instru¬
ment for dealing with issues of difference. Jewish dancers attempted to
expand modernist boimdaries to include constituencies outside the dom¬
inant social order. In doing so they also challenged notions of consen¬
sus culture where difference was muted. In this instance, the rules of
modernism aligned with consensus ideology to emphasize unity, as illu¬
sionary as that idea may have been.

In addition to grappling with problems of how to reconcile modernism


to an idea of essentialized Jewish identity, Sokolow took on another
issue of modernism and community. Here she did not approach com¬
munity from outside the mainstream; rather she attacked it from within,
focusing on the dysfunction that occurs in a society where community
has come apart. Sokolow explored this theme in two ways: through the
concept of alienation and through rebellious youth.
Alienation had always been closely connected with modernity, but
the concept took on particular force in the 1950s. Writers of every
stripe considered it until, by the end of the decade, it had worked its
way into the general vocabulary of the country. Alienation was tied to
98 A Game for Dancers

notions of rationalized society in ways that I would like to consider by


comparing two nearly contemporaneous works: Erich Fromm’s socio¬
logical study, The Sane Society (1955), and Sokolow’s Rooms (1955)- This
is not to imply that Fromm and Sokolow influenced each other, but
rather that they treated alienation in often parallel ways.
Like many intellectuals in the United States during these years,
Fromm had begun his career as a Marxist and ended, by the 1950s, a
liberal in the consensus vein (Jay 1996). He had been a member of the
Frankfurt School in Germany, emigrating to the United States in 1934
and soon after breaking his ties with the Institute of Social Research. A
psychiatrist, Fromm had been particularly interested in merging Marx¬
ism with Freudian theories; then after the war he rejected Freud’s view
of society as too pessimistic. He is of interest here because he treated
the subject of alienation at length, and his ideas had much in common
with other intellecmal thought of the day.
Sokolow had long been associated with leftist causes. Like Maslow,
she danced with Martha Graham’s company during the 1930s. She also
managed her own group, the Dance Unit, which appeared at union
halls and working-class social organizations. During the 1930s and ’40s
Sokolow lived with Alex North, a composer and leftist whose brother,
Joseph, was a founder of New Masses. Sokolow spent three months in
Russia with North in 1934. Although North considered Sokolow naive
in her political views and she was apparently never a member of the
Communist Party, she retained her critical viewpoint of capitalist soci¬
ety throughout her life (Warren 1998: 29-30). When Sokolow returned
to New York from Moscow, she renewed her association with Graham,
as well as continuing her own choreographic career. During the 1940s,
as noted, she worked extensively in Mexico City, returning from time
to time to New York. The 1950s found her again working with her own
company in New York, as well as taking assignments abroad in Israel
and Mexico. She premiered Rooms at a concert at the 92nd Street Y on
24 February 1955.
Rooms is a suite of dances that depicts various types of men and
women who are out of touch with themselves and cannot make contact
with others.^ It has a jazz score by Kenyon Hopkins. Sokolow once re¬
marked that Rooms dealt with the loneliness of the city; however, as one
critic noted, it was not about loneliness but aloneness (Terry 1955: 12).
Fromm defined alienation as “self-estrangement” and related it to so¬
cial rationalization, or as he awkwardly called it “quantification and
Embodying Community 99

The Juilliard Dance Ensemble in the opening dance from a 1977 production of Anna
Sokolow’s Rooms (1955). Photo: © Peter Schaaf.

abstractification” (1955: 110-152). His explanation, largely based on


early Marxist theory, found the source of alienation in the development
of capitalism. As the division of labor took hold, exchange value took
precedence over use value, and people were separated from the making
of a whole product, which they had pre\dously controlled. They could
no longer see or understand the whole with the result that the power
they unleashed through their labor, primarily through science and tech¬
nology, came to rule over them. Men and women thus lost themselves
as the center of their own experience. In a quantified and abstracted so¬
ciety, people viewed themselves as salable commodities rather than as
active agents and bearers of power. Consequently, they measured suc¬
cess by what others thought of them rather than by their true identity.
(Fromm, like most of his contemporaries, assumed that each person had
an essential human identity that was authentic but corruptible.) In a so¬
ciety in which the division of labor was so complete, community was
separated from the individual, as was public from private life. Individu¬
als could no longer understand their own interest as part of a greater
community, so selfish interest prevailed. By the 1950s, Fromm said, so¬
ciety had become almost completely alienated.
lOO A Game for Dancers

It is in the notion of fragmentation and a resulting sense of loss,


particularly in terms of community, that Sokolow’s works most closely
resemble Fromm’s idea of alienation. Rooms begins with a section entitled
“Alone,” in which eight men and women are seated on chairs, all facing
front. As a plaintive trumpet solo begins, one, then another dancer, rises
and is seated again. Sokolow equated the chairs with cell-like rooms,
which each individual inhabits. Seated, the dancers alternate between
tense and twisted clutches and stretches and awkward flops that leave
them prostrate across their chairs. From time to time one of the dancers
looks up and then collapses again. Each individual moves indepen¬
dently of the others with no apparent relationship to the whole. This
initial dance introduces several solos, the first of which is entitled
“Dream.” Here a young man rolls away from his chair and across the
floor. He does slow cartwheels and back bends, lies on the floor, feeling
forward with his fingers, then grips and claws his way along before
struggling to rise. He takes large steps forward in slow motion, runs
in circles, lies on his back, then suddenly sits up. Finally he walks slowly
back to his chair, his head bowed. Many years after creating Rooms,
Sokolow described how she had thought about each of the solos. The
first, she said, characterized the actions of a man who “dreams contrary
to what he is in life” (Sokolow 1974). Fromm might have said the man
dreams because his real work is unsatisfying. He dreams of a life in
which he controls his actions and is a bearer of power instead of being
merely a small part of a whole he does not understand or control.
The next solo, “Escape,” is set for a woman. Her movement is some¬
times languid as she shakes out her long hair, sways to the jazz rhythms
and slowly stretches out her body. At other moments, however, her ac¬
tions become frantic; she rushes from one empty chair to another, she
dances madly, tossing her hair obsessively from side to side, then sud¬
denly flings herself down, all her energy drained away. At the end of
the dance we find her walking aimlessly, then she stands behind her
chair gazing off into space. She appears strangely narcissistic, resembling
Eromm’s individual who views herself from a distance as an object, a sal¬
able commodity. Certainly such objectification has long been the plight
of women, but in an age of alienation it becomes a general condition.
The other dances in Rooms further demonstrate the loss of self that
characterize alienation. “Going” is a solo for a male dancer who acts as
if he is on amphetamines. This dance is the only one in which Sokolow
used jazz movement. The young man snaps his fingers to the beat, his
Embodying Community lOI

body bent forward as he prances in tight steps, hep-cat fashion. He


switches suddenly to a manic kind of boxing, punching and dodging in¬
visible blows, then he hammers out the beat with silent, clapping hands.
He runs, he slides, he falls, he hammers the beat, he punches, he snaps
his fingers until finally, seated on the stage he slowly raises and lowers
his outstretched arms, his whole body straining upward as if attempting
to fly. The dance ends with the man seemingly nailed to the floor and
straining to escape.
The next dance, “Desire,” is for six men and women who remain
sealed in their own worlds as they crawl, curl, and stretch within the
precincts of their limited space. Their movement is full of both yearn¬
ing and sexual tension as they reach, then twist and hunch in on them¬
selves. The dance is dominated by a step in which the dancers rhythmi¬
cally rub their feet forward and back on the floor in a way that suggests
a masturbatory soothing of frustrations. They remain in their hermetic
worlds, close to each other but unseeing, mirroring each other’s move¬
ment but never connecting. “Panic,” another male solo, contains a recur¬
ring image: that of a man bent from the waist who reaches out blindly
in front of him. He staggers as he moves this way and that. He covers
his eyes and shakes his head as if in the grip of migraine. He rushes for¬
ward, retreats, and finally crouches behind his chair as the dance comes
to an end.
“Daydream,” is for three girls and perhaps was meant as a lyrical mo¬
ment in an otherwise wholly bleak work. The music is sweet and the
girls dance, supported by their chairs. Although they face the audience
they appear to be regarding themselves in a mirror. At the end they stand
behind their chairs, sling a leg over the back and lean forward, elbows
on knee, as if contemplating their faces. Since, as Fromm said, commu¬
nity is separated from the individual in an alienated society, individuals
no longer perceive themselves as part of a community and self-interest
takes over. The narcissism of the characters in Rooms, speaks of that as¬
pect of alienation. The penultimate dance is entitled “The End?” and is
a female solo, very percussive, that Sokolow associated with madness
(Sokolow 1974). Among the woman’s gestures is a tense wriggling of
the fingers of her outstretched arms. This solo is followed by a return
of all the dancers, who bring on their chairs and seat themselves. The
work ends with them sitting, staring blankly ahead.
In his discussion of alienation Fromm mentioned that early psychi¬
atrists were called alienists because they dealt with people who were
102 A Game for Dancers

both lost to themselves and unable to make contact with others. That
sense of loss permeates Rooms. The characters have little idea where they
are or who they are, as they alternately embrace and attempt to escape
their fragment of real estate. There is also about them an obsessiveness
and yet aimlessness that is neurotic and in some cases pathologic. Above
all, there is the dancers’ gaze, which is never allowed to connect the
performers with their fellows. This hermetic disconnect from others
through the device of the gaze is ironically what most binds the work
together. For Sokolow, this was the world that modernity had wrought.
As close as Sokolow and Fromm were in their views of alienation,
they parted company on the solution to the dismal condition that was
the lot of modern men and women. Fromm sought an answer not in
revolution or organized communal pressure but in individualism, the
courage to be different, to become in his words, “productive.” Fromm
looked to changes of life practice to bring about this productive indi¬
vidual who must learn how to handle continuing conflict, to be aware
of suffering, and to act on his or her own initiative. The sane society
would be one that attacked the problems of rationalization and atten¬
dant alienation from all standpoints—economic, political, spiritual, and
philosophical, as well as through character and culture (1955: 270-363;
see also 1947, 1950). Fromm’s optimistic turn and his emphasis on pro¬
duction brought him into conflict with some of his peers, notably Her¬
bert Marcuse, while linking him to consensus liberals like Seymour
Lipset, Daniel Bell, and Arthur Schlesinger Jr. (Marcuse 1955: 248-
274; see also Jay 1996; Pells 1989). Sokolow saw no such happy solution
to modern ills. For her, the world was as it was. In her pessimism, she
was closer to Adorno, Horkheimer, and Marcuse than to Fromm. For
Sokolow there did not seem to be a way to regain communal power or
for the individual to be at once herself and part of an integrated whole.
Yet if Sokolow was relentlessly pessimistic in her critique of modern
society, she nonetheless found effective ways to embody its pain. She
did this through the structure and form of her dance. She began by cre¬
ating the piece in suite form, which allowed her to dispense with any
sort of story line. Her theme made its effect through accumulation, one
segment added to another to reinforce a point, rather than through a
series of causal events. Sokolow was careful to avoid narrative vignettes,
staying within broad thematic boundaries.
Rooms also showed that it was possible to treat contemporary social
problems in a highly metaphorical manner. Although it seems contra-
Embodying Community 103

dictory, Sokolow may have developed her successful strategy through


her work with actors. Sokolow had taught movement to actors at the
Group Theatre in the 1930s. When Elia Kazan, Cheryl Crawford, and
Robert Lewis, who had been associated with the Group Theatre, started
Actors Studio in 1947, they asked her to teach movement classes for them
(Warren 1998: 89-94). Actors Studio specialized in a training method
that was loosely based on that of Constantin Stanislavski, which taught
actors to search deeply within their own experience to discover the feel¬
ings and actions for their characters. Sokolow became interested enough
in the workshop’s approach to participate in some of its acting classes.
She also worked with the actors on specific projects of her own.
Sokolow actually started to make Rooms at Actors Studio but was
disappointed in the results and decided instead to make a dance piece
(Warren 1998:115). However, she may have found the basis of the move¬
ment she wanted for Rooms by working with the actors to locate move¬
ment and gestures of people under mental and emotional stress. Sokolow
had studied with Louis Horst in the 1930s, and it appears she returned
to his idea of modern dance’s beginnings in vernacular movement. In
his hook Modem Dance Forms, Horst wrote: “Both [German and Ameri¬
can modern dance] built up the art form by abstracting familiar every¬
day movement. The realistic gesture or posture is taken as a point of
departure on which to construct a poetic metaphor” (Horst and Russell
[1961] 1987: 18). Sokolow returned to the roots of modern dance to
find powerful new metaphors for an alienated age.
Sokolow said she discovered her mature movement vocabulary while
choreographing Lyric Suite of 1954. Although certainly Lyric Suite con¬
tains the seeds of much of the vocabulary for Rooms, it also employs a
great deal of conventional modern dance movement. In contrast. Rooms
is a more radical work. It contains hardly a recognizable step from the
standard modern dance vocabulary that by 1950 had coalesced from
the techniques of the pioneers. Rather, Sokolow started fresh from non¬
dance movement, which she then heightened through exaggeration,
accumulation, and especially, repetition. She also made use of what is
called in Laban Movement Analysis, “bound flow,” in which bodily
energy is put in tension with itself to hinder the flow of movement.
She Aen alternated this with a total depletion of energy. The constant
wrenching of the body from one extreme energy level to another helped
to produce an effect of abnormality. Einally, there was the gaze, which
Sokolow began to develop in Ly?ic Suite. In that work, the dancers tend
The Juilliard Dance Ensemble in a 1977 production of Anna Sokolow’s Rooms (1955). Photo:
© Peter Schaaf.

to Stay focused into the distance rather than on each other, but it isn’t
consistent throughout. Particularly in the central duet, the man and
woman frequently make eye contact. In Rooms, however, Sokolow made
the separation complete.
Rooms was premiered on 24 February 1955 at the 92nd Street Y, and
critics had the chance to assess it at various performances over the
course of the next few months. The work was immediately recognized
as important, but also dismrbing and unpleasant. Several reviewers
called it a theater piece, saying that it was realistic. Doris Hering, how¬
ever, noted that, “though her [Sokolow’s] viewpoint was philosophical,
it did not fall into the trap of being literary.” What impressed Hering
“was the great complexity of dance material woven into what seemed
like the most simple of dance surfaces” (Hering 1955: 75). One would
tend to side with Hering in this assessment. For although the move¬
ment in Rooms is gestural, Sokolow’s treatment of it is anything but
realistic. Gestures are danced, that is, they are conceived as rhythmic
pattern rather than as mimetic elements that interrupt the dance flow.
If she sought the basis of her vocabulary in the non-dance movement
described by Horst, Sokolow was also prompted to use that vocabulary
Embodying Community 105

in dance terms, imbuing it, in Denby’s words, with dance qualities. This
is not to say that Rooms is objectivist. It is expressional; that is, it is for¬
mally abstracted from life experience to perform a social function. The
work may have done this a little too well for John Martin, who found it
objectionable on several points, particularly in its unrelentingly dismal
tone. But be also conceded that it was impressive. “The explanation,”
he said, “probably lies in the extreme concentration of movement, of its
music and especially of its performance” (1955a: 26).
Most critics viewed Rooms as psychological, as pertaining to indi¬
viduals who are cut off from society. Only George Beiswanger, in a
thoughtful piece v-Titten after seeing the work at the American Dance
Festival in 1956, commented on it in sociological terms (Beiswanger
1957: 21-23). To view Rooms as psychological was to keep it at a dis¬
tance, to consider it involving simply individual cases at odds with the
world. To view capitalist society as a whole in such bleak terms was far
more damning. If alienation were an inevitable result of capitalism,
therapy would be of little help. However, whether psychological or so¬
ciological, part of the reason Rooms impressed even if it did not please,
was surely due to the fact that it could pass for imiversal truth. Audi¬
ences perceived in its alienation an essential element of modernity. As
such. Rooms proved to be one of the rare instances in the 1950s when a
dance of devastating critique was acknowledged as successful, even if its
full implications were deflected.

In 1958, Sokolow began a project of related works that continued her


exploration of social critique and that occupied her for a decade. All
were set to jazz scores by Teo Macero, with whom Sokolow often worked
beginning in tbe 1950s, and all concerned disaffected youth, or so the
works were interpreted by critics at the time. Sokolow herself did not
speak of any particular age group in these dances, but perhaps because
she used contemporary social dance in them it was assumed the works
dealt with young people. In the 1950s, youth was also very much on the
minds of Americans; the rise of a postwar youth culture riveted public
attention.
Although rebellious youth is probably as old as human relationships,
the flappers were the first generation in the twentieth century to become
celebrated for flaunting society’s rules in often self-destructive behav¬
ior. Rebellion had little place in a country in deep economic depression
io6 A Game for Dancers

and then world war, but the 1950s ushered in an era of prosperity ac¬
companied by the baby boom, both of which accelerated consumerism.
This provided a climate for developing a specific culture of white
middle-class youth in conflict with adult society (Lipsitz 1981, 1989;
Gilbert 1986; Diggins 1988; Goodman i960). At the same time, politi¬
cians, law enforcement professionals, social scientists, and the media
became focused on what was perceived as an increase in juvenile crime.
This occurred even though the statistics on whether or not there was an
actual crime wave were far from conclusive (Gilbert 1986: 63-78). The
disparity between reality and perception appears to have been the result
of experts and the media finding it difficult to separate youth culture
from delinquency. The concern with juvenile crime resonated with par¬
ents who could barely recognize their children in the adolescents living
in their homes.
In the 1940s, mainstream America had fistened to the same pop music
whether the listener was sixteen or sixty. Singers like Frank Sinatra,
Perry Como, and Rosemary Clooney dominated the Top 40 charts. Ten
years later the situation was vastly different, as teenagers danced to the
aggressive sounds of rock music and imitated the provocative gyrations
of Elvis Presley. By the mid-1950s, teenagers not only listened to their
own music, danced their own dances, wore their own clothes and hair
styles, they seemed to be more influenced by their peers and the mass
media than by their parents. In addition, as increasing numbers of work¬
ing-class adolescents began attending high school after the war, middle-
class parents worried about influences from these unfamiliar, and possi¬
bly dangerous, sources. Many adults feared they could no longer impress
their own values on their children. This translated into the specter of
rampant juvenile crime, which was cast in the media, both by social sci¬
entists and journalists, as nothing less than a breakdown of society. The
focus on adolescent misbehavior was further fueled by congressional
hearings on juvenile delinquency throughout the 1950s and by an out¬
pouring of magazine and newspaper articles on both adolescent crime
and life styles. Films such as The Wild One (1953), Blackboard Jungle
(1955), and Rebel without a Cause (1955) fanned the flames of concern as
they romanticized adolescents acting-out to the horror of adults.
Sokolow had dealt with youthful rebellion long before the 1950s in
Case History No.-(1937). In this solo she treated juvenile delin¬
quency from the standpoint of how social inequality led to crime. Her
program note read: “A study of a majority of case histories shows that
Embodying Community 107

petty criminals usually emerge from a background which begins with


unemployment and follows its course from street corner to pool room,
from mischief to crime.Margaret Lloyd described the dance as “a
study in juvenile delinquency by the poverty route, showing a youth,
bored and restless in his grim environment, releasing stifled energies in
the petty offenses that lead to felony, and culminating in a tense crouch
of fear against the backdrop” (1949: 217). Sokolow’s new series of works
did not portray youth as a victim of society but rather as representative
of it. Here she was not dealing with juvenile delinquency among the
lower classes, as she had in Case History No.-, but instead with a
more generalized and ambiguous group that was seen by observers as
contemporary youth.
Sokolow also had used social dance in several earlier works, among
them a jitterbug duet in the Kurt Weill-Langston Hughes musical.
Street Scene, of 1947 (Lloyd 1949: 214, 217). Nor was Sokolow the only
choreographer to be interested in youth culture in the 1950s. The most
celebrated of them was Jerome Robbins, whose West Side Story (1957)
and N.Y. Export: Opus Jazz (1958) were far better known than Sokolow’s
works. Sokolow and Robbins were friends and respected each other’s
choreography, but Sokolow, typically, took the vanguard high road. Her
studies did not glamorize youthful bad behavior, but once again showed
the bleak alienation of capitalist society. To reinforce her vision Sokolow
eschewed the kind of quasi-commercial music by Leonard Bernstein
and Robert Prince that Robbins favored, choosing instead the complex,
often harsh sounds created by Macero.
Sokolow began her series with Session for Six, which premiered in
February 1958 in New York. She continued it with Session f8, which
had its first performance in April 1958 at Juilliard, where Sokolow was
teaching. That same year she presented Opus f8 in Amsterdam, Ses¬
sion for Eight at Juilliard, and Opus Jazz ipy8 in Israel. Then came Opus
'60 in Mexico, Opus ’62 in Israel, Opus ’6y at Juilliard, and finally Opus
’dy, which she made for the Joffrey Ballet. Speaking of Opus f8,
Sokolow said, “I used jazz for an over-all aura of the sounds and
rhythms of today. I wanted the feeling of a new era, one where life is vi¬
olent and precarious, and the individual seems unimportant” (Sokolow
1965: 39). Although the works in this series differed over the years, a
sense of violence, aggressiveness, and boredom endured.
Filmed versions exist of Session for Six, Opus ’hy, and Opus ’hy. From
these and accounts of some of the other works it is possible to assemble
io8 A Game for Dancers

at least a partial idea of how the series developed. Viewed in conjunc¬


tion with the later pieces, Session for Six looks related to, but nonethe¬
less quite different from, the later dances, which are closer to each other
in structure and movement. However, although Session for Six is not an
earlier version of the Opus works, it is nonetheless a clear precursor.
Session for Six, which premiered at the 92nd Street Y, begins with a
man and woman alone on stage, who do a combination of coupe jete steps,
leave, then quickly reemerge with the group.^ Teo Macero’s score re¬
sembles a baroque trumpet motif as the couple begin to move, then
quickly segues into jazz variations as the group of dancers replaces the
initial couple. It’s as if the music tells us what will happen. This first
dance is dominated by coupe jete, grand plie, and petit echappe, all seen in
continuously changing combinations of steps and dancers. While exe¬
cuting these movements, the dancers maintain clearly arranged patterns
of lines and groups so that the structure is always emphasized. Their
comportment for the most part is held-up and academic. Yet within all
the strict patterns and steps there are small ruptures—a handshake, al¬
ternating clenched and open hands, a hand to the head, or undulating
hips in 2Lplie. A man executes one twist—it’s a harbinger. The first dance
ends with the performers executing grands jetes, first crisscrossing the
stage, then in a circle. The hand-to-head gesture is increasingly seen
until it becomes a motif in its own right.
The second dance begins with the performers entering the stage as
if at random. Now instead of maintaining upright, academic comport¬
ment, they bend their torsos forward slightly and execute small swing
and jitterbug steps. They become couples, join hands as if doing a fast
arm-wresde that then turns into full body twists. They are doing twist¬
ing social dance steps, but the dancers still stay in the same structured
patterns as earlier. They change patterns in jazzy prances with their hands
held in front of them in the pawlike gesture borrowed from social and
jazz dance.
In the third and final dance, the three girls take center stage and do
grands battements and plies interspersed with jazzy steps. Halting in deep
plie, they look around for the boys, who enter and lift them, still in a plie
position into the air. They separate and all exit, except one boy and girl
who look intensely at each other for a split second before running into
the wings. Everyone returns, and the dance ends with the group doing
fast jump turns in fifth position, their arms outstretched, their bodies
erect and their heads bent forward, their gaze on the floor.
Embodying Community 109

Doris Hering wrote of Session for Six that it had a note of “bitter
hunger.” Yet despite its emotionality, the work “was etched in spare out¬
lines colored by a bold use of spatial and rhythmic counterpoint. Struc¬
turally it was the most disciplined work Miss Sokolow has created to
date” (1958: 27). Hering mentioned the emotionality of the piece, but
other critics were less clear about that aspect. Louis Horst, for example,
found it “bright, gay and perky and was definitely contemporary with
its suggestions of jazz.” He also noted “an explosive nuclear ending”
(1958a: 55), while P. W. Manchester at Dance News called it an abstract
work that was over before it had a chance to solidify into “what prom¬
ised to be an exciting exposition of male and female dance” (1958: 9).
Looking at Session for Six today, one can see how all these ideas might
be formd in it, depending on the viewpoint. It is certainly short, lasting
barely eight minutes, which may account for some of the tentativeness
Manchester felt. Sokolow also placed more emphasis on formal elements
than she usually did. Yet at the same time she included certain gestures
and movements that suggested an emotional texture. Hering, who found
the work bitter, may also have been responding to the dancers’ gaze,
which tended to be focused inward and which gave the performers the
look of cool hipsters. On the rare occasions when the dancers made eye
contact, it was like an exclamation point.
In Session for Six, Sokolow seems to have been trying to combine el¬
ements of different genres and styles to see what kinds of moods and
movement qualities she could achieve. Her goal was not simply to start
in academic dance and end in jazz, but to see how different combina¬
tions of movement could work together. Judging from the critical re¬
sponse, it was uncertain at that point what she had achieved. But Ses¬
sion for Six was only the beginning, in a sense the material she would
be dealing with when she came to make a more definitive statement.
Certainly, Sokolow’s own comment about her goals in Opus f8 indi¬
cate that after nearly a year she had a clear idea of what she wanted to
accomplish. Reviewing Session f8, which Sokolow made for Juilliard
students and premiered in April, Horst spoke as if Sokolow had already
found a more emphatic emotional tone. He wrote: “This work, in a
jazz vein, seemed to reflect the negative, almost existentualist [frr],
state of mind as evidenced by the young of today. As such the chore¬
ography projected some telling and bitter comments” (1958b: 86). In
a later review of Session f8, Manchester noted a “remarkable solo for
Patricia Christopher reclining in a veil of black hair with all the mys-
no A Game for Dancers

teiy of a sphinx and rising to move like a goddess” (1959a: 7). Horst
also remarked on the same dance (1959: 39)- This solo may have been
a forerunner of a dance in the later Opus works, in which a woman
maintains a sphinxlike calm amid her frenzied male companions. By
Session for Eight, which Sokolow made for her company and which pre¬
miered in December 1958, ten months after Session for Six, Hering
noted that the work achieved “an impact of frenzied emotionality”
(1959: 28). She added that it was a variation of Session for Six. It appears
that after nearly a year Sokolow was well on her way to producing the
work that would become Opus ’6y.
By 1963, the relatively simple Session for Six had developed into a
twenty-minute piece of five movements. The work contained echoes of
Session for Six, particularly in the use of the jive and twist steps and in
certain recurring gestures such as the hand to the head.^ Sokolow also
used a strong formal structure, although it was far more complex than
Session for Six, and she maintained some of the alternating academic
and jazz comportment found in the earlier piece. However, now both
the choreography and Macero’s score were far more violent than the
relatively sedate Session for Six.
Opus ’6^ starts with a girl running toward two men who grab her and
slide her forward on her back.^° She turns as she slides and ends sprawled
face-down on the stage. The other dancers enter, and two other trios
repeat the slide. The boys straddle the girls’ prostrate bodies, looking
down at them. Couples form, they crouch and clutch each other. Then
a group dances while a boy and two girls stare out at the audience. Their
posture is tough and aggressive; they slouch, but their arms are tense,
as if ready to act. Next the boys hold the girls upside down and move
forward looking like multilimbed creatures.
The second dance is a twist for a girl and four boys. Macero’s score
sounds like an extended train crash, with a fast driving rhythm behind
it. The movement is coldly orgasmic. The twisting is very tight, rapid,
and obsessive. A boy moves on the floor between the legs of the other
dancers, writhing wildly; no one appears to notice. The dancers remain
locked into their own sexual worlds. The girl, here much taller than
the boys, remains especially untouched by the violent movement. At
one moment the boys hold her aloft; she sits upright on their out¬
stretched arms staring out at the audience; she runs her hands through
her luxurious hair and opens her mouth wide in a silent cry—or a yawn.
After holding the girl horizontally over their heads, the boys repeatedly
Embodying Community III

LjTine Fippinger and men of the Juilliard Dance Ensemble in Anna Sokolow’s Opus '6^
(1963). Photo: Radford Bascome. Comtesy of The Juilliard School.

throw her in the air. She then slides down their bodies, and the dance
ends with all of them on the floor. They give their heads one staccato
shake.
The third movement looks as if it had developed from the second
dance of Session for Six. The dancers take couple positions, as they did
in the earlier dance. But although they are in couple formations, they
actually dance alone, ignoring their potential partners and moving
about the stage in frantic but introverted motion. They collapse, pick
themselves up, then move on seemingly at random. They plod or wildly
execute a few mambo steps. Everyone exits except one girl, who re¬
peatedly flings her arms and body in fierce, jagged fits. The others enter,
and a couple moves to the center, dancing tightly together. As they
II2 A Game for Dancers

move, they gaze into each others’ eyes with rapt attention, then quickly
part. The dance ends with the performers in two lines along the sides
of the stage, making weakly threatening gestures at each other.
The fourth dance includes numerous moves for couples that involve
bending and lifting. Sometimes the boys press themselves tightly over
the girls’ limp bodies, sometimes they lift the girls, whose bodies seem
leaden. The boys carry the girls while attempting to execute simple
jazzy steps, but they are hampered by their oppressive burdens. The
couples separate and face off, then repeat the lifts. Finally the boys place
a hand over the girls’ eyes and they all blindly stamp, their heads raised
as if trying to see the light.
The last dance begins with couples linking elbows at the front of the
stage. They move in synchronous steps backward and forward. They lie
on the floor, arms outstretched. They stare out at the audience. They rise,
and for the first time, the music softens. They raise their arms alter¬
nately with the palms upward. They take a retire position and rest there.
Finally they join arms again and move forward as the curtain drops.
Although Sokolow’s view of contemporary society in Opus ’’6^ is
similar to that of Rooms in its narcissism, aimlessness, and disconnec¬
tion, the alienation is less complete. There are a few brief moments of
connection, and at the end there is a softening, perhaps a hint of change.
But if optimism is what Sokolow had in mind, she made no attempt to
clarify her position. When she created Opus ’dy for the Joffrey, which
was similar in a number of ways to Opus ’dy, she ended the work with
the dancers moving ominously to the front of the stage and then leap¬
ing into the audience as if on the attack. It is difficult to see this aggres¬
sive action as a sign of hope rather than as an example of a ruling class
turning on itself. In any case, Sokolow’s stance on this issue remained,
at best, ambiguous.
Much of the movement Sokolow used in her youth works drew on
the vocabulary she had begun to develop in Lyric Suite and greatly ex¬
panded in Rooms. It was here she found the gestures, steps, rhythms, and
comportment of stress and neurosis. To these she added social dance
movement that imbued the pieces with the suggestion of youth. As in
Rooms, there was no story and little use of mimetic gesture to depict spe¬
cific meaning so that the modernist formula was generally adhered to.
And while one might point out that the social dance, in its specificity,
moved the work away from a notion of essential experience, it never¬
theless read as universal in its focus on dominant-class concerns. So once
Embodying Community 113

again, Sokolow was able to critique the ruling order from the inside.
She did it by finding ways to put pressure on modernist rules without
reaching the breaking point. Still, Sokolow’s works were rarely greeted
with enthusiasm. Rather, they were recognized as worthy, and they were
tolerated. Her vision was too dark and uncompromising. Jill Johnston
succinctly summed up the dance field’s view when she dismissed Sokolow
as “an excellent choreographer with a death-rattle message” (1968: 32).
Yet Sokolow had her followers, and over the next decade they made
a considerable impact on what would be called postmodern dance. Two
of her dancers, Jeff Duncan and Jack Moore, founded Dance Theater
Workshop in 1965. Many of her company members were part of DTW,
and Sokolow worked there as well, in addition to offering advice and
encouragement. Others of her company also distinguished themselves.
Martha Clarke went on to join Pilobolus in the early 1970s before start¬
ing an independent career. Paul Sanasardo created works in the late
1950S and 1960s that owed much to Sokolow’s preoccupation with
alienation and that used some of the techniques that she employed, such
as repetition and accumulation. Pina Bausch was part of Sanasardo’s
Studio for Dance group, and, through him, she also absorbed some of
Sokolow’s approach before returning to work in her native Germany.
However, that was later. In the 1950s, Sokolow was a respected chore¬
ographer but one who was little known outside the dance world. Her
leftist past and her critical stance would have worked against her in a
time of consensus. It is hardly surprising she was not chosen to be one
of America’s “cultural ambassadors,” when the State Department began
to fund dance companies on foreign tours in the 1950s. One cannot
imagine government officials wanting Sokolow, with her vision of cap¬
italist society, to represent the United States abroad.
AF RI CAN -AME RI CAN
VAN GUARD ISM: 1940S

L ike so much having to do with race in America, black dancers’ po¬


sition in the postwar modernist vanguard was particularly compli¬
cated. In order to gain acceptance within a modern dance that was de¬
fined and dominated by whites, black artists needed to be able to obtain
access to the vocabularies and techniques of the genre as well as avenues
to patronage. But equally important, they had to learn the unspoken
norms of modern dance and, more broadly, of modernist aims, whether
to absorb or challenge them. This began to take place in the 1930s and
accelerated in the 1940s. However, the process was hardly straightfor¬
ward, considering the problematic nature of race in America. Wliite
critics struggled in the 1940s to define a “Negro” high-art dance that
was distinct from other forms, and then struggled again to decide how
a black person might be a modern dancer rather than a “Negro” one.
By the 1950s, these convolutions eased as more black dancers made
their way into the modern genre and found acceptance there. Yet prob¬
lems of discrimination and stereotyping persisted, just as they did in the
United States at large.
The difficulties that black dancers and choreographers faced in the
1930S, ’40s, and ’50s have been well documented.^ My intention is not
to contest these findings but to add to the discourse on African Ameri¬
cans in modern dance through an examination of black artists’ relation¬
ship to modernism. I will focus attention on two intertwining issues that
have attracted little notice but that affected black vanguardism in the
postwar years. The first is a critique that arose in the 1940s that served
to distinguish and separate black concert dancers from white. This cri¬
tique centered upon the black body and on Africanist elements in black
movement vocabularies. Sporadic but persistent, it tied black modern
dancers to commercialism, specifically to the exploiting of a high-art
African-American Vanguardism: 1940s 115

form through sexual and virtuosic exhibition. The critique is important


because it reflects modernist concerns regarding commerce and at the
same time points to ways in which such concerns were used to differ¬
entiate and control black dancers within a white-dominated form.
The second point, which will be dealt with in chapter 6, has to do
with the universalizing of black experience in modern dance that oc¬
curred during the 1950s. While this signified a new acceptance of black
choreographers and their dances within the modern genre, it also made
it more difficult for dancers to deal with racial issues. As was seen in
works concerning Jewish identity, modernism’s demand for universality
tended to put limits on objectives that fell outside dominant (white)
standards, values, and interests.

One of the important legacies of the 1930s revolutionary dance move¬


ment and New Deal programs was that they helped open the doors of
modern dance to black artists (Graff 1997; Perpener 2001: 189-190).
Although, as Susan Manning has pointed out (2004), leftist organiza¬
tions were more given to lip service than to action on the issue of racial
equahty, inroads were made through greater access to ballet and mod¬
ern dance training and in some cases through performing oppormnities
under the aegis of both leftist groups and New Deal programs. Kather¬
ine Dunham, for example, produced her dance-drama L’Ag’Ya (1938)
under the sponsorship of the Federal Theatre Project in Chicago, and
in 1939 she was invited to New York to choreograph some of the dances
for the Labor Stage’s production of the musical revue Pins and Needles.
The Labor Stage was affiliated with the International Ladies Garment
Workers’ Union. When Dunham presented her first company concerts
in New York in 1940, it was at the Labor Stage’s home, the Windsor
Theatre, and with the help of its producer, Louis Schaeffer. The New
Dance Group also did much to nurture black dancers. Artists such as
Ronne Aul, Pearl Primus, and Donald McKayle, all of whom began
their careers in the 1940s, studied, taught, and/or performed with the
New Dance Group.
Black dancers’ access to a white-dominated modern dance paralleled
to a degree the gathering momentum of the civil rights movement.
More than a million black soldiers fought in World War II, and when
they returned to the United States, they wanted more equality (Dal-
fiume 1968; Lawson 1976; Chafe 1982; Sitkoff 1993). Veterans took part
ii6 A Game for Dancers

in many of the early civil rights battles and some even succeeded in reg¬
istering to vote in the South in 1946 and ’47. In addition, a million
blacks moved north to jobs in the 1940s. They created a significant vot¬
ing block, and in response to this new constituency President Truman
created a Committee on Civil Rights in 1946 that recommended anti¬
lynching legislation, protection of voters’ rights, desegregation of the
armed forces, and other antisegregation legislation. Few of these recom¬
mendations were acted upon, but they were the beginning of the wedge
that would eventually lead to laws passed and enforced in the coming
decades. In the meantime, violence continued in the South, often in
conjunction with blacks trying to register to vote.
Like the small advances made in civil rights while discrimination
continued, African Americans made headway into modern dance while
in many ways continuing to be excluded. White critics didn’t fault black
dancers for being on white stages; in fact by the 1940s the dance press
in general welcomed black dancers. The criticism was more subtle. Al¬
though diffuse and often vague, it primarily hinged on an idea that black
dancers were commercializing a vanguard form. In this case commer¬
cialism was perceived primarily in two ways: as an excess of sexuality
and of virtuosity. This excess appeared to be natural to the black body
and, according to contemporary critique, was something that needed
curbing. The critique of commercialism in African-American dance will
be examined here through the work of Katherine Dunham, Pearl
Primus, and Talley Beatty, all of whom rose to prominence in the 1940s.

Katherine Dunham was the most famous black high-art dancer of the
1940s. However, Dunham’s relationship to modern dance was ambiva¬
lent, and although most black dancers of these years held similar posi¬
tions, hers was particularly uneasy. Born in Glen Ellyn, Illinois, in 1910,
Dunham followed her older brother to the University of Chicago
where she studied with Robert Redfield, one of that generation’s most
important anthropologists. In 1935, Dunham was awarded a Rosenwald
Foundation grant, which allowed her to do fieldwork in the Caribbean
for eighteen months. It also made it possible for her to do her field re¬
search under another eminent anthropologist, Melville Herskovits, who
was head of the African studies department at Northwestern Univer-
sity.2 Dunham’s work with two of the most gifted social scientists of the
interwar years not only aided her in forming her own dance theory
and philosophy, it was vital in giving her a grasp of modernist aims.
African-American Vanguardism: 1940s 117

This knowledge, coupled with her fieldwork and her training in west¬
ern dance forms, provided her with an unusually sophisticated founda¬
tion on which to develop her own dance.
While Dunham was pursuing her university education, she was also
deeply involved in dance practice. She had become interested in danc¬
ing in high school, where she learned what appears to have been a Dal-
crozian and Laban-influenced free dance. When she arrived in Chicago,
she added ballet to her studies, working with Mark Turbyfill, a leading
dancer with the Chicago Opera Ballet. In 1929-1930, she and Turbyfill
attempted to start a black ballet company called Ballet Negre, but the
project foundered for a variety of reasons, many having to do with the
racism of the times (Barzel 1983). Dunham also worked with Ludmilla
Speranzeva, who had been trained in ballet in Russia and in modern
dance with Mary Wigman in Germany. Speranzeva was a friend and
mentor who allowed Dunham to use her studio to teach classes. She also
rented a studio for Dunham in her own name after the latter’s return
from the Caribbean. With Speranzeva’s urging, Dunham formed the
Negro Dance Group, which was primarily devoted to modern dance. In
1933, Ruth Page, ballet director of the Chicago Grand Opera, invited
her to dance in her ballet La Guiablesse, set in Martinique, which helped
to make Dunham better known in the community. After her Caribbean
journey Dunham added Afro-Caribbean movement to the classes she
taught and began to choreograph works in the fusion vocabulary she
was developing. In 1937, her group was invited to perform in New York
as part of the Negro Dance Evening at the 92nd Street Y. In 1939, Dun¬
ham returned to New York with some of her company for the Pins and
Needles revue.
Dunham premiered her first fall concert in New York on 18 Febru¬
ary 1940 at the Windsor Theatre. The production was at first billed
simply as Katherine Dunham and Dance Group, but after its initial suc¬
cess the title was changed to Tropics and Le Jazz “Hot”: From Haiti to
Harlem. The first part of the tide was drawn from two suites of dances
that were part of Dunham’s program; the last indicated the breadth of
dances presented. The production met with such demand that it ran
for thirteen consecutive Sundays at the Windsor. The Dunham com¬
pany program consisted of three segments. The first included Primitive
Rhythms, which was comprised of Rara-Tonga, Tempo-Son (a Cuban slave
song and dance of possession), and Tempo-Bolero; this was followed by
two rumbas. After the first intermission there was Peruvienne, which
Katherine Dunham as the Woman with the Cigar in Tropics—Shore Excursion (1940). Courtesy
of Jerome Robbins Dance Division, The New York Public Library for the Perfoming Arts, Astor, Lenox and
Tilden Foundations.
African-American Vanguardism: 1940s 119

consisted of two dance-drama vignettes; Island Songs, which included a


Haitian carnival meringue and a Martinique folk dance; another drum
interlude; and Tropics—Shore Excursion, which featured Dunham as the
Woman with the Cigar. After the second intermission the company
performed Lejazz “Hot,” which included a boogie-woogie, Barrelhouse,
and Honky-Tonk. The program ended with another narrative piece, Bre’r
Rabbit an' de Tah Baby.^
Dunham’s program demonstrated objectives she often expressed
through articles and interviews. Foremost among these was to show
white audiences the contributions blacks had made to American culture
and, at the same time, the cross-influences between black and white
forms. She told an interviewer in 1947, “I was determined to have a
group of dancers who would be able to show the people of the U.S.
what others have contributed to our culture” (Pierre 1947: 13). One of
the key elements in her statement are the words “our culture” and the
inclusiveness they assume. For Dunham, American culture was a mix of
black and white, not two separate entities. She expanded on this idea in
an article published in 1941 ([1941a] 1978). After documenting African
retentions in Caribbean and American religious and secular dances of
the past, she spoke of European influences, particularly through English
square dance and the French quadrille, on plantation dances. Out of these
dances emerged minstrelsy, which Dunham said was important because
for the first time it was possible to see the influence of black culture on
American white culture. What Dunham called “this instrument of fu¬
sion between the two cultures” gave recognition to black cultural expres¬
sions. The result in the United States was twentieth-century dances that
drew on black tradition and grew out of black inspiration. Such cross¬
fertilization gave African tradition “a place in a large cultural body
which it enjoys nowhere else” (ibid.: 73). The concert program that
Dunham brought to New York made her thesis clear, from the African
retentions in the Caribbean dances, to European influences in the plan¬
tation and minstrel dances, to the cross-fertilization in the social dances
of the twentieth century.
Dunham not only attempted to make cross-fertilization of cultural
expression evident in her programming and selection of material, she
also demonstrated it in her choreography. This was shown first by the
fact that she choreographed her dances to fit the standard forms and
practices of the Western theater. She did not try to reproduce au¬
thentic African and Afro-Caribbean rituals and practices. Although an
120 A Game for Dancers

anthropologist, Dunham never transferred dances directly from the field


to the stage.
Dunham also incorporated techniques dominated by whites into her
choreography. Ballet and modern dance were the major techniques she
knew before going to the Caribbean, and after her return she continued
to employ them as a part of her choreographic vocabulary. Two films
made in the 1940s of extracts from a number of Dunham’s dance works
show how often she used elements of ballet vocabulary. For example, in
the duet from UAg’Ya (1938) such steps as saut de basque, poses in front
attitude, and hops in arabesque are to be found.‘^ In many of the dances
she created for the men of the company, which tended to be more vir-
mosic than the dances for the women, she also frequently employed
steps such as grand jete, cabriole, and pirouettes in a variety of positions
including en dehors, a la seconde, and attitude. These can be seen in the
film excerpts of Choros #i (1944) and VAg’Ya. In addition, she sometimes
used balletic lifts, which are shown in the film of Rhumba Jive (1941)
where a man lifts a woman onto one shoulder.^
Dunham didn’t end her ideas of racial cross-fertilization with her
choreography; she started early to develop and codify a technique that
incorporated ballet, modern dance, and Afro-Caribbean movement.
She said that among her goals was “to develop a technique that will be
as important to the white man as to the Negro” (Orme 1938: 46). To a
large extent, she accomplished that goal. In the late 1970s, Alillicent
Hodson was able to write that Dunham had contributed to modern
dance a liberation of the pelvis and knees as well as a new movement
vocabulary of the lower body (Hodson 1978: 187). She might have men¬
tioned a liberation of the shoulders, neck, and spine, as well. In looking
at the films of the Dunham company from the 1940s, it is notable how
the shoulders, in particular, are given pronounced articulation. Also
emphasized is the use of body-part isolation, fundamental in Dunham
technique, which allows the body to work in countermovements and
thus to create complex dynamic rhythms. Today many of the movement
elements Dunham codified have become an integral part of American
high-art dance as a whole.

When Katherine Dunham and Dance Group opened at the Windsor


Theatre, the critics were generally enthusiastic. They treated the pro¬
duction as a high-art recital, and they understood that it was likely to
African-American Vanguardism: 1940s 121

mark a turning point in black high-art dance. However, they showed


less understanding of Dunham’s specific intentions. Walter Terry wrote
that “last night at the Windsor Theatre Katherine Dunham broke the
bonds that have chained the dancing Negro to the alien techniques of
the white race and introduced us to a dance that was worthy of her own
people” (1940a: 13). In a lengthy Scmday piece John Martin noted: “Miss
Dunham has apparently based her theory on the obvious fact so often
overlooked that if the Negro is to develop an art of his own he can begin
only with the seeds of that art that lies within him” (1940b: 114).
Dunham was assumed to want to distinguish her dance from white
dance and maintain its separation. Terry and Martin thought this a
laudable endeavor and wrote as if it were the central issue for Dunham.
When they noticed white-dominated techniques in her work, they
generally did not approve, as witnessed by Martin’s comment: “There
is among certain of the male dancers, including Talley Beatty, a distress¬
ing tendency to introduce the technique of the academic ballet. What
is there in the human mind that is so eager to reduce the rare and gen¬
uine to the standard and foreign!” (1940b: 114; see also 1940a: 23). In
some cases, however, the absorption of “white” elements was considered
commendable. Terry wrote that although Dunham’s dances appeared
spontaneous (a quality considered typical of black dance), they obvi¬
ously were not, “for the works are beautifully constructed with an eye to
stage pattern, movement, development and compactness” (1940a: 13).
In other words, Dunham had overcome black spontaneity for white
discipline and restraint. Although such comments reveal cultural as¬
sumptions that have long been discredited, Terry and Martin were, in
fact, excited by the possibilities of Dunham’s work and were writing
extremely positive reviews. They saw her dance as an important break¬
through, even if they did not fully appreciate her goals.
Dance Observer did not review the Dunham company until it opened
the Dance Theatre Series at the 92nd Street Y on 10 November. This
venue placed Dunham more fully within a modern dance context than
Broadway, and it reinforced her high-art aims. However, there were as¬
pects of the production and company that were not typical of modern
dance recitals and that made Dunham harder to categorize. To a degree
these differences may have set the stage for future criticism. Dunham’s
vision was always ambitious. Her company was large by modern dance
standards, consisting of fourteen dancers plus two pianists and two drum¬
mers. Also, thanks to her designer husband, John Pratt, the company
122 A Game for Dancers

had elaborate sets and costumes. Few modern dance companies short of
Martha Graham’s could mount such a sophisticated production. Mary P.
O’Donnell, who reviewed the 92nd Street Y performance for Dance
Observer, seemed to sense these differences. She compared the concert
to the productions of Argentina in that “authentic folk ideas and move¬
ments were molded by formal aesthetic standards,” and she noted that
the company presented its material with “extraordinary showmanship
and a real sense of the theatre.” She ended her review by saying that
Dunham warranted “appreciation as a dancer, as a choreographer and
as an arresting personality” (O’Donnell 1940: 148).
Critics from the black press were as positive as white commentators
about Dunham. However, they were far less concerned with how she fit
into high-art dance than that she had gained success with white audi¬
ences. Dan Burley wrote in Amsterdam News that “some day a careful
chronicler will set down on paper in detail the story of how Katherine
Dunham conquered Broadway” (Burley 1940: 21), while Edgar T. Rou-
zeau in the Norfolk Journal and Guide wrote “Catherine [rzV] Dunham’s
art takes her into the very best circles. When it comes to hobnobbing
with the Astors, the Morgans, the Vanderbilts and the Rockefellers, she
is away head of us” (Rouzeau 1940: 17).
For all the positive response to Dunham’s first New York season,
there were also hints of the negative criticism to come. Terry noted that
while the dances in Tropics “possessed a veneer of cheap sophistication,”
those qualities were essential to the street and bar dances they por¬
trayed. Similarly, the jitterbug and shimmy dances of Lejazz ''Hot, ” al¬
though full of “unrestrained cavortings resolve themselves into healthy,
yet extremely lusty, folk dances” (Terry 1940a: 13). He concluded that
if such dances appeared cheap, it was undoubtedly owing to white per¬
formers who had cheapened them. Rouzeau of the Norfolk Journal and
Guide echoed Terry in saying that the “cheap sophistication” of Island
Songs and Tropics was necessary to convey the style and feeling of the
dances. He felt that Dunham was at her best in Peruvienne, because the
graceful movements showed “her subde technique at its best.” (Rouzeau
1940: 17). Margery Dana of the Daily Worker was positive overall;
however, she felt the second half of the program “was too influenced by
musical comedy,” by “gaudy costumes, cheap sensuality and, of course,
‘Uncle Tom’” (Dana 1940: 7).'^ Dana appears to have misunderstood
Dunham’s intentions in depicting minstrelsy in the section entitled
BreT Rabbit an' de Tab Baby. As noted, Dunham felt that minstrelsy was
African-American Vangnardism: 1940s 123

the first theatrical form to show the influence of black culmre on white
and wanted to make that point, whereas Dana saw the depiction of min-
sti'elsy as succumbing to stereotypes.
1943) when Dunham presented her next set of concerts in New
York, the uneasiness hinted at in earlier comments about the theatri¬
cality and sexuality of her dances began to coalesce into more overt
accusations of commercialism. Dunham had starred in Cabin in the Sky
on Broadway in 1940, then toured with the show. When it closed in
California, Dunham and the company stayed on, performing in several
Hollywood films and in nightclubs. Sol Hurok, one of the most power¬
ful impresarios in the United States, then invited Dunham to appear
under his management. Hurok added a singer and a Dixieland jazz en¬
semble to the piano and percussion Dunham had previously employed.
He also gave her production the title Tropical Revue, although Hurok
himself described the production as “somewhere between revue and
recital” (1953: 59; see also 1947: 267). However, the categorizing of
Dunham’s dances as a revue changed audience expectations, despite the
fact that with one exception the 1943 program did not differ markedly
from that of 1940. This exception, though, was important and may have
influenced critics, although for the most part they did not acknowledge
it. The major addition to the repertory was an ambitious work entitled
Rites de Passage. It replaced Peruvienne and Island Songs from the 1940
production, with the rest of the program similar to that of Tropics and
he Jazz “Hot.”
Rites de Passage was composed of three sections: “Fertility Ritual,”
“Male Puberty Ritual,” and “Death Rimal.” The fertility rimal centered
on a man and woman who in their duet advanced, retreated, and circled
each other while executing a rhythm of backward and forward pelvic
movement. The ritual aspect of the dance was emphasized through the
repetition of movement; through the dancers’ gazes, which did not meet;
and through their faces, which remained neutral throughout. The male
puberty ritual depicted an adolescent boy who dreamed or underwent
trance, then faced ordeals before being welcomed as an adult into the
community. The last segment dealt with the death of a chief, the soci¬
ety’s mourning, the crowning of a new king, and his marriage to the
queen.
In the original program note Dunham described Rites de Passage “as
a set of rituals surrounding the transition of an individual or group of
individuals from one life crisis to another. The ritual period, often at
124 A Game for Dancers

Lucille Ellis and James Alexander in the “Fertility Ritual” section of Katherine Dunham’s
Rites de Passage (1943). The photograph was taken during the company’s London season in
1948. Photo: Roger Wood. Cotirtesy of Jerome Robbins Dance Division, The New York Public Library fm' the
Petforming Arts, Astor, Lenox and Tilden Foundations.

once both sacred and dangerous, is under the guidance of the elders of
the community: the entire community joins in this critical transition so
that the individual may, in a changed status, have a complete rejoining
with the society.”^ The program note went on to say that the work did
not represent any particular culture but was an abstraction meant to
capture the “emotional body of any primitive community and to pro¬
ject this intense, even fearful personal experience.” The costumes by
John Pratt reinforced the idea of nonspecificity in fantastic garments
that recalled no definite source. The women wore earth-colored ban¬
deau tops and short skirts, braided to resemble natural fiber. Their hair
African-American Vanguardism: 1940s 125

was hidden beneath high, crownlike wigs of braids. The men wore briefs
and over them loincloths. Like the women, they also wore braided wigs.
As Ramsay Burt has pointed out (1998: 166-179), Dunham was
concerned not just with understanding the technical aspects and overt
social function of the dances of so-called primitive people but also the
deep psychological needs and desires that motivated them.^ There is
ample evidence that Dunham saw in these psychological motivations
particularly powerful examples of universal experience that both the
artist and anthropologist sought to tap into. As such, Dunham placed
herself on the side of modernism. She wrote in an article in 1941 that
the artist “seeks to recreate and present an impression of universal human
experience—to fulfill either human needs or wants. The instrument is
the specific art form which may have been chosen; the effectiveness de¬
pends upon skill in handling the form and upon the originality of the
individual imagination. But the experience which is given expression
cannot be either too individual or too specific; it must be universal”
([1941b] 1978: 55). She then went on to explain the value of anthropo¬
logical study: “In such a survey, the student of anthropology gradually
comes to recognize universal emotional experiences, common alike to
both the primitive Bushman and the sophisticated cosmopolitan” (ibid.).
At the time, however, commentators did not make a connection be¬
tween modernist goals and the universal emotion Dunham tried to
embody through an abstraction of primitive ritual. When Tropical Revue
opened on 19 September 1943 at the Martin Beck Theatre, Louis Horst
and John Martin agreed that Rites de Passage was the best and most se¬
rious work of the evening, but they didn’t say why and made no attempt
to analyze it (Martin 1943b: 24; Horst 1943: 90). Roberta Lowe, critic
for the Amsterdam News, felt that Rites de Passage, which she called
“Haitian inspired,” was not authentic in the sense that it did not portray
a specific culture convincingly (1943: 9). John Meldon of the Daily Worker
also saw Rites de Passage in specific cultural terms, saying that it was based
on an authentic Haitian folk dance (1943: 7), although he disagreed
with the Amsterdam News in that he felt the work was the highlight of
the evening.
Rather than focus on the modernism of Rites de Passage, commenta¬
tors turned their attention to what they considered the commercial na¬
ture of Dunham’s production, showing greater ambivalence about her
work than they had earlier. This was undoubtedly aided by the fact that
the fertility section of Rites de Passage caused a flurry of publicity as the
126 A Game for Dancers

company toured the United States. Descriptions of the Dunham reper¬


tory as “hot” and “sizzling” abounded. When the group reached New
York, John Martin mentioned that Dunham now called her production
a revue, although he said it was in truth “part recital, part nightclub”
(1943b: 24). Anatole Chujoy, writing of Dunham for the first time in
Dance News, thought she presented “an excellent show,” but treated it
as light entertainment, refusing to intellectualize her dances or to try
“to find a justification for what she is doing and for the way she is doing
it” (1943: 3). Louis Horst gave Tropical Revue only cursory treatment
in Dance Observer, saying its material was “frankly of revue and night¬
club caliber” and so needed only brief comment (1943: 90).
Martin’s remarks in 1943 reflected a shifting attitude toward Dun¬
ham’s work that would become more emphatic as time went on. He
wrote: “As an anthropologist the gist of Miss Dunham’s report seems to
be that sex in the Caribbean is doing all right. She even seems to have
remembered certain testimony along this line that eluded her when last
she appeared here-abouts, and which no doubt Hollywood has brought
to her mind” (1943b: 24). Martin’s suggestion was that Dunham had
made her dances sexier after being in the Hollywood marketplace and
that because of it they were anthropologically less authentic. Martin
overlooked the fact, which he had noted in his 1940 reviews, that Dun¬
ham’s works were never meant as anthropological reconstructions. Yet
Martin praised the dances of the first part of the program, “dealing with
primitive rituals and the like” which, he said, “fell well within the recital
category” (ibid.).
Soon, the Dunham company left on another national tour during
which Rites de Passage was dropped in Boston after complaints led to a
visit by the censor (Perpener 2001: 152-153). When John Martin saw
Tropical Revue again the next year on its return to New York, his opinion
had taken a much harder edge. He found it to be “from anthropology
a la Minsky to the frankest kind of honky-tonk. There are rumbas and
rituals, barrel-house boleros and blues, and the characteristic accent of
all of them is pelvic” (1944a: 15). Martin’s indiscriminate grouping of
rituals with folk and bar dances as all somehow distastefully sexual was
an idea that had not appeared in his earlier reviews, and his comparison
of Tropical Revue with the burlesque shows at Minsky’s theater suggests
moral outrage more than reasoned argument. But although his reaction
was more extreme than most, a thread ran through several other re¬
views tying Dunham’s production to commerce. Writing in New Masses
African-American Vanguardism: 1940s 127

under the name Francis Steuben, Edna Ocko linked Dunham’s work to
Broadway: “Miss Dunham’s recitals several years ago set an enviable
standard for authenticity and impeccable taste. It is no news that Broad¬
way is not exacting on either score” (Steuben 1943: 26). However, both
John Meldon in the Daily Worker and Henry Simon in PM praised the
company, Meldon calling the group one of the brightest and most en¬
tertaining on Broadway (Meldon 1943: 7; Simon 1943: 22).
Significantly, Roberta Lowe of the Amsterdam News defended Dun¬
ham’s use of sexuality. In doing so she was perhaps responding to criti¬
cism of the production. Lowe contrasted the sophisticates who would
see the beauty and vitality of Tropical Revue with the barbarians who
would be attracted to the “hip-swinging” that she noted was a hallmark
of the show. Her reference to barbarians may well have been a swipe at
prurient white interest. She commented more than once on what she
called “hip” rather than “pelvic” movement, but said that it was rele¬
gated to its proper place in the street and bar dances (1943: 9).
Edwin Denby, in a New York Herald Tribune review, did not conflate
sexuality with commercialism in Tropical Revue. He didn’t care for Rites
de Passage, but he faulted it on formal grounds, saying that Dunham had
not been able to fit African expressive method to that of the Western
stage. “ The movement,” he said, “is based on African dance elements but
the choreographic plan is that of the American modern school” ([1943a]
1986: 143). He noted that the latter, like all Western dance, made its ef¬
fects through contrasts of weight, volume, speed, and dynamics, whereas
African methods favored reiteration. Yet contrasts were not absent from
African art, he argued, citing African sculpture as a form that showed
“distinctness of contrasts and their surprising reconciliation” (ibid.).
Denby preferred Dunham’s bar and street dances. He singled out the
three different “hot” styles of Dunham’s Brazilian Bahiana, Cuban Shore
Excursion, and American Barrelhouse: “She has observed these fashions
of tropical entertainment intelligently, she knows what they are each
after. Better still, she knows there is nothing arch about a hot style, that
its expression is serious and sometimes even angry” (Denby [1943a]
1986: 142). He thought Dunham’s gestures “provocative and yet dis¬
creet and she can even keep a private modesty of her own. As a dance
entertainer, she is a serious artist.” Denby linked art, entertainment,
and sexuality in positive terms that a number of other critics found im¬
possible to do.
Sociologist St. Clair Drake wrote in 1981 that “the deep strain of
128 A Game for Dancers

Puritanism in American life that tended to turn sexuality into prurient


interest was a constraint that serious black performers had to break
through” (Drake 1981: xii). He went on to say that Dunham showed her
audiences that “sexuality as expressed in some aspects of African and
New World black tradition has symbolic meanings relevant to utility as
well as to sexual satisfaction, and that ostensibly erotic dancing can be
cherished for the sheer joy of the bodily movement and display of danc¬
ing sldll” (ibid.).
Dunham’s accentuation of pelvic movement accurately reflected the
African retentions she wished to show, but such movement also related
to an element of form and function that was important to her. In an
article written in 1941, Dunham made the point that in Caribbean
peasant societies social and sacred dances with pelvic movement were
directly associated with sexual activities; that is to say, form and func¬
tion were closely linked (Dunham [1941c] 1978:192-196). She described
the banda dance of Haiti, which was performed at funerals and had a
sexual character, “in keeping with the African philosophy which closely
associates procreation with death, perhaps as a compensatory effort”
(ibid.: 193). She mentioned, too, the danses grouilliere, which included
both sacred and secular dances and which featured the grinding of the
hips. Dunham was extremely interested in the links between form and
function as part of her general choreographic strategy. She took her cue
from that connection in making her fertility duet in Rites de Passage. The
movements she chose were ones she had seen in the Caribbean, and
they reinforced the purpose of the dance.
However, for some American observers form could only follow func¬
tion under certain conditions, at least if the result was to be considered
art. Commentators who willingly accepted sexuality as art when it was
depicted as neurotic or pathological, for example in the works of Gra¬
ham or Tudor where the use of the pelvis was also in evidence, were far
less eager to give sexuality a place in art when it was portrayed as a uni¬
versal and often joyous part of life. This was particularly the case when
sexuality was being expressed by black bodies. Then the tendency was
to tie sexuality to commerce and to the exploitation of a high-art form.
Dunham’s dances, clouded by Puritanism and racism, lent themselves
to the fear of commercialism that modern dance commentators were
already feeling.
However, it must be said, too, that Dunham fed the fire of sexual
controversy. The public attention it brought her helped ensure the sur-
African-American Vanguardism: 1940S 129

vival of her company while allowing her to demonstrate some of the con¬
tradictions of racial assumptions. Among her tactics was, in interviews,
to encourage the image of herself as the sexualized primitive then put it
in tension with the educated sophisticate. Afargaret Lloyd described
her this way in an interview: “Lithe and slender in her tight, leopard-
cloth slacks, gentle of voice and manner, she was poised and charming,
and very keen in her observations. It was difficult to realize that not many
generations ago her ancestors had left Africa in slave ships” (1949: 247).
Lloyd’s patronizing tone is a little amusing considering how thor¬
oughly Dimham controlled the interview. Certainly Dunham was aware
of what she was doing. She promoted the dichotomy in her publicity
materials, spoke about it often, and wrote at least one article on the sub¬
ject. In “Thesis Turned Broadway” ([1941 hi 1978: 55-57) she explained
that she saw synthesis rather than conflict in the two aspects. As an
anthropologist she had no interest, she said, in presenting the exotic on
stage. She wanted to investigate the cultural and psychological frame¬
work of why people danced. She sought the common patterns that
bound them together as human beings. Sexuality was part of a univer¬
sal human totality that was no more to be denied than birth and death.
Dunham, in emphasizing the compatibility of sexuality and serious
artistic intent in her dances and her person, chose to disrupt stereotypes
both on the stage and off.
There was another element to the issue of sexuality in Dunham’s
dances that was not openly discussed but that she understood and dis¬
turbed. This was the stereotype of blacks as primitive, in the sense of
uncivilized, and sexually uninhibited. As Sander Gilman has pointed
out in his study of racial stereotypes and pathology, black women in the
nineteenth century came to be associated with unbridled sexual appetite
in contrast to white women. This was part of a broader agenda of dif¬
ferentiating the races, with blacks portrayed as antithetical to whites.
Both art and science in the nineteenth century depicted black women
as inherently immoral and uncontrollable due to their primitive sexual
makeup. That is to say, they were not only primitive in spirit but in the
physical construction of their sexual organs. It was a small step to link
the image of the black woman to that of the prostimte in nineteenth-
century iconography (Gilman 1985: 76-108; Giddings 1992). Other
scholars have shown that black men were similarly cast as sexually in¬
satiable and for similar reasons (Fanon 1967; Mercer 1994: 131-220;
Collins 2004: 149-180).
130 A Game for Dancers

Although any scientific basis for such stereotypes had long been dis¬
missed by midtwentieth century, the stereotypes themselves persisted,
as Gilman shows in his analysis of modernist literature (1985: 109-12 7).
Dunham unsetded these stereot3/pes in her dances by seeming to con¬
form to them through her movement vocabulary and then countering
them with anthropological insights. As is evident in her interview with
Lloyd, she could also conjure similar tensions through her own per¬
sona, dressing to suggest sexuality and the jungle while speaking with
“civilized” charm and intelligence. Denby said there was provocation in
Dunham’s dances, and it is likely he was correct. When Dunham re-
mrned to New York in 1944, Denby reiterated many of the ideas he had
expressed earlier, concluding that “Miss Dunham’s wit and charm and
complete knowingness are the key” (i944d: 10). Thus, Dunham used
sexuality to make a number of points about race in America, but in the
1940s few commentators understood her intent.

Dunham was not alone in being criticized for the sexuafity of her dances.
Pearl Primus, Talley Beatty, and others were also faulted, particularly
for pelvic movement, which offended white sensibilities and was linked
to the exploitation of modern dance. Dunham, however, was seldom
condemned for emotional or virtuosic excess, other elements linked to
commercialism in black dance. Here Primus and Beatty were especially
singled out.
Primus’s career was different from Dunham’s in that her training
placed her firmly within the New York modern dance scene. Born in
Trinidad in 1919, Primus’s family moved to New York when she was
three (Green 2002; Barber 1992). She attended Hunter College High
School and then Hunter College, where she studied biology and pre¬
medicine with the idea of becoming a physician. After deciding that
medicine was not for her, she entered Columbia University for a mas¬
ter’s degree in anthropology. In the meantime she had started dance
classes at the New Dance Group and also studied with Martha Graham
and Louis Horst. Primus’s association with the New Dance Group, under
whose auspices she often appeared, gave her legitimacy as a modern
dancer. Also, her performances fit the concert format, being mostly solos
and small group works with modest costumes, music, and scenic effects.
Such details were important, as they allowed a dancer to be categorized
and judged according to the norms of the genre.
African-American Vanguardism: 1940s 131

Dance Observer followed Primus’s early career avidly. When five young
dancers presented a concert at the 92nd Street Y on 14 January 1943,
Primus earned the best reviews. Gervase Butler wrote: “Pearl Primus is
not a “Negro” dancer in the conventional sense of the term. Trained in
the New Dance Group and now studying with Martha Graham, her
aptitude and feeling for racial rhythms is secondary to a direct and un¬
usually ample response to the modern idiom. Pearl Primus is about the
most exciting evidence this reviewer has seen that there are no inher¬
ent confines to the dance capacities of her race” (1943: 27). Butler de¬
fined a “Negro” dancer as one who used “racial rhythms,” that is, African
and diasporic sources of movement. He felt that for Primus these were
secondary to the movement vocabularies of modern dance. The fact
that Primus gave primary importance to the modern idiom meant that
her race would not limit her to specificity; she could communicate uni¬
versal emotion and in this way become truly a modern dancer. Butler
suggested that it was Primus’s training and vocabulary that made the
difference, rather than her subject matter. This was an important dis¬
tinction, for Primus’s dances were in many instances specifically about
black experience. Her program included A Man Has Just Been Lynched,
later called Strange Fruit, and Hard Times Blues, which dealt with the
plight of tenant farmers. These social protest themes were welcomed
during the war when modern dancers were expected to deal with prob¬
lems of modern life. In addition, her concerts included African and
diasporian-inspired suites, spirimals, and jazz-based abstract dances.
John Martin was also impressed with Primus, saying “it would be
hard to think of a Negro dancer in the field who can match her for
technical capacity, compositional skill and something to say in terms
that are altogether true to herself both racially and as an individual artist”
(1943a: 5). Martin particularly pointed out the originality of Primus’s
choreography and its strength of composition. In African Ceremonial, he
said she employed “genuine primitive movement and shapes it precisely
and authoritatively into artistic theatre form,” while in Hard Times Blues
she explored the music’s rhythmic possibilities to achieve original cho¬
reography that had great impact. In A Man Has Just Been Lynched, she
capmred the passion and terror of the onlooker’s experience in extra¬
ordinary movement.
Leftist and black critics greeted Primus with as much enthusiasm as
Martin and Butler but with a different emphasis. Edith Anderson an¬
nounced in her headline in the Daily Worker that Primus was the first
132 A Game for Dancers

Negro to make her debut at the Y, thus breaking the color barrier. She
spoke of Primus’s “terrific power, exuberance, ease and control” and
that she “tore the house down” in her debut (1943: 7). Peter Suskind of
the Norfolk Journal said Primus embodied the spirit of the dance and
had something new and original to say. He also noted that Primus had
appeared with four other dancers who were white, and she was the only
one who was cheered for a curtain call after each dance (1943: 14).
In October 1944, less than two years after her first appearance at the
92nd Street Y and several months after her first solo concert, also at
the Y, Primus made her Broadway debut. As Hurok had done with
Dunham, Max Jelin, Primus’s impresario, arranged her material in the
shape of a revue, with musical numbers by Frankie Newton’s band and
solos by Josh White placed between her dances. Although some critics,
including John Martin, were still highly laudatory, the general response
to her work was less positive than it had been earlier. Some blamed the
introduction of elements unrelated to the dance, in particular the band
numbers, which they thought intrusive.^ But others found something
amiss with Primus’s choreography and presentation.
Edna Ocko, writing under the name Francis Steuben in Nero Masses,
summed up the problems. She found Primus’s dances to be deeply felt
but lacking in strucmral coherence: “These very social and artistic con¬
victions demand imaginative and connotative movement of a creative
caliber not always achieved by Miss Primus” (Steuben 1944: 28). Ocko
went on to say that often the only element that bound the dances to¬
gether was their emotional intensity. “The stuff and substance of the
craft itself—its inner connectives, its patterns of developing movement,
its structural continuity—basic dance elements which cannot be con¬
jured out of thin air—are resolved by Miss Primus in what seems too
easy and too unoriginal a manner” (ibid.).
Edwin Denby also found the production only partially successful, al¬
though he still thought Primus “a thrilling dancer.” He felt the protest
dances, which included the recently choreographed The Negro Speaks
of Rivers, were less compelling than the playful jazz pieces, in which
Primus’s choreography was inventive and the dances well-made. He
said the protest works were less successful because when the audience
agreed beforehand with the artist, the protest no longer became the cen¬
ter of attention. Instead viewers focused on the elegance of the dance ex¬
ecution, and Primus occasionally “hammed” her protest pieces ([1944c]
1986: 247-248). N. S. Wollf of Dance News agreed that it was in the
Pearl Primus in Hard Times Blues (1943). Photo: Gjon Mili/Time Life Pictures/Getty Images.
134 A Game for Dancers

social protest dances where Primus failed. Wollf found them too literal
(1944: 7).
The most extensive analysis of Primus’s work in the wake of her
Broadway debut was Lois Balcom’s two articles in Dance Observer (1944a,
1944b). Balcom couched her analysis in terms of three different types
of audience. The first was the least demanding. It consisted of specta¬
tors who came with certain preconceptions about “Negro” dance, all
of which Primus fulfilled. This audience liked the dances inspired by
African tribal sources (stereotypes of the primitive), the jazz dances
(stereotypes of minstrelsy), and the modern protest pieces, because they
all seemed to reinforce preconceived ideas of race. The second audience
was more sophisticated and less easily pleased. It believed that the
Negro dancer could not be modern because she could not overcome
her race. This audience declined to like the “primitives” because they
were interpretations rather than recreations. Nor did it appreciate the
modern protest dances because they did not reflect the style of Graham
or other established moderns. The preference here was for the jazz
dances because Primus was performing them best at that moment.
The last audience believed that modern dance and the Negro race
were not mutually exclusive. Clearly Balcom considered herself among
this group and as such sought to give Primus advice. She began by quot¬
ing John Martin’s concept of modern dance as he expressed it in 1933
in The Modem Dance: “From the desire to externalize personal, authen¬
tic experience, it is evident that the scheme of modern dancing is all in
the direction of individualism and away from standardization” (Balcom
1944a: 111). Balcom did not consider such a modern dance to be in con¬
flict with race. Being modern was a matter of being oneself in the deep¬
est sense, and if one were Negro, then racial heritage would be included
in being modern. The problem for Balcom was that she did not believe
Primus was offering authentic experience in her modern dances. She
located the trouble in a lack of discipline. Using Strange Fruit as an ex¬
ample, Balcom said it was the most effective of Primus’s modern dances,
but it was “formless, undisciplined, unconcentrated” (1944b: 123). She
noted that “it is not enough to say that ‘discipline’ is not notably a
Negro characteristic” (ibid.). The artist must discipline herself in order
to be able to use her medium to express her intention. In flinging her¬
self about the stage, doing spectacular leaps, stamping her feet, and beat¬
ing her thighs. Primus was not showing discipline. Balcom enlarged on
her theme: “Economy, tautness, like ‘discipline,’ may seem not to be
African-American Vanguardism: 1940s 135

characteristic of the Negro temperament, but the Negro artist has no


choice but to master the seeming paradox of expressing with economy
and tautness the emotional prodigality of her race!” (1944b; 124). If
Primus did not get herself under control, Balcom said, she would re¬
main little more than a talented and popular entertainer.
Certainly in these articles Balcom showed her concern for nurturing
a black modern dancer. She appared to want Primus to succeed but was
worried she would fall into the arms of easy entertainment where she
would never become modern. One could argue that as a talented indi-
tddual Primus had moved too quickly from unknown to star, a view that
was reflected in several reviews of her Broadway debut. But there is in
Balcom’s criticism a shift from the problems of a young dancer to an
issue of race. In her view, blacks were by nature undisciplined and emo¬
tionally profligate, and Primus’s dances, in which she flung herself about
and made spectacular leaps, were examples of it. Balcom’s comment
resonates with criticism of Dunham as sexually extravagant, and in both
cases this excess was tied to exhibitionism and commerce.
Although Balcom’s articles constituted only one opinion. Dance Ob¬
server was a highly influential voice in the dance field and devoting two
lengthy articles to a single dancer’s performance was extremely unusual
for the magazine. In addition, once Balcom set the critical tack, the other
Dance Observer critics followed her direction. Besides finding fault with
the undisciplined character of Primus’s modern dances, they began to
link sexuality, exhibition, and commerce in them. Horton Foote spoke
of the “pelvic dullness” of Primus’s “primitive” dances, while the spiri¬
tuals and songs of protest “emerged thin and superficial” (1945a: 44-45).
In 1946, Mary Phelps was not altogether clear about what she objected
to in Primus’s work, although she harshly criticized the choreographer
for insulting her own dance, a paean to the Watusi people, which Phelps
found undignified (1946a: 24). Several months later another Dance
Observer critic, Eleanor Anne Goff, again connected pelvic movement
to exhibitionism. Speaking of the “primitive” dances she said: “As for
the rest, what could be beautiful savagery descends to the blow by blow
repetition of uninspired torso and hip flexions, play-dances, reminiscent
of, but not as successful as, some of Katharine [sic] Dunham’s works,
and tinged with a theatricalism and exhibitionism which rightly do not
belong in sincere works of this nature” (1946: 76). Goff, in what can
only reflect her own rather fevered imagination, looked for “beautiful
savagery” from a black dancer, but then criticized the dancer for a use
136 A Game for Dancers

of repetition and pelvic movement, which were Africanist elements,


though hardly “savage.” The jazz and play-dances, Goff said, were the¬
atricalized and exhibitionist.
The charge of exhibitionism in Primus’s dance extended beyond sex¬
uality and emotional excess to technical virtuosity. Margaret Lloyd,
writing in her 1949 book on modern dance, exemplified the suspicion
that surrounded Primus’s remarkable elevation, something that Balcom
also mentioned. She commented:

Hard Time Blues to a record by Josh White, is phenomenal for its ex¬
cursions into space and stopovers on top of it. Pearl takes a running jump,
lands in an upper corner and sits there, unconcernedly paddling the air
with her legs. She does it repeatedly, from one side of the stage, then the
other, apparently unaware of the involuntary gasps from the audience.
The feat looks something like the broad jump of the athlete, but the take¬
off is different, she tells me, and the legs are kicked out less horizontally.
The dance is a protest against sharecropping. For me it was exultant with
mastery over the law of gravitation, and the poor sharecroppers were for¬
gotten. “Going up in the air does not always express joy,” she explained.
“It can mean sorrow, anger, anything; it all depends on the shape the
body takes in the air.” So what appeared to be a triumphant assault was
evidently a projection of defiance or desperation. (1949: 271-272)

Lloyd was skeptical about Primus’s intentions in her jump, feeling it was
out of keeping with the themes of her dance. However, for Primus it
was a means of expressing a number of different emotions, depending
on its form. It was part of her means of expression, and to have sup¬
pressed it would have been like asking Graham to suppress her signa¬
ture contractions. It is important to note that, according to Primus, she
did not simply stage protest through subject matter. Hard Times Blues,
The Negro Speaks of Rivers, and other dances may have expressed protest
in their themes, but they also embodied it in movement. In this she ful¬
filled John Martin’s notion of a modern dance as one that abstracted
from life experience through individual means, a quality he, himself, saw
in her dances. Yet for Lloyd and others the virtuosic nature of Primus’s
dance was a means of exhibiting herself rather than calling forth au¬
thentic emotion.
In 1948, Primus, who eventually earned a Ph.D. in anthropology from
New York University, won a fellowship from the Rosenwald Founda-
African-American Vanguardism: 1940s 137

tion to study dances in Africa. She spent nine months there, and it was
only with her return that she was once again wholeheartedly welcomed
into the high-art dance fold, this time as an anthropologist and expert in
African dance as much, if not more, than as a choreographer of modern
dance. She returned to Africa often after her first trip, and her appear¬
ances in the United States included numerous lecture-demonstrations
of African dances and culture as well as dance concerts. In this way, her
virtuosity was folded into broader interests and was given anthropolog¬
ical legitimation.

Although critics may have faulted Primus, they also treated her as one
of their own. Talley Beatty, on the other hand, came from the Dunham
company and was frequently condemned for his early choreography,
as well as criticized for his dancing. Like Dunham and Primus, he was
charged with exhibitionism. Born in 1923 in Chicago, Beatty began his
career as a youngster with Dunham in the 1930s (Beatty 1990, 1993;
Nash 1992). He had been studying for only a few months before he
began appearing with her group. At that time Dunham had not yet been
to the Caribbean and was teaching primarily ballet and modern dance.
When she left for her fieldwork, she suggested Beatty study with other
teachers. He took modern classes with Kurt and Grace Graff, but he
was not allowed to mix with white students, so he went early in the
morning and took a class alone in the dressing room. Edna McCrae, a
well-known Chicago ballet teacher, also agreed to have him as a student,
but again he could not take classes with her white pupils. He therefore
followed what was happening in the studio from an office that had a
small window looking out on the classroom. This kind of discrimina¬
tion was common in Beatty’s early career and made a lasting impression
on him. He later commented: “That was part of our lives then in the
South Side. I mean it was really quite a hard role. We were very aware
of it and I think that perhaps, I was kind of a rough character. I mean I
was rather frank so I’m sure I must have said something about it. But
there was nothing. ... It was either doing it that way or not doing it at
all” (Beatty 1990).
When Dunham’s company began appearing in New York in the late
1930s, Beatty was often singled out for the brilliance of his dancing.
However, he was also one of the men John Martin castigated for having
an inappropriate balletic virtuosity. Martin seems not to have understood
Talley Beatty as a young dancer in New York. Courtesy of Jerome Robbins Dance Division, The New
York Public Library for the Pei forming Arts, Astor, Lenox and Tilden Foundations.
African-American Vanguardism: 1940s 139

that Beatty had been trained as a ballet dancer and therefore bore the
marks of that training on his body. This must have been particularly
visible in Dunham’s choreography, which utilized ballet along with Afro-
Caribbean and modern dance techniques. That Beatty was an accom¬
plished ballet dancer is supported by the fact that Lincoln Kirstein
asked him to appear in Lew Christensen’s Blackface (1947) with Ballet
Society and to choreograph a work for the company. Although at
that time Martin was expressing a commonly held opinion in saying that
ballet was inappropriate for black bodies, Beatty himself could not have
disagreed more. He later commented, “I think that [ballet] is part of our
culture, a part of the American culture. . . . It’s just as close to blacks
as it is to whites. This is a part of us” (Beatty 1990). Like Dunham, he
saw ballet as a legitimate source for all Americans, and like her, he used
ballet technique and vocabulary in his own choreography.
Beatty began his choreographic career on 3 November 1946 with
two dances for the inaugural concert of Choreographers’ Workshop.
He made a duet for himself and Lavinia Williams entitled Blues, and
another duet for himself and Tommy Gomez called Southern Landscape.
The only publication to cover the concert was Dance Observer, and
Louis Horst said of Beatty’s offerings simply that they showed him to
be a better dancer than choreographer (1946: 124). Two months later,
Beatty had enlarged Southern Landscape to three scenes, which he again
presented under the aegis of Choreographers’ Workshop. Again Horst
reviewed. He wrote that the scenes, which were listed as “Defeat,” “The
Mourner’s Bench,” and “The New Courage” had exciting and beautiful
passages, “but one longs for more restrained choreography and less dis¬
play of obvious technical stunts by Air. Beatty himself” (1947: 20).
On II May 1947, the Workshop presented the five best pieces of
more than twenty given that season, and Southern Landscape was one of
them. By this time Beatty had completed the entire work, which was set
to a score by Elie Siegmeister. The inspiration for Southern Landscape
was Howard Fast’s novel. Freedom Road (1946), which in part dealt with
blacks being forced off the land they had been given after the Civil War.
Beatty choreographed a series of dances that focused on the grief and
loss that African Americans felt at being dispossessed.
We are fortunate that Southern Landscape survives today in a video
recording of a reconstruction overseen by Beatty in 1992. There is
also a film made in 1948 of Beatty himself in the work’s central solo.^^
The first section, “The Defeat in the Fields,” shows men and women
140 A Game for Dancers

working the land and then being terrorized there. It is the closest thing
to dramatic action in the work. Next is “Mourner’s Bench,” which
Beatty later said he imagined as a solo for a man who had just returned
from a nighttime burial.'^ The third section, “My Hair Was Wet with
the Midnight Dew” is a dance of sorrow performed by two couples. A
program note from 1948 states that the “survivors of the attack wan¬
dered at night through their ravaged fields.“Up, Ring Shout” fol¬
lows. Historically the ring shout was a slave dance of religious release
in which the participants moved counterclockwise in shuffling steps to
the accompaniment of spirimals. The dance went on for long periods
of time during which possession sometimes took hold of one or more
of the members (Emery 1988: 120-126; Stuckey 2002:214-47). Beatty
did not attempt an authentic recreation of the dance but rather used it
as the basis for his own choreography to signify the release of commu¬
nal grief. The dance begins with shuffling steps, which gradually lead to
more varied movement. The movement becomes increasingly fast and
more densely packed, creating a sense of building intensity. The last
scene of Southern Landscape, entitled “Settin’ Up,” was described in the
program as a dance of mourning for the dead. According to dancer/
historian Joe Nash, who saw early performances of the work, Beatty
changed the choreography of the last section when he revived the piece
in the 1990s.
The early reviews of Southern Landscape were generally encouraging.
When Dance Observer critic Mary Phelps saw the finished work at the
Choreographers’ Workshop concert, she did not mention stunts, as
Horst had done earlier (1947: 65). She called Southern Landscape “a
straight impassioned cry of suffering and determination.” She said it
only needed increased contrast in the overall design to make it entirely
effective. In Dance Magazine, Doris Hering described Southern Land¬
scape as a thunderstorm and Beatty as the lightning in it. He was, she
said, “an exciting, inventive, intuitive, virtuosic dancer,” and his chore¬
ography, like his style, was “wild, undisciplined, and pixy-like” (1947:
20-21). Southern Landscape itself was over-long and diffose, and con¬
tained “much that is merely stagey and spectacular.” She singled out
“Mourner’s Bench” as an example. The work also contained “much that
is sincere and shows real talent and theatrical sense.”
In 1948, Beatty formed a company that presented its first full con¬
cert on 24 October 1948 at the 92nd Street Y. It was at this point that
criticism of Beatty’s virtuosity began to harden. The concert program
Kim Bears-Bailey and members of the Philadelphia Dance Company in Talley Beatty’s
Southeifi Landscape (1948), which Beatty revived for the company in 1992. Photo: Patented
Photos. Courtesy of the Philadelphia Dance Company.
142 A Game for Dancers

consisted oi Rural Dances of Cuba, Southern Landscape, Saudades do Brazil,


Blues, and Kanzo. Of these, Kanzo, which depicted Haitian vodun initia¬
tion rites, and Rural Dances of Cuba were new. The New York Times and
the Herald Tribune ran announcements of the concert, but neither paper
covered the debut. The concert was also ignored by Dance News and by
the black press. Doris Hering, giving Southern Landscape another view¬
ing, was not impressed. She wrote that it was ambitious but naive. “Even
Mr. Beatty’s solo, ‘Mourner’s Bench,’ while visually arresting, does not
grow from the deep roots that make sincere concert dance. There is a
feeling of movement chosen merely for its eye appeal” (1948: 35).
In her brief review for Dance Observer, Harriet Johnson said the Affo-
Caribbean dances were too closely aligned to Dunham and advised
Beatty to pursue work more on the order of Southern Landscape (1948:
122). This was mild criticism considering that a number of the Dance
Observer writers had begun to focus attacks on Beatty’s virtuosity. Ear¬
lier in 1948, Martha Coleman had written that “Mr. Beatty’s technical
skill continues to express only his virtuosity, and to confuse exhibition
with expression” (1948: 7). By 1950, Robert Sabin was speaking of the
need to pare down “attempts at virtuosity” in Beatty’s Vieux Carre, a
work that survives only in Sabin’s description of it as imitative of Mas¬
sine’s Gatte Parisienne (1950; 71). That same year, Joan Brodie wrote a
particularly harsh review of a concert that was similar in content to the
92nd Street Y performance of 1948. She was incensed by what she
called Beatty’s “cheap sensationalism,” which she blamed on both the
sexuality she perceived in some of his material and on his technical skill
(1950: 138-139). “It is,” she wrote, “upsetting to see a group of well-
trained, hard-working dancers ... performing in such trash” (1950: 139).
Even Southern Landscape did not escape her censure. Conceding that it
was a serious work, outstanding for its dignity and attempt at real com¬
munication, she wrote that “Mourner’s Bench,” while affecting, needed
more conviction “so that a more genuine emotion might come through
the brilliant movement” (1950: 138-139).
One has to turn to Cecil Smith, editor of Musical America, to find a
more thorough and balanced analysis of Beatty’s work. Writing of the
92nd Street Y concert in 1948, Smith commented:

His YMHA program was consistently brilliant, and at times as breath¬


taking as any dancing to be encountered on the American stage today.
But it offered hardly more than a suggestion of Mr. Beatty’s potential in-
African-American Vanguardism; 1940s 143

vention and scope as a choreographer. Three of the four longer com¬


positions were opportunistic, in the sense that they were derived lock,
stock, and barrel from the style of Miss Dunham’s folk hallet, though
they are better composed, for the most part. It was only when Mr. Beatty
turned from Cuba, Brazil and Haiti to a highly individual and heartfelt
modern-dance work, Soiithem Landscape, that a new dimension of depth
appeared. Drawing its inspiration from Howard Fast’s chapters on the
Reconstruction period, Soiithet~n Landscape seeks to express the defeat,
the grief, and the religious release of a people who had briefly tasted
freedom and had seen it taken away from them again. The second and
third of the five movements, in particular, capture the essence of this grief
in genuinely moving form, and Mr. Beatty’s solo dance is remarkable for
the way in which his superb virtuosity remains constantly meaningful.
(1948: 10)

Smith’s review gives a clearer idea than most of the content of Beatty’s
concert and the relative merits of the works. He found the dancing bril¬
liant rather than tasteless and exhibitionist, as had other critics. He took
no particular offense at the Afro-Caribbean dances, although he found
them derivative, while Southern Landscape impressed him for its sincer¬
ity and emotional force, especially Beatty’s solo. Undoubtedly referring
to Southern Landscape, he said the concert suggested Beatty’s potential
and scope as a choreographer.^^ Yet what appeared to Smith as both vir-
tuosic and “constantly meaningful,” was for the critics of Dance Observer
and Dance Magazine excessive. For Brodie, Beatty’s works lacked “form,
meaning, unity, and good taste” (1950: 138).
In order to try to come to grips with the radically different views
Beatty’s work elicited and the recurring critique of his virmosity, I
would like to compare “Mourner’s Bench” with two solos by Jean Erd-
man of about the same year. In the late 1940s, Erdman was considered
one of the most promising of the new generation of modern dancers. I
have chosen to use her dances for comparison rather than a solo by an¬
other male dancer because the great majority of modern dance chore¬
ographers and performers in the late 1940s were women. For example,
eight out of the eleven dance performances reviewed in Dance Observer
in January 1948, when Southern Landscape was also reviewed, were by
women. What I want to emphasize here are differences in the approach
to choreography and performance that race, class, and gender might
have made during these years.
144 A Game for Dancers

The extant film of Beatty in “Mourner’s Bench” shows him perform¬


ing a technically demanding solo that depicts the spiritual anguish of a
grieving man. But not only is the dance demanding, it is choreograph-
ically innovative, that innovation made possible in part by the technical
abilities of the choreographer-dancer. Beatty used the bench as a part¬
ner and support on which to balance in a wide variety of ways. He
stands while moving into high extensions, he braces his legs beneath the
bench and cantilevers his body out into space like the bowsprit of a ship,
he rolls off onto the floor in a controlled extension to the side, and
more. The dance is a technical tour-de-force, but it is also a metaphor
for pain as the body struggles against the gravity that pins it down.
Joan Erdman’s Ophelia was called “probably Miss Erdman’s finest
solo,” notable, in particular for its dynamic quality (Larkey 1949: 152).
This dance can be seen on video in a reconstruction Erdman oversaw in
the early i99os.^'^ Looked at solely in terms of the movement, it consists
primarily of walking steps, now slow, now fast, sometimes coming to
rest in the hitched hip pose of medieval sculpture and paintings. The
walking steps are most often combined with turns, sometimes step¬
ping turns, sometimes larger circles made while doing walking steps. It
also includes occasional high-swinging leg kicks that swirl the dancer’s
long skirt in an arc above her head, high side extensions, and shallow
lunges to the side. The Graham contraction is also in evidence from
time to time. There is much use of arm movements, which tend to be
nonspecific gestures. Eor example, at certain points the dancer extends
her arms overhead, or holds her arms close to the body and moves her
lower arms from side to side in unison. None of this movement extends
the general modern dance vocabulary, and according to Erdman, as noted
earlier, she had little interest in creating a vocabulary of her own. Like
many other young dancers of the time, she was more concerned with
using existing dance vocabularies in an impersonal way, rather than de¬
veloping new personal forms.
In contrast, Beatty extended a ballet and conventionalized modern
dance vocabulary in “Mourner’s Bench.” He did this first by abstracting
movement and gestures derived from African and diasporic movement.
Eor example, he used torso articulation in which shoulders and torso
move forward and back, adding this to modern dance contractions. In
addition, he extended modern dance movement through his experi¬
mental use of the bench, in which he found a number of steps that could
work in concert with the prop he had chosen. Eor instance, at one point
African-American Vanguardism: 1940s 145

in the film Beatty lies across the bench perpendicular to its length and
rolls his body slowly from one side to the other, turning each time into
a high arabesque. At another moment he plies into a crouching fourth
position that he uses as a springboard into a series of quick turns that
take him behind the bench, his back to the audience, from which he as¬
sumes a wide second position bending back over the bench toward the
spectators.
In contrast to Erdman’s dance, Beatty also stretched and filled out
the movement to the maximum reach of his body, making the dance
image appear large and forceful. The steps flow from one to the other,
rarely coming to a halt in a pose, giving the impression of dynamism
and a stretched tension. The movement in Erdman’s dance often stops
before being fully developed, as if held back and made small. Rather
than powerful, the dance image is one of neatness and control. The
many halts between steps also continually stop the flow of movement.
Of course this may have something to do with the mood she was trying
to convey in the dance. In the video, she speaks of wanting to commu¬
nicate a sense of “psychic dismemberment” in the piece. But here one
might also point to another of Erdman’s dances of the same period,
which had quite a different aim, but nevertheless still employed move¬
ment in a similar way. Hamadryad (1948) is a dance that Erdman said
was meant to convey the “sheer enjoyment of movement, a voluptuous,
joyous outdoors feeling. . . . There was nothing civilized about it.”^^
This dance contains more varied and energetic action than Ophelia,
but it, too, constantly freezes the flow and stretch of the steps. Although
one must always consider the fact that we are not seeing Erdman, her¬
self, but another dancer of a later period performing, nonetheless, I
think it is fair to say that the arrangements of the steps make the move¬
ment appear highly controlled, despite the sense of abandon Erdman
wanted to convey. The steps move from pose to pose. In fact the dance
begins with a pose, head looking up, torso twisted, legs bent in a demi-
plie, arms low and open. Erom this position the dancer moves into a pro¬
file pose in a deep contraction out of which she reaches upward, then faces
front repeating the first pose; her arms undulate in a quick waving mo¬
tion; she resumes the first pose; and then repeats the phrase. This start-
and-stop construction constantly breaks the force of the action. Each
pose is clear, but there is little sense of a powerful thrust of movement.
Critics may well have found the control in dances by Erdman and
other white performers reassuring. African Americans invaded this polite
146 A Game for Dancers

world and celebrated the power of the human body. Primus’s jump,
Beatty’s pounces and balanced high extensions, and Dunham’s undulating
hips proved disturbing in their unabashed energy and forthrighmess.
Critics read these elements as uncontrolled and perhaps in a subliminal
way, dangerous. And it can be argued that there was, indeed, danger to
the status quo in these dancers’ movement. Dunham’s use of the pelvis
and shoulders challenged American Puritanism, while Primus and Beatty’s
use of virtuosity called into question restrictions placed on blacks’
physical means of expression, particularly means that were ruled alien
to them. All three dancers threatened barriers constructed to control
blacks and separate them from whites. Denby had commented that the
injustice of slavery and its legacy was hardly protest, as it was generally
agreed that slavery had caused egregious harm. Therefore, so-called
protest dances could be celebrated with little disturbance to long-held
behefs and practices. But Dunham, Primus, and Beatty attacked discrimi¬
nation and hypocrisy where it could sting, and they did it not so much
with stories told but through a corporeal politics of danced movement.
The resulting case made against them was that they trafficked in com¬
merce, that virtuosity and sexuality were simply employed for display
and as such could not contribute to authenticity. Accusations of com¬
mercialism lessened markedly as black artists made their way into mod¬
ern dance in increasing numbers during the postwar years. In the 1940s,
however, the sense that black dancers were exploiting a vanguard form
played into white stereotypes of blacks as physically unrestrained and
emotionally profligate while at the same time reflecting legitimate
concerns within the field about the influence of the market on modern
dance.
6

AF RI CAN -AM E RI CAN


VANGUARDISM: 1950S

I n the 1940s, charges of commercialism and exploitation devalued the


work of black modern dance choreographers even as black artists be¬
came increasingly visible in the field. What is surprising is how thor¬
oughly these accusations disappeared in the 1950s. In fact, dance com¬
mentators’ references to any aspect of race became far more muted.^
There were several possible reasons for this shift. Certainly by the be¬
ginning of the 1950s, African-American dancers and choreographers
were becoming more firmly established within modern dance than they
had been earlier. Most modern dance and ballet teachers in New York
accepted African Americans as students, and although ballet companies
still resisted black dancers, African Americans found their way into a
number of white modern dance companies, including those of Sophie
Maslow, Anna Sokolow, Martha Graham, Jean Erdman, Alwin Niko¬
lais, and Merce Cunningham (Greene 1997). At the same time, white
dancers occasionally performed in black companies, as Remy Charlip,
Louis Falco, and Esta and Eve Beck did in Donald McKayle’s group. In
addition to increased numbers of dancers, there was a gradual accept¬
ance of Africanist movement in modern dance. During the 1950s, black
popular culture, although generally diluted for white audiences, greatly
expanded in the mainstream. When teenagers routinely shook their
hips to rock-and-roll and Elvis (“the pelvis”) Presley appeared on prime¬
time television, it was no longer shocking to see such movement in van¬
guard dance. As the works of black choreographers came to be viewed
as modern rather than “Negro,” their dances were universalized in keep¬
ing with modernist imperatives. While this helped integrate black cho¬
reographers into a white genre, it tended to erase concerns that were
considered specific to black experience.
The muting of race also came at a moment when fears of the Soviet
148 A Game for Dancers

Union and a communist takeover were at their height. In such an at¬


mosphere few commentators wanted to call attention to America s prob¬
lems for fear of seeming unpatriotic or of lending aid to the enemy. In
her groundbreaking book, Cold War Civil Rights, Mary L. Dudziak
documents the U.S. government’s concern over international criticism
of American racism and the vulnerable position in which it placed the
country. Dudziak asks, “How could American democracy be a beacon
during the Cold War, and a model for those struggling against Soviet
oppression, if the United States itself practiced brutal discrimination
against minorities within its own borders?” (2000; 3). The “Negro prob-
lem,”to use Gunnar Myrdal’s term, caused foreign relations difficulties
with countries in Asia, Africa, and Latin America, as well as the Soviet
Union. Although international criticism of American racism was com¬
mon during the postwar years, Dudziak notes that to criticize racial ills
from within the United States was to risk being labeled subversive.
For commentators of the time, one way out of the predicament of
dealing with racial problems was to avoid the subject altogether. For
example, Daniel Bell, in his discussion of American society and its
troubles in The End of Ideology, failed, except in the most cursory of ref¬
erences, to mention the existence of racial issues. Instead he spoke of
alienation and fear of mass culture being major social concerns, along
with the threat of communism. Other observers of the midcentury
American scene from William H. Whyte Jr. to Arthur Schlesinger Jr.
also ignored the subject. Like Bell, they talked of the fear and malaise
they saw around them, focusing on middle-class angst.
The degree to which race was off the table in dance was most fla¬
grantly illustrated by Katherine Dunham’s Southland, which premiered
in Chile in 1951 and dealt with a lynching. After the work was performed,
Dunham met with increasing harassment from the State Department.
This continued not only through the remainder of the South American
tour but on subsequent appearances abroad. After performing South¬
land in Paris in January 1953 to mixed reviews and silence from the
American Embassy, Dunham decided she didn’t have the strength to try
to perform it in the United States. But the ordeal wasn’t over; through¬
out the 1950S, the U.S. government found numerous small ways to
attack Dunham when her company was on tour outside the country.
It comes as no surprise, then, that the State Department neglected to
award Dunham one of the touring grants that were started in1954 as
part of the country’s cultural diplomacy (Hill 1994; Dunham [1951]
African-American Vanguardism: 1950s 149

1978). During the 1950s, the State Department also attempted to si¬
lence Josephine Baker for criticizing American racial discrimination
while on tour in Latin America. Her passport could not be revoked be¬
cause she had become a French citizen; however, Baker found it in¬
creasingly difficult to obtain engagements in Latin America and at one
point was barred from landing in the United States on her way from
France to Mexico City (Dudziak 2000: 67-76). Although Dunham and
Baker’s experiences were the exception rather than the rule, they sug¬
gest the climate that prevailed at the time regarding the discussion of
racial issues.
The chilling effect engendered by the Cold War, coupled with
modernist demands for universality and consensus culmre’s erasing of
difference combined to make themselves felt in dance discourse in a
number of ways. For example, a subtle shift occurred in the categoriza¬
tion of dance and dancers. In the 1940s, dances had been divided between
“Negro” and modern with most black dances categorized as “Negro,”
because they were viewed as dealing with specific rather than universal
experience. In the 1950s, as more black dancers came to be considered
modern, the two categories shifted to a more generalized “ethnic” dance
and modern dance. Ethnic dance occupied a lower position in the dance
hierarchy than modern dance and the even more prestigious ballet.
Dances that appeared to deal with the roots of black experience were
categorized as ethnic, whereas dances that were viewed as universal
were modern. So, for example, Geoffrey Holder, who was from Trinidad,
frequently choreographed works that were inspired by aspects of Afro-
Caribbean culture. Critics routinely used “ethnic” to describe his sources.
Although such a categorization might have been to some degree un¬
derstandable in Holder’s case, at least one writer also used “ethnic” to
describe Alvin Alley’s sources for his bitterly despairing Blues Suite
(Bernstein 1958: 88). This work drew on Alley’s memories of the bars
and brothels of the area in which he lived as a child in Texas. Nonethe¬
less, the work, which depicted the effects of poverty and discrimination
in rural America, was classified as ethnic in its source material.
In another example, Walter Terry categorized Dunham and Primus
as ethnic dancers in his book The Dance in America ([1956] 1971), even
though neither would have described herself in such terms. Terry did
place Janet Collins and Ronne Aul in his section on modern dance and
mentioned Primus there, although he said she had turned increasingly
to ethnic dance after having traveled to Africa.
150 A Game for Dancers

Gerald Myers, writing in the 1980s in an essay on ethnic and mod¬


ern dance, addressed the issue of black dancers being designated as eth¬
nic rather than modern. He defines ethnic dance as one that attempts
to conserve tradition by replicating the past, whereas modern dance is
characterized by originality and rebellion against the past; . modern
dancers are found rebelling against the tradition that begets, sustains, and
to some degree entraps them” (Myers 1988: 24). However, he notes that
even when it is acknowledged that black choreographers are simply re¬
ferring to dances from, for example Africa or Haiti, those sources are
considered so overwhelming that the new dances “represent a contem¬
porary ethnic genre.” He adds that “connotations of aesthetic rawness,
of unsublimated primitivism, mingle in the mischievousness of the label
‘ethnic’” (ibid.). At the least, the “ethnic” designation given to African-
American dances in the 1950s indicates a move away from dealing di¬
rectly with race, and at most, to the reduction of a potentially explosive
issue into an innocuous matter of colorful folk and exotic cultures.
A more effective method than euphemism for avoiding the discus¬
sion of race was simply, like Daniel Bell, not to address it. Among the
few articles in the dance press that met racial issues head-on in the 1950s
was a Dance Magazine interview with Janet Collins that chronicled the
difficulties she had experienced breaking the color barrier in ballet. Be¬
fore migrating to New York, Collins had been a student of Carmelita
Maracci in Los Angeles. Collins told Norma Gengal Stahl that while
still living in California she had auditioned for Leonide Massine, at that
time choreographer for Ballet Russe de Monte Carlo. Massine had com¬
plimented her on her dancing but said he could not accept her because
the company did not have enough specialty roles for her, “and for the
corps de ballet, he said he’d have to paint me white” (Stahl 1954: 28).
Asked what her response had been, Collins said she “cried for an hour,
and went back to the barre” (ibid.). Collins was finally accepted by the
Metropolitan Opera Ballet, where she became premiere danseuse while
also choreographing modern dance works for concerts of her own.
It is clear from Stahl’s interview with Collins that the dance com¬
munity was aware of the difficulties black dancers faced, and it may be
assumed they were aware of racial issues on a broader level. However,
in most instances, members of the dance press avoided speaking of race
during the 1950s in much the same way they avoided decoding works
that dealt indirectly with current troubles. There are many examples
of writers failing to mention what would seem to be obvious racial ref-
African-American Vanguardism: 1950s 151

erences in works. A case in point is a review written by Lucy Wilder in


1952 that described Donald McKayle’s early dance, Saturday's Child, as
“a stunning solo piece ... in which he delivers Countee Cullen’s poem
as the accompaniment for the dance.” Wilder continued to speak at
length about the dance, how the movement heightened the poem’s ef¬
fect, how the solo “was confined in space, physically taut and abrupt,
emotionally direct and intense” (1952: 58). Yet she did not mention that
Cullen had been a major figure in the Harlem Renaissance, or that the
subject of the poem dealt with the anguish of grinding poverty, or how
McKayle’s dance, created and performed by an African-American cho¬
reographer, may have related to issues of race in the poem. The racial
element of McKayle’s Saturday'^s Child did not escape the notice of the
black community, however. McKayle says in his autobiography that
the solo, which he danced in several informal performances around New
York, resulted in his being invited to join the Committee for the Negro
in the Arts, an elite group that included such members of the Harlem
Renaissance as Langston Hughes and Paul Robeson and younger rising
black artists, writers, and performers including Sydney Poitier and Harry
Belafonte (McKayle 2002: 35-36).
In another instance Doris Hering wrote exclusively in aesthetic terms
of Louis Johnson’s Spiritual Suite, presented in a concert at the 92nd
Street Y on 27 November 1955. The opening section, she said, had to
do primarily with design, while his solo to “Motherless Child” “was af¬
fecting because it was a free, spontaneous expression of emotion” (1956:
77). Similarly, Robert Sabin, reviewing a concert by Janet Collins at
the 92nd Street Y on 3 December 1951, spoke of her works Three
Psalms of David and Spirituals simply as dances of religious inspiration
(1952: lo-ii). Both, he said were derivative of Graham, but he made no
reference to the fact that the former suite drew on Jewish sources and
the latter on Aifican-American ones, let alone how these two works
might have been treated in company with each other. This was some¬
thing John Martin had mentioned in his review of the spirituals when
he had seen them in 1949, remarking that although Collins was “aware
of racial backgrounds in the spirituals . .. they are in every sense dances
rather than an exploitation of heritage” (1949). He went on to note that
Collins was similarly interested in Hebraic dances and was working
with composer Ernest Bloch on a series of them, which she would chore¬
ograph. Setting aside Martin’s attempts to separate modern dance from
“an exploitation of heritage,” the point here is that he addressed the
152 A Game for Dancers

subject of race in dances that he viewed as related to it and tied the spir¬
ituals to the group of dances drawn from Jewish material as works that
dealt with particular culmral and racial backgrounds. We might be led
to believe that the lack of racial references in writing of the 1950s meant
that race was simply no longer an issue in modern dance, were it not
that it had figured so prominently a few years earlier and that its disap¬
pearance occurred when the social and political atmosphere in the
United States discouraged discussions of difference. In order to explore
what the muting of race in dance discourse meant for Afi-ican-American
choreographers, I want to mrn to two important works from that decade:
Donald McKayle’s Rainbow Round My Shoulder and Talley Beatty’s The
Road of the Phoebe Snow, both choreographed in 1959.

McKayle, who was born in 1930, was the youngest child of a middle-
class immigrant family from Jamaica who settled in Harlem (McKayle
2002). His father was a maintenance man for various theaters in New
York until the war, when he started a career as an airplane mechanic.
His mother worked in the garment district. McKayle went to the best
public schools his parents could find and was an excellent student. A
high school friend who was taking classes at the New Dance Group in¬
vited him to a Pearl Primus concert, which proved to be transformative.
Although he had little dance experience, McKayle auditioned for a
scholarship at the New Dance Group and won it. A year later he was
performing with the New Dance Group and choreographing dances of
his own. His first major work Games (1951) was a hit, and McKayle be¬
came a young choreographer to watch.
During the 1950s, McKayle not only made works for his own group,
he danced with a number of other companies, including those of Anna
Sokolow (he was part of the original cast of Rooms), Martha Graham,
Jean Erdman, and Merce Cunningham. He also continued to dance with
Maslow and appeared on television and in several Broadway shows, in¬
cluding House of Flowers (1954, Balanchine/Herbert Ross), West Side Story
(1957, Robbins), and Copper and Brass (1957, Sokolow). By the time
McKayle choreographed Rainbow Round My Shoulder he was a mature
artist and dance maker.
McKayle conceived Rainbow Round My Shoulder almost by accident
(McKayle 2002).^ He had been hired at a summer resort in the Catskills
as a weekend entertainer. There he met the great blues singer Leon
African-American Vanguardism: 1950s 153

Bibb, who was also on the entertainment staff. They decided to do a


number together, set to Bibb’s singing of the chain gang song “Tol’ My
Captain.” These songs were sung by prisoners in the South who were
sent out to break rock for roads. The songs kept the rhythm of the pris¬
oners’ work as well as expressed their despair and hopes. In the number
they worked out. Bibb sang while he pulled McKayle across the stage
on a piece of chain that bound them together. As he sang. Bibb held
rocks in his hands, which he pounded together for emphasis. McKayle
gradually freed himself of the chain and danced unrestrained for a few
moments before once again binding himself to the other man and col¬
lapsing to the floor. The number was enormously successful, and Mc¬
Kayle went on to create a full dance work from this idea.
AIcKayle gradually identified other chain gang songs he wanted to
use, including “Rocks and Gravel,” which opens Rainbow Round My
Shoulder wi^dn. the words:

Well it’s early in the morning, huh!


Baby when I rise, oh-well-a, huh!
Got the aches and pains, lordy mama, huh!
Make a man wanna die, oh-well-a, huh!

As the song is heard, seven men enter in a line, their arms clasped.^
The “huh!” at the end of each line reflects the rhythmic effort of the
prisoners’ work. McKayle’s choreography embodies the same rhythmic
effort, as the men’s heads and torsos snap in harsh unison at the end of
each line.
The next song, “I Had a Gal,” introduces the sole woman in the
piece. As the men lie collapsed on the stage, she conjures up fantasies of
love and caring. The movement is balletic and lyrical. Her arm move¬
ments embrace and encircle. She disappears, and the men rise as the
title song is heard:

I got a rainbow, huh!


Tied around my shoulder, huh!
I’m goin’ home, huh!
My Lord, I’m goin’ home

The men move to the song with sharp contractions as they embody the
work being done. But the theme is soon interrupted by another brighter
An early cast of the men in Donald McKayle’s Rainbow Round My Shoulder (1959): From
back row left, William Louther, Tommy Johnson, Donald McKayle, Claude Thompson,
Morton Winston, Charles Neal, and Don Martin. The photograph was taken at Jacob’s
Pillow. Photo: John Lindquist ©, The Harvard Theatre Collection, Houghton Library. Courtesy of the
Donald McKayle Papers, University of California, Irvine, Libraries.

one that again introduces the girl, this time as a young woman with a
bouncing flower in her hair. She flirtatiously shakes her shoulders and
hips as she moves about the stage. One of the men tries to catch her, but
she remains teasingly elusive. She disappears as “I Got a Rainbow” is
heard once more. Then she is back, this time as a mother gently ad¬
monishing her neglectful child, then as a wife or lover helping up and
encouraging a fallen man. The next song, “Take This Hammer,” speaks
of an escape:

Take this hammer, carry it to the captain


Tell him I’m gone
I don’t want no cold iron shackle
Round my leg
If he asks you was I laughing
Tell him I’m crying
African-American Vangnardism: 1950s 155

If he asks you was I running


Tell him I’m flying

The men gather in a crowd. Two run off. Shots are heard. One man
somersaults on, and is picked up and carried off. The other man enters
with the girl and they dance to “Another Man Done Gone”;

They killed another man


He had a long chain on
Another man done gone.

The man collapses, and the curtain falls.


McKayle’s use of a socially conscious theme was not unusual for him;
in addition to Saturday'’s Child he had created Games (1951), which dealt
with childhood poverty and discrimination, and Her Name Was Harriet
(1952) (later reworked as Her Name Was Moses), based on the life of
Harriet Tubman of Freedom Train fame. This subject matter reflects
the influence of the New Dance Group ethos and McKayle’s work with
Maslow and Sokolow. But although McKayle used pantomime in Rain¬
bow in a way that sometimes recalls Maslow, he gave far more attention
to metaphoric movement and to its relationship to the flow and pattern
of the dance. For example, although there are suggestions of ham¬
mering as the men’s arms reach over their heads with their hands locked
together, the gesture is abstracted. The resulting metaphor is then in¬
terspersed with other movement so that it becomes part of the dance
pattern. This may be why the men’s dancing is so compelling at this
moment; it is transformed into a poetic abstraction of physical labor
rather than being an imitation of everyday activity.
McKayle also used the musical rhythm to suggest movement. The
clearest example occurs in the work songs where the dancers’ sharp
contractions coincide with the “huh!” at the end of the musical phrase.
The embodied sound reflects both the musical rhythm and the rhythm
of the prisoners’ work. McKayle has pointed out another way in which
the relationship between movement and music is especially close."^ When
he rehearses Rainbow, he said he often sings the songs as the dancers
move in order to encourage them to listen closely to the music rather
than simply to count it. The phrases are uneven because of the mode
of labor the men of the chain gang were doing. By singing the ragged
musical phrases, he tries to instill in the dancers’ bodies the rhythm that
156 A Game for Dancers

Donald McKayle and Shelley Frankel in a 1962 photograph of McKayle’s Rainbow Round
My Shoidder (1959). Photo: ©Jack Mitchell. Courtesy of the Donald McKayle Papers, Univerisy of California,
Irvine, Libraries.

gives the timing to the movement and helps form the physical shape of
the dancers’ bodies in motion. In the original production, further con¬
nections between the dance and music were made through a male cho¬
rus that accompanied Leon Bibb and that was arranged in a pyramid on
steps leading from the orchestra floor to the stage with Bibb at the top
of the pyramid (McKayle 2002: 116). The chorus members swayed as
they sang, which made them participants in the stage action. In this way
African-American Vanguardism; 1950s 157

the music and dance created a whole that is missing when the dance is
done, as it often is today, with recorded music.
When Rainbow Round My Shoulder w2ls premiered on 10 May 1959 at
the 92nd Street Y, it was not greeted with the same level of enthusiasm
that Ga?nes\\2id been eight years earlier, although the work has since be¬
come a classic.^ Games dealt with the street games children played when
McKayle was growing up in East Harlem, a neighborhood of black,
Puerto Rican, and Jewish immigrants. It depicted not just ways in which
the neighborhood children entertained themselves but also the insecure
and violent lives they led as a result of poverty and discrimination. The
work revolved around an incident that happened to McKayle as a child
when a friend was wrongly arrested for street fighting (McKayle 2002:
46-47). In the dance work, a child is violently beaten by a policeman.
Although the policeman was originally performed by a white dancer,
Remy Charlip, the children were played by an integrated cast, and it
is understandable that reviews did not mention race as an element of
the work.
Rainbow also referred to a situation that involved blacks and whites,
but African-American references were more obvious in the later work
because of the music and Africanist aspects of the movement. McKayle
had studied briefly with Primus (who left the New Dance Group shortly
after he arrived) and then with Haitian dancer Jean Leon Destine, as
well as having grown up in a household of Jamaican-born Americans.
He was therefore famifiar with various African diasporic forms. Although
McKayle has said he was not aiming to create black characters in Rain¬
bow, his use of pelvis and shoulder articulation, particularly in the flir¬
tatious girl’s dance, is based in Africanist movement. Certainly, too,
McKayle was thinking of racial injustice when he created Rainbow. He
wrote in 1965: “Then I became intrigued with the music of the South¬
ern Negro chain gang. What first captured me were the pulsing, rest¬
less rhythms. They seemed wrapped in the chains that bound the suf¬
fering men together, and they seemed to explode in desperation and
anger. The lyrics were sardonic and then, in turn, biting, sensual, and
filled with protest” (1965: 59). In a more recent interview, McKayle told
a story that linked Rainbow Round My Shoulder to the growing civil
rights movement in the dance’s focus on African-American dreams of
freedom and equality. He said that shortly after the work’s premiere he
was invited to present it on CBS television. After the broadcast, a news
program followed with Walter Cronkite reporting on the lunch counter
158 A Game for Dancers

sit-ins in the South. McKayle said it was an unforgettable moment for


him because, “it was a time of great change and people putting their
lives on the line to effect that change. I thought the dance was part of
that in its own way” (McKayle 2000).
Race, though, made little appearance in reviews of Rainbow Round My
Shoulder. Harry Bernstein of Dance Observer even neglected to mention
that the theme involved a chain gang, saying simply that ''Rainbow Round
My Shoulder, the new choreography of Mr. McKayle to traditional
music, produced, at its best, an exciting emotive response. But an over¬
literalness of the choreography and incompatible movement for Mary
Hinkson, as the Woman, lessened the work generally and made of it
one of narrow dimension” (1959; 93). In the limited space generally al¬
located for each review in Dance Observer, Bernstein had little opportu¬
nity to elaborate his comments, but he did manage to say that the work
was set to traditional music (without specifying the source), that it was
over-literal and its dimension narrow in part because of incompatible
movement for the woman. One assumes that Bernstein objected to the
narrative, mimetic aspect of the dance because of his use of the term
“over-literal.” Was it the last scene that put him off, where the men es¬
cape and one is killed? Certainly that is the most mimetic and the one
that most specifically refers to the injustice of the men’s situation. But
Bernstein also wrote that the movement for the woman was incom¬
patible, which with its literalness gave the work a narrow dimension. This
sentence is far from transparent, but I don’t think it is unreasonable to
surmise that he meant the woman’s use of hip and pelvic movement
gave the work an Africanist character that made it less universal and
more specific in character. Whatever the details of Bernstein’s cau¬
tiously coded verdict, he did not refer directly to race while employing
the epithet used routinely for any work that dealt too unambiguously
with controversial subjects. He also couched his comments wholly in
aesthetic terms without reference to social meaning.
In her review for Dance Magazine, Selma Jeanne Cohen did mention
the theme of Rainbow Round My Shoulder, but she also spoke in much the
same aesthetic language as Bernstein. She said that McKayle’s ideas were
built on characterization and so had functional integrity, while lacking
advenmrousness (1959: 17). Since Cohen was sympathetic to the experi¬
ments of Merce Cunningham and other objectivists, she was undoubt¬
edly speaking of the expressional thrust of McKayle’s work, which she
found conservative. She added that only the last scene “lacked internal
African-American Vanguardism: 1950s 159

motivation and seemed contrived.” Cohen found Hinkson’s perform¬


ance stunning for its “remarkably fluid grace and dramatic sensitivity,”
while the dancing of the men “had a persistent forcefulness.”
P. W. Manchester differed from her colleagues, finding the work “very
fine” in her Dance News review (1959b: 14). But except for noting that
it was based on “Negro prison songs” she did not mention race. Instead
she described the dance in terms of its emotion and form. McKayle, she
said, had encompassed in the dance “the despair, rage, terror and long¬
ing of men from whom even hope has departed. Mary Hinkson, glori¬
ously feminine, seductive yet gentle, is the image of all womanhood and
the subtle placing of her movements against the male background gives
the work its essential balance.”
Several years later Walter Sorell wrote in Dance Observer that in Rain¬
bow Round My Shoulder McKayle had “created haunting images and
conjured up many moods of despair and hope, of struggle and ultimate
defeat. In every phrase he pictured the stifled and strangled being in
man and his desire for freedom” (1962: 91). This review was more sym¬
pathetic than some of those of three years earlier, but Sorell couched his
comments in universal terms, not connecting the work in any way with
racial injustice.
The integrated character of McKayle’s company contributed to his
work’s being perceived as nonracial. So did his use of a predominately
modern dance vocabulary, which in viewers’ eyes rendered the dance
more universal than specific. But while the response to Rainbow Round
My Shoulder demonstrated the degree to which black artists had been
absorbed into modern dance, it also meant that injustice was seen in
abstract terms. The rules of modernism made it difficult to deal with
problems that could not be viewed as universal from the standpoint of
a dominant white society, and this situation may not always have served
African-American interests.

Talley Beatty’s The Road of the Phoebe Snow was a more immediate hit
than Rainbow, but it also was treated in generalized terms. In addition,
the work showed the degree to which Beatty’s movement style and
vocabulary had been accepted in modern dance and how difficult it had
become to use these elements to political purpose. Beatty’s virtuosic
amalgam of Africanist, modern dance, and ballet movement, which
critics had found deplorable ten years earlier, was now celebrated as
i6o A Game for Dancers

exhilarating, and Beatty himself was accorded a kind of success he had


never known.
The Phoebe Snow was the name of a train that passed through an
area of Chicago where Beatty’s father, a house painter, had sometimes
worked when Beatty was a boy. As Beatty explained it, in those days his
father was not allowed to join the painters’ union and so had to take
work in the poorest neighborhoods (Beatty 1990). The ghetto beside
the railroad tracks was one of these areas. The dance work centers on a
group of disaffected youth. Like Sokolow’s dances on the same theme,
Phoebe Snow is set to a jazz score, here by Duke Ellington and Billy
Strayhorn. Beatty’s young people, however, are far more alienated than
Sokolow’s—angrier and more destructive.
When The Road of the Phoebe Snow was premiered on 28 November
1959 at the 92nd Street Y, the program mentioned no narrative. It listed
only the names of the choreographer, composers, and dancers. How¬
ever, the work does have an outline of a story: a couple is attacked by a
group, and as the girl runs from her tormenters, she is killed by a pass¬
ing train. The curtain goes up on a boy looking down at a girl, lying at
his feet.*^ A crowd gathers, running on from various directions, during
which the girl is carried off stage. This beginning action turns out to
be nearly identical to the ending of the work so that, after the fact, the
viewer realizes that the rest of the piece is a flashback of sorts. Treating
the narrative in this way allows the work to be read as almost completely
“pure” dance. Until those last moments, the focus remains on a driving,
inexorable flow of movement, which finally becomes a metaphor for
anger and disaffection.
The first dance begins with two men leaping sideways from the wings
onto the empty stage. Three other men follow, dancing to a jangling,
percussive musical score. They contrast long stretched pulls of the body
and arms with snaking hips, snapping fingers, and then slow, drawn out
poses. Fluidity is constantly set against hunched over hep-cat steps.
Then the five girls come on with hard, unison movement. The men and
women form into two independent groups dancing against each other in
elegant, dense phrases. At the end the girls run and jump onto the boys.
The next dance starts in a circle, all the participants facing each
other. They move out confidently, then as they go from the brightly
lighted circle to the shadows beyond it, they seem to lose their nerve.
They stumble back into the center then crouch and point into the
dark. But what looks like the set-up for a dramatic scene never happens.
African-American Vanguardism: 1950s 161

Rather, the dance takes over again in highly complex stage patterns in
which the movement is set into a now contrasting, now synchronized
clapping and finger snapping. Sometimes each dancer riffs indepen¬
dently, then the individuals coalesce into groups, the whole stage con¬
stantly on the move. Suddenly everyone leaves, and one boy remain*;.
His short solo features big jumps and liquid transitions that shift to a
series of kneeling poses. This dance segues into more ensemble move¬
ment, and this, in turn, is followed by a major duet. Here a boy and
girl approach each other from across the stage in undulating steps.
Their dance is like a deep, fast-moving river. Her movement is based on
high extensions that interact in various ways with her partner’s body.
Throughout the dance there is a sense of tension, although the tem¬
perature remains cool. At the end, the other dancers move around them
as the girl lies spread-eagled on the floor with the boy on top of her. The
crowd moves back, and he rises, slapping her hand away as she reaches
toward him.
A second couple enters, and they dance a tender duet that is highly
balletic in its many arabesques and lifts. They stop from time to time
and embrace. This dance leads to the last scene in which the second
couple is attacked by the crowd, the boy is beaten, and the girl assaulted
and taunted by the crowd. She runs and falls as lights flash and the
sound of clanging signals is heard. The boy lifts the girl and lays her in
the center of the stage as the other girls pose on their partners’ shoulders.
The Road of the Phoebe Snow is unusual both for how Beatty reconciled
thematic material with nonnarrative movement and how he used move¬
ment metaphorically. The work conveys some of Beatty’s own anger at
the discrimination he experienced as a student and dancer. This had not
stopped in Chicago. When he reached New York, George Balanchine
had been impressed with him in Cabin in the Sky, which Balanchine di¬
rected and which featured Dunham and her dancers. Balanchine gave
Beatty a scholarship to the School of American Ballet, but when he
appeared for class, Beatty said the receptionist refused to allow him to
participate (Beatty 1990).
Beatty often gave his dances a driving use of technique that both em¬
phasized the virtuosity of his dancers and at the same time spoke of his
own anger at rejection. Beatty, like Primus, used virtuosity as a means
of protest, but he went further than she did, not simply relating form to
themes he wished to convey but using virtuosity to accuse a white soci¬
ety of discrimination and oppression. Speaking of the dancers in Phoebe
i6i A Game for Dancers

Enid Britten and Ulysses Dove as the central couple in an Alvin Alley
American Dance Theater production from 1970 of Talley Beatty’s The
Road of the Phoebe Snotv (1959). Photo: Alaji Bergma?i. Courtesy of Alvin Alley Dance
Foundation Archives.

Snow, he commented: “In the opening dances of Phoebe, the dancers are
presenting themselves. They are saying look how beautiful I am, see
what I can do, do I get the job? You sure don’t. I was making a statement
that no matter how good you are you will be pushed back. You see that
in the ballet” (Beatty 1993). At the time, however, critics saw nothing of
the kind, although they recognized an anger and violence in the work.
Doris Hering, for example, praised it extravagantly and spent a good
African-American Vanguardism; 1950s 163

deal of her review speaking about the atmosphere it created: “And like
the endless chains of freight trains from which it presumably draws its
title, it is a dance with no special beginning, no defined ending. Just the
miraculous capacity to alter the landscape as it passes” (i960: 80-81).
Hering spoke, too, of the angry strut and reach of the men and of a
lonely duet that was jazzy yet somehow contained and classical. She
described the second duet as yearning and “tainted by the presence of
an intruder. Then danger, separation, the group furiously closing in on
them, and again the red and green lights.”
A few months later, P. W. Manchester wrote in Dance News:

If this [The Road of the Phoebe Snow] was exciting the first time (and it was)
how much more so on this occasion! It has been shortened and this pulls
the whole thing together. If we have thereby lost a little of the amazing
dancing of Tommy Johnson we have gained a clearer pattern of the whole
work which now has a rhythm driving clean through it from beginning
to end. This rhythm pulses, quickens, slows but never relaxes and the
spectators are caught up in it as surely as it has captured the dancers.
There is a wonderful continuous series of short entries with various
combinations of dancers sweeping in and out, and two miraculous dance
duets—Georgia Collins of the gorgeous legs with Ernest Parham, lan¬
guorous in their sensuality, and Candace Caldwell and Tommy Johnson,
tender in the midst of passion until the casual brutality of the world
breaks into their dream. (1960b: lo-i i; see also 1960a)

And Jill Johnston, writing in Dance Observer, commented: “The jazz


ballet, to a scintillating Duke Ellington-Billy Strayhorn score, with its
frenetic rhythms born of ruthless city jungles, has all the searing excite¬
ment which may have been lacking in other productions. Georgia
Collins and Ernest Parham’s duet was easily the highlight of the pro¬
gram, arousing spontaneous cheers which were very well deserved”
(i960: 89). Once again, race disappeared from discourse, absorbed into
a generalized account of urban poverty and youthful alienation. Unfor¬
tunately, the black press did not cover Beatty’s concerts during the
1940S and ’50s, so we have no record of what African-American critics
may have thought of the work.
Only John Martin mentioned race and only in passing. He also ex¬
plained at more length than most how Beatty had subsumed narrative
into abstraction in Phoebe Snow:
164 A Game for Dancers

To put a finger on any cut-and-dried program for the piece is difficult; its
program is basically the projection of moods and states of mind and the
formalized activities that grow out of them. In outline it seems to be vir¬
tually a suite of related dances, almost like some Negro folk “Sylphides,”
with its duets and trios and ensembles; until at the close we find ourselves
suddenly in the midst of a highly dramatic situation, with anger and fear
and violence in the air. Perhaps only then does it become clear that the
numbers that have gone before may not have been quite so abstract as
they have seemed to be. (1959: 18)

Martin also noted that to use ballet terms to describe the dances was not
inappropriate “for the work, whatever its underlying human compul¬
sions, ends by being essentially classic in form and texture” (ibid.).
Martin’s comment illustrates the extent to which opinion had changed
regarding hlack bodies and “alien” techniques. There appears to have
been no question in his mind that Beatty was a modern dance choreog¬
rapher or that ballet technique was open to him.
Martin was one of the few critics who continued to speak about
race through the postwar years. In his 1963 survey of high-art dance,
tnxixl&d John Martin'’s Book of the Dance, he devoted a section to “Negro
Dance” (in which he included Asadata Dafora, Dunham, Primus,
Beatty, McKayle, and Alley) and another to “Integrated” dance compa¬
nies, which were white companies that accepted African Americans. As
quixotic as these divisions seem today, they were a sign that race in the
New York dance world was not a settled matter.
In all then, the acceptance of African-American choreographers into
modern dance proved to be a two-edged sword. While it marked a new
level of equality, it did so at the risk of blunting black choreographers’
ability to use dance as a weapon of protest. It also pointed up the limi¬
tations of modernism to encompass difference. Like Jewish choreogra¬
phers, African Americans were considered modernist only to the degree
that they could be viewed as one with a dominant race and culture.
When this was not the case, African Americans were pushed into a cat¬
egory of ethnic dance, which both demoted them in the dance hierarchy
and once again positioned them as other. In addition, commentators’
perception of Africanist movement and virtuosity as thrilling rather
than dangerous raised the possibility of once more viewing black mod¬
ern dance simply as display. In reviews of The Road of the Phoebe Snow,
there was a sense of the sheer excitement of seeing highly skilled dancers
African-Anierican Vanguardism: 1950s 165

in motion. Although this did not keep most critics from also noticing
the anger and violence of the work, in the 1960s, black modern dance
would come to be marked by the degree to which observers reduced it
to beautiful bodies in virtuosic movement.^
The ways in which race. Cold War issues, and consensus ideology
met and were refracted through modern dance’s own needs and inter¬
ests during the 1950s give some idea of the complex relationship be¬
tween specific fields and the larger social field. African-American chore¬
ographers won a permanent place in a dominantly white form, but it
was at the price of having racial issues minimized or ignored. At the
same time, black choreographers’ positions remained to a degree am¬
bivalent, evident in the categorizing of some dancers and dances as
ethnic rather than modern.
7

OBJECTIVISM'S
CONSONANCE

I n November 1957, Dance Magazine initiated a series of articles aimed


at surveying the modern dance field. The reason for this unusual
attention, according to editor Lydia Joel, was that sweeping changes
over the last decade had brought modern dance to what she described
as “an uncertain point.” Modern dancers, she wrote, were no longer
interested in opposing ballet, no longer concerned with social con¬
sciousness, or with psychological introspection. Dance Magazine “with
its usual willingness to step into the hornet’s nest,” would try to discern
some order in the field through articles to be published over a number
of months under the general title, “Close-Up of Modern Dance Today”
(Joel 1957: 20).
The series, however, did not so much attempt to discern order as
impose it. Two of the six articles dealt with what Joel referred to as “the
extremists,” choreographers who demanded “the right of dancing to be
its own subject matter.” The first of these two articles included the writ¬
ten statements from a symposium that had been organized a few months
earlier by David Vaughan, entitled “Four Dancers Speak and Dance.”
The dancers were Merce Cunningham, Merle Marsicano, Katherine
Litz, and Shirley Broughton. The second article, written by Walter
Sorell, concerned Alwin Nikolais and his work at the Henry Street
Playhouse (Sorell 1958). Joel pointed out that these were not the only
choreographers who focused on movement rather than the communica¬
tion of emotion; there were a number of others, including Midi Garth,
Marie Marchowsky, Erick Hawkins, Paul Taylor, and James Waring.
She might have added Murray Louis, Phyllis Lamhut, and Gladys Bailtn
at the Henry Street Playhouse. With the exception of the Playhouse,
these dancers did not constitute a coherent group but rather were tied
only by their interest in making their subject the materials of dance.
Objectivism’s Consonance 167

The rest of the Dance Magazine series consisted of two articles on


pedagogy published in December 1957 and February 1958 and two
articles devoted to another symposium that included Doris Humphrey,
May O’Donnell, Sybil Shearer, Carmelita Maracci, and others, which
appeared in March and April 1958. Oddly, these were dancers who as a
whole produced the kinds of works Joel said were no longer of interest.
Dance Magazine was not alone in its declaration of a new vanguard.
In an article for Dance Obsei'-verm 1955, Jill Johnston had identified two
trends in modern dance, one expansionist and one rebellious, but had
not yet been able to articulate how the rebels were breaking with the
old guard (Johnston 1955: 101-102). Two years later she was more suc¬
cessful in another article written for Dance Obse}'ver (Johnston 1957a:
55-56). Here, she again described two directions, one that expanded on
the work of the dance pioneers and another that broke with it. Then she
went on to characterize the first group as using a dramatic-realistic ap¬
proach, exemplified by Lim6n, while the other was devoted to deperson¬
alization and experimentation, exemplified by Cimningham, Nikolais,
Shearer, and Litz. In still another article a few months later, Johnston
continued her thoughts on the vanguard by tackling the question of the
meaning of abstraction in dance (Johnston 1957b: 151-152). Abstrac¬
tion, she stated, meant simply the absence of dramatic subject matter:
“The basis of abstraction in dance is its proximity to or distance from
dramatic subject matter, the carrier of which is gesture. And gesture is
movement invested with meaning.. . . Representation in dance means
more or less fidelity to dramatic character and situation” (1957b: 151).
Although Johnston credited modern dance as a whole with making
dance an abstract art, her division between abstraction and drama was
one that objectivists commonly used to separate their work fi-om ex-
pressional dance.
Why, one might ask, did Dance Magazine and Dance Observer find it
advantageous to anoint objectivism as the new vanguard? What did ob¬
jectivism do for modern dance? To begin to address these questions, I
want to examine objectivism through the work of two of its most influ¬
ential exponents, Merce Cunningham and Alwin Nikolais.^ Cunning¬
ham and Nikolais were particularly important because they did not
simply formulate an approach to modern dance that was antimimetic
but articulated philosophies, embodied in their work, that put them in
opposition to the dominant social order. However, the resistant aspects
of their dance for the most part went unrecognized, with emphasis placed
i68 A Game for Dancers

instead on the dances’ lack of emotional basis, dehumanization, con¬


cern with form alone, and by extension their meaninglessness.

By 1945, Merce Cunningham (1919-) had been on the New York mod¬
ern dance scene for more than fifteen years (Cunningham 1985; Vaughan
1997a). Born in Centralia, Washington, Cunningham attended the
Cornish School (now the Cornish College of the Arts) in Seattle, where
he met John Cage, who was accompanying the dance classes. Bonnie
Bird, Cunningham’s teacher at Cornish, encouraged him to attend a
summer session with Martha Graham at Alills College in Oakland, Cal¬
ifornia. Impressed by his technical skill, Graham invited Cunningham
to New York, where he joined her company in 1939. There he quickly
earned a reputation as a charismatic performer. Cage arrived in New
York in 1942 and was soon urging the young dancer to strike out on his
own. Cunningham and Cage gave their first joint recital in 1944, and
Cunningham left the Graham Company permanently in 1945.
Over the next decade, Cunningham developed a dance that focused
on movement and dance structure rather than on the communication
of emotional essences. He has said that even during his years with Gra¬
ham, he was most interested in how people moved and how dances
could be assembled, and he worked long hours alone in the studio ad¬
dressing these concerns (Cunningham 1985: 39-43). To begin with,
Cunningham focused on dance technique, slowly building an approach
to movement in which he clearly had ballet in mind. While studying at
the School of American Ballet during his Graham years, he paid careful
attention to Balanchine’s attempts to give the dancing body a maximum
of visual legibility. Cunningham developed a technique that incorpo¬
rated a number of ballet steps as well as ballet turn-out (in addition to
parallel positions) and that stressed high extensions of the legs and a
comportment in which the weight was lifted off the legs to give the
body a sense of lightness. However, he sought to engage the torso in a
different way from ballet. He told Jacqueline Lesschaeve that in classi¬
cal ballet he had noticed that the dancers moved the shoulders {epaule-
ment) against the movement of the legs, but the torso did not move much.
Consequently when the legs moved with great speed, as Balanchine
demanded, the torso lost its clarity. Cunningham developed a more
flexible back and torso that would make the entire body more legible
when the legs moved with speed. “All of my work comes from the trunk.
Objectivism’s Consonance 169

Jo Anne Melsher, Merce Cunningham, Carolyn Brown (partially hidden), and Viola Farber
rehearsing Cunningham’s Septet (1953). Photo: George Moffett. Courtesy of Jerome Robbins Dance
Division, The New York Public Library for the Performmg Arts, Astor; Lenox and Tilden Foundations.

from the vv^aist, nearest the hip,” he said. “And you tilt it or you twist it
in every direction. It doesn’t come from the shoulders, but from much
farther down. Further than that, I relate it to or against the leg” (1985:
62). At the same time, though, Cunningham’s carefully constructed
technique included a relaxation in the shoulders and arms that gave it a
far more informal appearance than ballet. This sense of informality was
reinforced in choreography that incorporated walking, standing, and
other movement of everyday life.
Cunningham’s early technique can readily be seen in a film of Septet
(1953) made in 1964.^ The work, choreographed in seven sections for
six dancers, is set to Eric Satie’s Trois morceaux en forme de poire. Cun¬
ningham has said it was one of the last works he made using “a wholly
170 A Game for Dancers

intuitive procedure” (Vaughan 1997a: 76). In the sixth section (an ada¬
gio danced by Cunningham, Carolyn Brown, Viola Farber, and Barbara
Lloyd) the movement clearly shows Cunningham’s indebtedness to
ballet in the vocabulary of steps, which range from attitudes and
arabesques to developes, grands battements and plies. Important too is the
degree of turn-out the dancers use in addition to parallel positions, and
their use of weight, which is often held up off the legs. However, it is
also clear that Cunningham has freed the torso to a far greater degree
than occurs in ballet and, as mentioned, relaxes the shoulders and arms.
As the dancers do their slow developes and grands battements, their upper
bodies are canted in varied directions, and their arms do not appear to
take set positions. The back remains engaged and the hips stable no
matter how the torso bends. This stability is necessary when speed
is added to the dancing. Cunningham’s use of torso movement can be
contrasted with Balanchine’s in The Four Temperaments, where the trunk
is allowed to shift from back to front or side to side, but is not given
nearly the range that Cunningham gives it. In the parts of Septet that
include faster movement, such as Cunningham’s second-section solo and
the dance for two couples at the beginning of the fourth section, it is
possible to see the quick changes of direction Cunningham demanded
at speed and the consequent necessity of a stable back and hips. It
should be added that although lifting the weight off the legs also aids
speed, Cunningham’s technique does not call for abandoning a sense of
weight altogether. This is clearest when the dancers take lunges; as they
move forward or to the side, they emphatically push the weight of the
descending foot into the floor.
The technique Cunningham developed opposed expressional mod¬
ern dance in several important ways. First, older modern dance tech¬
niques stressed weight and the floor; Cunningham’s technique included
weight, but counterbalanced it with lightness. In addition, he aban¬
doned the floor exercises that were so much a part of older techniques
because, he said, humans mainly move while standing. He started his
classes with standing exercises that warmed up the back. Although he
retained the bare feet and flexible torso of traditional modern dance, as
noted, he added a great deal more speed. Most important, Cunningham
concerned himself with how movement worked and the shape it took
rather than with communicating meaning.
Although Cunningham’s dance was certainly rationally based, he de¬
veloped several strategies aimed at disrupting the cause-and-effect logic
Objectivism’s Consonance 171

that resulted in what he called the sending of social messages. The first
step was to treat dance as concrete, in the sense of being an activity. In
a defining essay entitled “The Impermanent Art” written in 1955, Cun¬
ningham said that a new dance existed that did not yet have a name but
did have ideas. “These ideas seem primarily concerned with something
being exactly what it is in its time and place, and not its having actual or
symbolic reference to other things. A thing is just that thing.” ([1955]
1978: 310). Dance should not attempt to represent something else,
but rather should be itself. “When I dance, it means: this is what I am
doing” (ibid.). This focus on the concreteness of dance “eliminates the
necessity to feel that the meaning of dancing lies in everything but the
dancing, and further ehmmates cause-and-effect worry as to what move¬
ment should follow what movement, frees one’s feelings about continu¬
ity, and makes it clear that each act of life can be its own history” ([1955]
1978: 310-311).
Dance was not about emoting, Cunningham said. Rather, “in its es¬
sence, in the nakedness of its energy it is a source from which passion or
anger may issue in a particular form” ([1955] 1978: 311). He added: “I am
no more philosophical than my legs, but from them I sense this fact: that
they are infused with energy that can be released in movement—that
the shape the movement takes is beyond the fathoming of my mind’s
analysis but clear to my eyes and rich to my imagination” (ibid.: 312). A
concrete dance, then, not only disrupted cause-and-effect meaning, it
also escaped reasoned analysis and instead found a reality born of prac¬
tice and bodily intelligence.^
Cunningham further allayed the temptation to impose meaning on
his works by divorcing the choreographic process from music. In col¬
laboration with Cage, Cunningham had by 1944 begun to base his
dances on a structure of time units that allowed music and dance to be
relatively independent of each other. This soon led to an arrangement
in which composer and choreographer worked independently. Music
and dance only came together at the moment of performance, where
they coexisted in the way that pedestrians and the sounds of traffic co¬
exist on city streets. Modern dancers had often separated music and
dance but had done so by subjugating music. Cunningham made music
and dance separate but equal. Cunningham also worked with designers in
a way that was similar to his working methods with composers, giving
them a minimum of information about a dance and then allowing them
to develop ideas as they saw fit. For example, in 1958 Cunningham
172 A Game for Dancers

wrote to Robert Rauschenberg about a new dance that Rauschenberg


was to design: “I have a feeling it’s like looking at part of an enormous
landscape and you can only see the action in this particular portion of
it” (Vaughan 1997a: no). For the dance, which was eventually titled
Summerspace (1958), Rauschenberg devised an abstract pointillist back¬
drop and costumes in Day-Glo colors.
Although all the processes mentioned here helped to disrupt mean¬
ing in Cunningham’s work, the most extreme element of anti-meaning
was his use of chance procedures. In the early 1950s, Cage introduced
chance into his music, and in 1951 Cunningham began to use it in his
dances, as well. At first this was done simply, by letting sequences of
movement or of individual dances be decided by a coin toss. Eventually,
however, it became a complex system in which Cunningham constructed
charts to indicate a variety of elements, including steps or movement
sequences and their order, space and direction, and the duration of
movement, then used chance procedures to determine how the various
elements would be combined. Once the selections were made, the dance
was set and rehearsed.
Suite by Chance (1953) with music for magnetic tape by Christian
Wolff, was among Cunningham’s early fully realized works using chance
techniques. In this piece he made up charts that took him a few hours a
day for several months to complete (Cunningham 1968: n.p.). The
movements he had given the dance, Cunningham has written, were as
flat and unadorned as he could make them (ibid.). Remy Charlip, one
of the Cunningham dancers who participated in the piece, wrote a sum¬
mary of the procedures Cunningham used for the work:

For this dance, a large series of charts was made: a chart numbering body
movements of various kinds (phrases and positions, in movement and in
stillness); a chart numbering lengths of time (so that a phrase or position
could be done in a long or short duration, or, in the case of the impos¬
sibility of lengthening the time of a movement, as for instance, a single
step, it could be repeated for the length of time given); a chart number¬
ing directions in space (floor plans).
These charts, which defined the physical limits within which the con¬
tinuity would take place, were not made by chance. But from them, with
a method similar to one used in a lottery, the actual continuity was found.
That is, a sequence of movements for a single dancer was determined by
means of chance from the numbered movements in the chart; space, di-
Objectivism’s Consonance 173

rection and lengths of time were found in the other charts. At important
structural points in the music, the number of dancers on stage, exits and
entrances, unison or individual movements of dancers were all decided
by tossing coins. In this way, a dancer may be standing still one moment,
leaping or spinning the next. There are familiar and unfamiliar move¬
ments, but what is continuously unfamiliar is the continuity, freed as it is
from usual cause and effect relations. Due to the chance metliod, some
of the movements listed in the charts were used more than once in dif¬
ferent space and directions and for different lengths of time, and, on the
other hand, many movements, to be found in the charts, do not appear at
all in the final choreography. (1954: 19)

Cunningham himself said of the dance:

It was almost impossible to see a movement in the modern dance during


that period not stiffened by literary or personal connection, and the
simple, direct and unconnected look of this dance (which some thought
abstract and dehumanized) disturbed. My own experience while work¬
ing with the dancers was how strongly it let the individual quality of each
of them appear, naked, powerful and unashamed. I feel this dance was
classical—precise and severe—however unfamiliar the continuity, how¬
ever imclassical the movements, in terms of tradition, and the stillnesses,
that is, held positions by the dancers, may have been. It was unprompted
by references other than to its own life. (Cunningham 1968: n.p.)

Both Cunningham and Charlip’s use of language in these passages


indicates the importance of certain ideas in Cunningham’s experiments.
Charlip commented on chance’s disruption of cause-and-effect relations.
Earlier in his article he mentioned that the surrealists and dadaists had
also experimented with chance procedures. However, he did not say
why they had done this, which was, according to Peter Burger, to at¬
tempt to regain some of the freedom denied by rationality: “Starting
from the experience that a society organized on the basis of a means-
ends rationality increasingly restricts the individual’s scope, the sur¬
realists attempt[ed] to discover elements of the unpredictable in daily
life” (Burger 1984: 65). For Cunningham, applying chance procedures
was a means of discovering the unpredictable in his own work. He has
said chance was a “mode of freeing my imagination from its own cliches”
([1955] 1978: 312). He noted that this method of countering the disposi-
174 A Game for Dancers

tions that sent him along familiar pathways produced movement that
was extremely difficult to perform (Cunningham 1985: 80; Vaughan
1997a: 59,62-63, 72)- "The customary logic of dance movement was con¬
fused, sometimes to the point of being physically limiting. Cunning¬
ham said that there were several dances he created that he was never
able to adequately perform. Nevertheless, it is important to note that
although the procedures Cunningham developed disrupted familiar
ways of structuring dances, they did not negate all rational processes or
personal choice. Rather, they enlarged the possibilities of sequencing,
shuffled continuity, interfered with norms. This in turn helped elimi¬
nate the imposition of narrative elements on the dance and gave it an
openness that led to greater freedom of interpretation.
Freedom of interpretation empowered the spectator at a time when
there was general social concern for decreasing individual freedom in
America. A body of writing ranging from the work of the Frankfurt
School to C. Wright Mills’s The Power Elite, from William H. Whyte
Jr.’s The Organization Man to David Riesman’s The Lonely Crowd, attests
to postwar anxiety over the individual’s loss of freedom in an increas¬
ingly bureaucratized society. Cunningham’s dance suggested that art
was no longer an authoritarian given. Instead, each individual was free
to find her or his own meaning in the dance.
If Cunningham’s use of chance can be linked to dada, Marilyn
Vaughan Drown (1997) has also connected his practices to Zen Bud¬
dhism. I would go further than Drown to argue that Zen permeated
Cunningham’s approach to his art, while at die same time being com¬
patible with certain modernist precepts. Zen allowed Cunningham to
offer, through his work, an alternative to Western, logically based sys¬
tems. John Cage first encountered Zen while at the Cornish School,
where he attended a lecture on Zen and dada by Nancy Wilson Ross
(Cage 1973: xi; Revill 1992: 107-125). In the late 1940s, he began to sit
in on lectures by D. J. Suzuki, who was teaching Zen philosophy at
Columbia University. He also became acquainted with Alan Watts, who
did much to popularize Zen in the English-speaking world through his
numerous books. Although Cunningham’s immersion in Zen may have
been more limited than Cage’s, he attended some of Suzuki’s lectures,
and he and Cage made a pilgrimage to visit Suzuki when the Cunning¬
ham company was inJapan in 1964 (Vaughan 1997a: 57,145). Cage also
undoubtedly introduced Zen ideas to Cunningham. Most important,
though, Cunningham’s working methods and his writings suggest that
Objectivism’s Consonance 05

he had found concepts in Zen that paralleled and perhaps enlarged upon
his own thinking.
Alan Watts wrote of Zen art:

The art forms which Zen has created are not symbolic in the same way
as other types of Buddhist art, or as is “religious” art as a whole. The fa¬
vorite subjects of Zen artists . . . are what we should call natural, con¬
crete, and secular things. . . . Furthermore, the arts of Zen are not merely
or primarily representational. ... Even in painting, the work of art is con¬
sidered not only as representing nature but as being itself a work of na¬
ture. For the very technique involves the art of ardessness, or what Sabro
Ffasegawa has called the “controlled accident.” (1957: 169)

As Drown notes, the controlled accident is an appropriate description


of the relationship of dance, music, and design in Cunningham’s work
as well as his use of chance procedures. These elements are highly in¬
dependent but come together in the same time and space. Within this
time and space of performance they may coincide at certain moments
in ways that reinforce each other, or that make unusual and interesting
juxtapositions. One can expand on Drown’s analysis to say also that
these disparate elements, paradoxically, accord with each other. This
is undoubtedly because the elements are not completely independent.
The composer has an idea, within a few minutes, of how long the dance
is to be and has to make his score flexible enough to accommodate the
dance. Cunningham also might tell his collaborators something about
his thinking for the piece, as he did with Rauschenberg in Summerspace.
Most important, Cunningham chose his collaborators carefully and
worked with them on a regular basis so that they were unlikely to pro¬
duce something out of keeping with his own artistic aims. Cage, of
course, helped form those aims and was Cunningham’s partner in life
as well as work. Earle Brown, who composed several of Cunningham’s
dances in the 1950s, was married to Carolyn Brown, Cunningham’s prin¬
cipal dancer, and Morton Feldman was a close collaborator of Cage’s.
Rauschenberg, who designed virtually all Cunningham’s sets and cos¬
tumes during the 1950s, was a friend as well as a vanguard artist. Cun¬
ningham’s circle of collaborators consisted of like-minded individuals
who worked and played together for years.
If the controlled accident describes Cunningham’s working relation¬
ship with his collaborators, it also describes his use of chance methods.
176 A Game for Dancers

Watts writes that “there is no duality, no conflict between the natural


element of chance and the human element of control. The constructive
powers of the human mind are no more artificial than the formative ac¬
tions of plants or bees, so that from the standpoint of Zen it is no con¬
tradiction to say that artistic technique is discipline in spontaneity and
spontaneity in discipline” (1957: 169). Cunningham’s use of chance
methods allowed him to retain the control of a virtuosic technique and
elements of organized movement, while achieving spontaneity through
tossing coins to govern space, direction, time, and other aspects of the
dance. Discipline is particularly important in Cunningham’s work, not
only the discipline it takes to acquire and perform his refined technique
but the discipline of long hours of making charts and tossing the coins
involved in the chance procedures of his dances. Yet it is these very things
that allowed him to attain the spontaneity he seeks.
There are other aspects of Zen that resonate with Cunningham’s
attitudes and his work. Watts speaks of Zen being located in the con¬
crete and of art not being artificial but a work of nature. Cunningham
was describing a Zen attitude when he wrote: “A thing is just that thing.
It is good that each thing be accorded this recognition and this love. Of
course, the world being what it is—or the way we are coming to under¬
stand it now—we know that each thing is also every other thing, either
acmally or potentially. So we don’t, it seems to me, have to worry our¬
selves about providing relationships and continuities and orders and
strucmres—they cannot be avoided. They are the nature of things”
([1955] 197^' 310)- Zen does not divide the world into binaries; rather,
all things are related. And since everything is related, humans are an
integral part of their environment. “Thus, our stark divisions of spirit
and nature, subject and object, good and evil, artists and medium are
quite foreign to this culture” (Watts 1957: 170). The concreteness that
Zen emphasizes means that much of Zen art is nonrepresentational.
Since the work of art is “that thing,” as Cunningham described it, it is
not necessary for it to represent something else.
The concreteness of Zen is closely related to direct action. Watts says
that Zen masters speak as little as possible about Zen because words are
abstractions, and they prefer to show its principles by the concreteness
of action (1957: 127). A story that Cunningham has related on many
occasions illustrates the point. During a class Cunningham was teach¬
ing, company member Marianne Preger asked how to do a certain move-
Objectivism’s Consonance 177

ment. Cunningham finally said, “Marianne, the only way to do it is to


do it” (1985: 48).
For Cunningham, dancing is instinctive from the standpoint that
it is not about ideas but about action. In 1955 he wrote: “If a dancer
dances—which is not the same as having theories about dancing or
wishing to dance or trying to dance or remembering in his body some¬
one else’s dance—but if a dancer dances, everything is there. The mean¬
ing is there, if that’s what you want” ([1955] 1978: 310). In speaking to
Jacqueline Lesschaeve, Cunningham noted that some dancers come
to study with him because they are attracted to the ideas of his dances,
particularly the use of chance, but because they are involved with intel¬
lectual processes, they cannot understand the act of dancing. “In most
people, there’s such a split between instinct and intellect. A technique
class should, in a way, within a certain scale, put them together so that
both are working in unison” (1985: 72-73).
In Zen, direct action leads to the possibility of freedom. But action is
not mere impetuosity; it is assumed the actor has spent long hours
practicing Zen principles. With this discipline one can act without the
danger of paralyzing abstractions that come with thought. Watts com¬
ments: “The marvel can only be described as the peculiar sensation of
freedom in action which arises when the world is no longer felt to be
some sort of obstacle standing over and against one” (1957: 132). Cun¬
ningham noted: “I find that it is the connection with the immediacy of
the action, the single instant, that gives the feeling of man’s freedom.
The body shooting into space is not an idea of man’s freedom, but is
the body shooting into space” ([1955] 1978: 311-312). Cunningham’s
writings suggest that he saw parallels between dance and Zen practices;
or perhaps, he considered dance itself a Zen practice, since, as Watts
notes, in Japan every profession and craft was at one time considered a
lay method of learning the principles embodied in Taoism, Zen, and
Confucianism (1957: 171).
There is one further aspect of action in Zen that is important in
Cunningham’s working methods, which is commitment or follow-
through—in other words, an action made unhesitatingly and whole¬
heartedly (Watts 1957: 148). In his 1955 essay, Cunningham mentioned
the necessity of this kind of commitment to the dance. “This is not
feeling about something, this is a whipping of the mind and body into
an action that is so intense, that for the brief moment involved, the mind
ij8 A Game for Dancers

and body are one. Xhe dancer knows how solidly he must be aware of
this centering when he dances. And it is just this very fusion at a white
heat that gives the look of objectivity and serenity that a fine dancer
has” (11955] 1978: 311).
Summerspace, which Cunningham choreographed in 1958, exempli¬
fies many of the points of contact with Zen and dada that have been
made here. When Cunningham created the work, he said he wanted it
to deal with space; “the principal momentum was a concern for steps
that carry one through a space, and not only into it” (1966: 52). He had
in mind the way birds alight and then move on as they pass in flight or
automobiles move relentlessly along freeways and cloverleaves. As a
starting point for the structure of Summerspace, Cunningham drew
twenty-one lines across the stage that corresponded to six wing spaces
for entrances and exits, devising a movement sequence for each line. Each
sequence was devoted to a certain kind of movement, such as turns,
runs, skips, or leaps. He then used chance procedures to determine the
direction each sequence would take, its speed, its height or depth (per¬
formed in the air or on the ground), how space was covered (for ex¬
ample in a straight line, diagonal or circle), number of dancers perform¬
ing an action, whether they did it together or separately, and whether
they ended the action on or off the stage. Again, the controlled acci¬
dent allowed Cunningham room to dictate the choice of steps and the
elements devoted to chance. The musical score was composed by Mor¬
ton Feldman, and the designs were, as mentioned earlier, the pointillist
backdrop and leotard costumes by Rauschenberg in bright, fresh colors.
In Summerspace the felicitous workings of the controlled accident
can be seen in the meeting of Feldman’s shimmering score, the dappled
set and costumes, and the movement, with its constant shifts of direction
and trajectories through the stage space. Whether one wishes to call it
a dada method or Zen, the cause-and-effect logic of the dance is broken
by the chance procedures, making constantly unexpected meetings and
juxtapositions both among the dancers’ groups and spacing and within
the movement sequences themselves. As for the dancing, a film exists of
excerpts from the first performances of Summerspace at the 1958 Amer¬
ican Dance Festival at Connecticut College."^ The film is without sound
and in black and white, but even this inadequate record cannot dimin¬
ish the impact of the dancers’ performance. Carolyn Brown, Viola Far-
ber, Remy Charlip, Cynthia Stone, Marilyn Wood, and Cunningham
move in unhesitating action, apparendy unblocked by any second guess-
Objectivism’s Consonance 179

Viola Farber and Carolyn Brown in Merce Cunningham’s Summerspace (1958). Photo: Richard
Rutledge. Courtesy of the Cunningham Dance Foundation.

ing about how they look or what they are doing. Their comportment is
upright but relaxed. They dance with unhurried clarity, each immersed
in the work in a way that looks free and unburdened; they are alert and
completely at home in their environment.^
Although Zen Buddhism and modernism may seem strange bed¬
fellows, they are compatible in two important ways: they both abjure
i8o A Game for Dancers

causality, and they both distrust language. Modernists sought to disrupt


ideas of causality as part of an antirationalization agenda, while Zen
practitioners view causality as the wrong-thinking that separates the
world into binaries. Zen holds to a relational view, or as Cunningham
put it, each thing as part of everything else. Above all, however, Zen
looks upon language as a rational system much as modernism does. In
An Introduction to Zen Buddhism ([1934] 1964) Daisetz Suzuki wrote:
“Zen is decidedly not a system founded upon logic and analysis. If
anything, it is the antipode to logic, by which I mean the dualistic mode
of thinking” (ibid.: 38). Words are words and nothing more, Suzuki de¬
clared: “Zen deals with facts and not with their logical, verbal, preju¬
diced, and lame representations” (ibid.: 61). “The reason why Zen is so
vehement in its attack on logic, and why the present work treats first of
the illogical aspect of Zen, is that logic has so pervasively entered into
life as to make most of us conclude that logic is hfe and without it life
has no significance” (ibid.: 63). On the contrary, Suzuki said, logic chains
the spirit. Logic must be conquered in order to see the whole of reality,
“which refuses to be tied up to names” (ibid.: 60). Dualities divide the
soul against itself; they cause discord, contradictions, and antagonism.
Zen attempts to make the whole of reality apparent.
This splitting of mind from the totality of mind-body causes other
problems. According to Watts, Zen teaches that there is no “myself’
apart from experience. It is an illusion for the mind to try to stand apart
from experience (1957: 121). Language aids this abstraction from real¬
ity. Zen is wary of language as a system of abstract signs that inhibits
uncontrived action: “spontaneity or naturalness {tzu-jan)... is the un¬
mistakable tone of sincerity marking the action which is not studied and
contrived” (ibid.: 13 3). In order to practice Zen, it is necessary to switch
from the symbolic to the concrete, from words to action. Watts speaks
about the “fatal confusion of fact and symbol” in splitting the mind
from itself. “To make an end of the illusion, the mind must stop trying
to act upon itself, upon its stream of experiences, from the standpoint
of the idea of itself which we call the ego” (ibid.). The Zen practitioner
understands “the importance of avoiding confusion between words and
signs, on the one hand, and the infinitely variable ‘unspeakable’ world
on the other” (1957: 130). Modernists, on their side, viewed language
as a system of signs that had been fully rationalized. Dada and surrealist
artists tried to disrupt the logic of language through strategies that in¬
cluded chance procedures, collage, and improvisation.
Objectivism’s Consonance i8i

Zen gave Cunningham an alternative to Western philosophies of du¬


ality, and chance procedures gave him a method for escaping the causal
logic of dance meaning. As such, these practices constituted a form of
resistance in his work. However, as noted earlier, critics did not recog¬
nize the critical aspect of Cunningham’s dances, even when Cunning¬
ham and his supporters (particularly Cage) spoke about them. Nor was
Zen little known at the time. Zen philosophy, calligraphy, and poetry
were all extremely popular in the early postwar years. The Beat poets
and artists were using Zen extensively in their work in the late 1940s
and ’50s, and Watts and Suzuki, among others, had written numerous
books and articles about Zen, which had become popular among a wide
public. However, commentators focused on what they saw as the lack of
emotion in Cunningham’s dances, their coldness, and their meaningless¬
ness, which set them in opposition to expressional modern dance.
A Dance Observer review by Nik Krevitsky in 1951 illustrates this
point. In it he struggled to come to grips with Cunningham’s first
chance piece. Sixteen Dances for Soloist and Company of Three. Krevitsky
noted that the work lacked continuity and meaning, but he didn’t ex¬
trapolate from this observation. Rather, he wondered why Cunningham
could not have given more hint of what the dance was about in the title,
instead of naming it so objectively and ciyptically. Even as a pure dance
piece, he said, “we find many of the sequences completely disturbing
from the standpoint of development or of mood” (1951a; 41). Krevit¬
sky did not suggest that the meaninglessness of Sixteen Dances was due
to the choreographer’s inexperience or ineptness; the absence of mean¬
ing appeared to him to be willful. Krevitsky twice spoke of the work’s
being aimed at a select few who shared a philosophy of art similar to
Cunningham’s. In emphasizing the choreographer’s lack of desire to
communicate and his concern with a small audience of peers, Krevitsky
placed Sixteen Dances in the realm of vanguard modernism, but his com¬
ments did not in any way reflect the view that the piece might have of¬
fered an alternative to the rationalized workings of Western society.
This was the general attitude of commentators, who focused on those
aspects of Cunningham’s work that tied it to a new objectivist vanguard
while ignoring or simply not seeing resistant qualities in it.

Alwin Nikolais (1910-1993), like Cunningham, saw his work accepted as


part of a new modernist vanguard while being denied a critical compo-
i82 A Game for Dancers

nent. Nikolais developed a dance that rejected narrative subject matter


and the communication of emotion. It also gave movement, sound,
and light an equal place on the stage. Nikolais formulated a dance the¬
ory rather than a technique, and working methods that gave improvisa¬
tion a central place. In addition he articulated an overarching philoso¬
phy, embodied in his dance, that stressed human beings’ place in a
larger universe. Nikolais wished to show, in his words, “man being a fel¬
low traveler within the total universal mechanism rather than the god
from which all things flowed” (1971a: ii). In 1958, he wrote an essay
entitled “The New Dimension of Dance” that explained his principles.*^
Speaking of new directions in the arts he wrote: “The arts, we find, are
now becoming vitally concerned with the direct and poignant transla¬
tion of those abstract elements which characterize and underline an art
subject” (1958: 43). The new modern dance was no longer concerned
with placing the individual at the center of attention, of probing psycho¬
logical and emotional essences. “Man now realizes his presence within
a universe rather than a world.” Nikolais went on to say, “the major
contemporary significance of this is the greater freedom from the lit¬
eral and peripheral self of man. This freedom, indeed, is one of the
striking and appealing characteristics of the new art” (ibid.). What it left
the artist free to do was to explore more fundamental energies than the
human psyche.
While acknowledging the importance of the Bennington pioneers
in exploring the basis of human emotions, Nikolais asserted that mod¬
ern dancers had nonetheless appeared as stylized characters through
which a psychological drama transpired. “The character still remained
literally and dominantly present.” Like other objectivists, Nikolais saw
in expressional modern dance representation rather than embodiment.
Equally important in his argument was the fact that the dance charac¬
ter dwarfed its surroundings. In the new dance,

the character or dance hero is no longer dominant. The new dance fig¬
ure is significant more in its instrumental sensitivity and capacity to
speak directly in terms of motion, shape, time and space. ... It is the
poetry of these elements speaking directly out of themselves and their
interrelationships rather than through a dominant dance character or
figure. The dance figure may often be present, but if it is, it is usually in
equilibrium with the aggregate of all elements in operation rather than
dominant or emotionally intrusive upon it. (1958: 44)
Objectivism’s Consonance 183

Nikolais said that subject matter in the new dance was often non¬
objective; consequently movement had less dramatic tension and dis¬
sonance now that “peripheral emotion” had been eliminated. The new
dance might, for example, include long periods of pause, comparable to
silence in music. Peripheral emotions such as sorrow, joy, and rage were
exchanged for primary emotions that were abstract feelings of heavy,
light, thick, thin, large, small, fast, slow, and so forth. These were more
basic than peripheral emotions, in fact were the ingredients of them.
The use of these primary emotive factors was to make humans “more
congenially aligned and relative to a greater natural orbit rather than an
unbalanced human island” (ibid.). In this way individuals could under¬
stand that they were part of an integrated totality rather than feeling
isolated and alone.
The performer of this new dance did not seek to impose his will on
it but left ego behind, making himself “pliantly available to his impul¬
sions” (1958: 45). “He stays totally engrossed in the motional content,
without personal emotional point of view towards the action, but, rather,
sentiendy and kinetically involved in its unfolding. In this way the mo¬
tion speaks—directly and clearly of itself—to both dancer and beholder.
Its emotional content is born out of itself, rather than from any exter¬
nal emotive factors grafted upon it” (1958: 46).
As for spectatorship, Nikolais conceded that the new dance he de¬
scribed might initially be difficult for viewers to understand, because it
did not conform to familiar practices. However, such a dance opened
the way to new freedom by “allowing the audience esthetic generation
of broader scope and personal associative experience” and by “plunging
the onlooker into more deep-seated visions and connotations” (1958:45).
Clearly, many of his ideas put Nikolais in opposition to expressional
modern dance, but if there were any doubt, he made it explicit when he
said: “New Modern Dance distinguishes itself from the Bennington
period in all areas. Subject, method, composition, technique, titles, cos¬
tume, accompaniment, all differ. We find the dancer, like the painter,
discovering directness of vocabulary immediately through the sentient
values of his media” (1958: 43-44). In opposing expressional modern
dance, Nikolais was not just engaging in modernist position taking,
trading one set of values for another in order to differentiate himself
from the past. Nor did he simply believe that psychological dance had
run its course. Nikolais viewed the basic assumptions of expressional
modern dance, and by extension of society at large, as wrong. Humans
184 A Game for Dancers

were not the center of things; they were not meant to dominate their
environment and each other but to exist harmoniously as one of many.
This was not an idea that fit comfortably with prevailing Western ide¬
ologies, based as they were on technological progress and the conquest
of nature.
World War II played a large role in the development of Nikolais’s
ideas. Murray Louis has commented that any belief Nikolais may have
had about the omnipotence of man faded after serving on the European
front in World War II.^ Claudia Gitelman, quoting Nikolais, writes
further that his “turn away from personal and emotive art was sealed by
the ‘apocalyptic explosion [of the atom] bringing awareness of invisible
realities of namre’” (Gitelman 2000: 204). And Susan Buirge, who danced
with Nikolais in the 1960s, also noted that Nikolais spoke frequently of
the impossibility of dance ever being the same after the explosion of the
atomic bomb.^ For him, the days were gone when a modern dance
could be content with representing what amounted to emotional prob¬
lems; a radical rethinking of art was in order.
Nikolais did not arrive at his concept of a new modern dance over¬
night. It was part of a process that percolated over a decade of work, be¬
ginning after the war. Nikolais came to dance through music (Louis i960,
1980; Grauert 1999). Born in Southington, Connecticut, of German
and Russian immigrant parents, he studied piano as a child. At sixteen
he became a pianist for silent movies, then eventually worked with sev¬
eral acting companies, doing a variety of jobs. He then moved to Hart¬
ford where he took over the directorship of the Hartford Marionette
Theatre. His dance studies started after seeing a performance by Mary
Wigman. It was her percussion orchestra that interested him, but when
Nikolais discovered Truda Kaschmann, a Hartford dancer who had
smdied with Wigman, he agreed to take dance classes as a means of
learning more about the percussion instruments he had heard. In 1938,
Nikolais started spending the summers at Bennington, working with
Humphrey, Weidman, Graham, Horst, and Holm. He then opened a
studio of his own in Hartford. His dance activities ended abruptly with
the war. When Nikolais returned from the army in the mid-1940s,
he settled on Holm as his principal teacher, and she soon asked him to
serve as her assistant. In 1949, Nikolais was offered the directorship of
the dance department at the Henry Street Playhouse on the Lower East
Side, which he accepted.
Nikolais’s move to the Playhouse was crucial to the evolution of his
Objectivism’s Consonance 185

.\]win Nikolais speaking to students at the Heniy Street Playhouse in the early 1950s.
Photo: Gene Dauber. Courtesy of the Nikolais/Louis Foundation for Dance.

ideas. There he was able to experiment with light and sound while de¬
veloping a school and a system of instruction with students who became
his company. His advanced class consisted of a handful of pupils, in¬
cluding Phyllis Lamhut and Gladys Bailin. Murray Louis soon joined
them after leaving the navy. Nikolais also had an assistant, Ruth Grauert,
who had worked with him in Hartford and was key to helping him re¬
alize his technical ideas as well as in aiding him with a number of orga¬
nizational tasks. The dancers of the soon-to-be-created Henry Street
Playhouse Dance Company were extremely young; Louis was the old¬
est at twenty-two, and Lamhut was still in high school. Nikolais was
teaching them how to dance as he was developing his approach to move¬
ment and choreography. But although inexperienced, the youngsters
were energetic and eager to experiment. Nikolais’s approach aimed at
training creative artists and not just dancers, so he required students to
attend technique, theory, composition, percussion, and notation classes.
He routinely shared concert programs with his company members, and
they also had the opportunity to show their work in their own concerts.
i86 A Game for Dancers

Nikolais stopped dancing soon after arriving at Henry Street. He had


come to dance late and said that he never felt comfortable as a dancer,
preferring choreography.^ Consequently, unlike most modern dancers
his work did not revolve around himself or the concept of a star per¬
former. This disposal of a central dancer-creator, reflected even in the
name of his company, reinforced his philosophical ideas, since it en¬
abled him to deconstruct the notion of the heroic individual, built into
the very structure of modern dance companies and works.
Nikolais’s working methods grew out of the heritage of Holm, Wig-
man, and Laban, to which he added his own ideas in order to attain the
direcmess, immediacy, and clarity of movement he wanted. His aim was
not to teach steps to be learned as a factory worker might learn the
motions of an assigned task. It was rather to teach a theory of motion
that would enable his dancers to help produce his vision on stage. He
therefore did not develop a dance technique, per se. In what was called
the technique class he gave exercises for general strengthening and at
the same time attempted to eliminate the personal tensions, rigidities,
and affectations in the bodies of his students that he felt impeded them
from achieving the kind of direct movement he sought.^® Classes were
usually arranged around a central theme: for example, rotations, direc¬
tional change, shifts of levels, or impulses in particular parts of the body.
The technique class paved the way for the real work, which took
place in the theory class that followed, where experimentation through
improvisation was a key element. Nikolais’s theory was based on a num¬
ber of premises, among the most important being “decentralization.”
Decentralization demanded a fluid center and flexible placement, con¬
trasting with the single, unvarying center of ballet and most modern
dance. It also called for the dancer to make the self available to the
movement rather than imposing the ego upon it. With a fluid center,
the dancer learned to quickly shift his or her center of movement and
balance while remaining in a relaxed body posture. According to Mur¬
ray Louis, “moving the center to any part of the body necessitated an
unusually quick and direct thinking. These shifts prevented the energy
from becoming rooted. They also brought into prominence parts of the
body (chest, hips, back) other than the extremities. The resulting move¬
ment seemed unpredictably and rhythmically complex” (1980: 138).
Decentralization is made particularly clear in Nikolais’s solo for
Louis, entitled “Fixation,” {torn Allegory (i959).ii The dance is set in
a highly constricted boxlike space in which the dancer can only move a
Objectivism’s Consonance 187

few feet in any direction. It consists of sequences of repeated, often


subtle isolations of the body. The first sequence begins with Louis,
turned slightly to the side, articulating breathlike movements that start
in the pelvis and move up the body to his torso, chest, and shoulders.
Then he executes a large, sweeping kick as his arms come over his head,
echoing the circular motion of the leg. Next, he resumes a straight
standing position, shakes his arms and hands rapidly in front of him,
and executes a hig hent-legged jump. He repeats this sequence several
times, all at rapid speed. He then moves on to sequences that explore
different parts of the body in jumps, turns, and crouches. Now the knees
receive attention, now the head, now the chest. In all of this Louis’s
balance and weight shift constantly. The dance has a precise, at times
mechanical, yet highly energized look about it, reinforced by a percus¬
sive score. The movement is lightning fast, the spine whipping, arms
darting, yet the individual isolations of movement are always lucid, and
the body posture never looks tense, giving the impression that Louis is
free to move in any direction. The qualities elicited by decentralization
become apparent in the flexibility and clarity of movement Louis
achieves within the extreme spatial limitations of the dance.
Nikolais’s theory sessions explored the theme of that day’s technique
class. For example, if the theme had been rotations of the body, the
dancers would then improvise accordingly. This was often done in
small groups that would build responsiveness and cooperation within
the group. Louis noted that the improvisation in theory sessions was
controlled, from the standpoint that if the dancers did not improvise ac¬
cording to specific theoretical concepts, it implied they did not under¬
stand the concept, and they were then corrected.
Nikolais taught improvisation to his dancers as a tool for choreog¬
raphy, since he expected them to create dances. He also asked his com¬
pany to use improvisational methods when he was producing his own
choreography. Phyllis Lamhut recalled that Nikolais often brought a
prop or image—sometimes a painting or sketch—to a rehearsal, and they
would begin to experiment from there.She used as an example the
dance “Discs,” from Kaleidoscope. Nikolais, she said, brought metal discs
about sixteen inches in diameter to the studio, asked the dancers to
strap them on their feet and explore what they could do. After they
experimented with a great deal of movement, Nikolais chose what he
thought might work best, added his own ideas, sent the dancers back for
more exploration, and eventually set the resulting dance. This was his
i88 A Game for Dancers

usual working process. His dances were created collaboratively within


the parameters of his dance theory. However, he always had the final
word on all elements.
Although Nikolais’s theory constituted a rational system, its overall
aim was to unfix habits and rigidities rather than impose them. In par¬
ticular, his use of improvisation was meant to give direct access to the
motional content that lay beneath emotional representation. Once again
it was the hidden truth of movement (or motion, as Nikolais preferred
to call it) that was sought, and it was located through a method that
circumvented conscious processes. In Nikolais’s words, “the subject is
guided by felt judgment rather than by objective or cerebrally domi¬
nated choice” (1958: 44). Murray Louis has noted that improvisation
demanded instant choreography and performance. Action occurred too
quickly for conscious thought to be brought into play, which in turn
created flexibility and enhanced bodily intelligence or “felt-judgment.”
Nikolais’s use of improvisation can be traced through his German dance
heritage, that is, through Holm to Wigman and Laban. And here there
are links to the Zurich dada group, which included several of Laban’s
students, among them Sophie Taeuber, Suzanne Perrottet, and Claire
Walther (Richter 1965: 31, 45, 69-70, 77, 79; Prevots 1985; Manning
1993: 68-69). Like chance procedures and free association, improvisa¬
tion played a role in what Hans Richter called the dadaists’ “conscious
break with rationality” (Richter 1965: 57). Nikolais also used a break
with rationality to gain access to what he believed was direct movement,
free from the taint of representation.
Nikolais employed sound and light to aid in the task of making his
philosophy visible on the stage. He had been attracted to dance through
percussion, and he used it in combination with the piano in his teach¬
ing. However, he did not begin his choreographic career using his own
scores. According to Louis, Nikolais gradually became frustrated with
trying to obtain appropriate musical compositions by others and conse¬
quently began to create his own. He started with scores for percussion
and piano then, with the advent of the tape recorder, he was able to pro¬
duce musiqiie concrete and finally electronic compositions (he is credited
as the first owner of a Moog synthesizer). These scores, Nikolais said,
emphasized a nonliteral response in the audience and encouraged “more
direct sense appeal” (1958: 45). Normally Nikolais created his chore¬
ography, the dancers moving in silence or to percussion, then he com¬
posed a score for it. In this sense, the sound and dance had closer links
Objectivism’s Consonance 189

than they did in Cunningham’s work. Early on, his scores also tended to
have a strong pulse, but gradually he deemphasized this element, mak¬
ing the sound less conspicuously supportive, then independent.
Nikolais not only discouraged ties to literal meaning through his
movement processes and sound scores, he further dissolved these con¬
nections through his use of light and color. In his early work Nikolais
disrupted the traditional hierarchy of the stage space, which focuses
attention on the center and which is normally supported by lighting.
His stage was an open area of abstract color and light that humans and
props might inhabit but, time to his philosophy, did not dominate. He
achieved his effects by placing lights at many unusual angles, including
low in the wings, which streaked the stage with light. He also used gels
in a variety of colors. Later, Nikolais would create fantastic lighting
designs with projections of great sophistication—in the early days the
effects he created were far more rudimentary, altliough no less experi¬
mental for their time. Ruth Grauert recalled that he would suggest a
new design he wanted to achieve, and Richie Brown, the technical di¬
rector at the Playhouse, would go to the cellar and improvise lights
from whatever he could find. Murray Louis noted that sometimes the
dancers operated lights when they went offstage to help create a desired
effect.
A glance at Nikolais’s early programs at Henry Street demonstrates
how he was developing his dancers and his dance in the direction his
mature work would take. A program in May 1950 included three Tech¬
nical Etudes (“Walking,” “Curve Study,” and “Fall Study”) as well as
Etudes Composed by the Dancers. The studies may have been extensions of
his technique and theory classes, indicating the point his young dancers
had reached in their training. But they also suggest the experimental
nature of his work at that time as Nikolais sought to realize his ideas in
motion. The fact that he produced a program that featured both his
dances and that of his students was also typical. As noted, he encour¬
aged his dancers to experiment along with him.
Noumenon is usually considered the first of Nikolais’s new dance works.
It was premiered on 9 November 1951 and performed by Beverly
Schmidt and Dorothy Vislocky. The dancers were completely covered
in stretch fabric. They used stools as props to sit and stand on, which
enabled them to enlarge their range of movement and allowed them to
take ever-changing shapes. In this case, Nikolais chose to reconfigure
the human body through costumes, but although he always considered
190 A Game for Dancers

such changes in body shape an option, he as frequently revealed the


body as concealed it. One of the costumes he favored most was based
on a unitard or leotard and tights, which took advantage of new tech¬
nologies in stretchable synthetic fabrics. The use of such costuming in
performance was unusual, but not singular. In the early 1950s, Balan¬
chine began to dress his dancers in practice clothes for some of his works,
and Cunningham, too, used leotards in most of his dances. However,
these skinlike costumes were a mark of difference in objectivist dance.
The usual costumes for modern dancers were skirts for women and
trousers and shirts for men. Leotards and tights put the emphasis on
movement and, since they were practice clothes, on the work of dance,
rather than on representational elements or thematic or narrative sub¬
ject matter. They also made it possible to see movement more clearly,
particularly small steps and gestures that previously would have been
hidden in less closely fitting garments.
On 26 January 1953 Nikolais introduced Etudes II: Masks, Props, and
Mobiles, in which he continued to develop his ideas. The title did not
indicate a specific work; it was simply a general description of inde¬
pendent dances by Nikolais and company members. In this case it in¬
cluded Aqueouscape and Noumenon, now called Noumenom Mobilis, by
Nikolais. There were also dances by the company, among them Enclave
and Harlequinade by Gladys Bailin, Annoyous Insectator by Phyllis Lam-
hut, 2ind Antechamber hyMntrzy Louis. On 10 December 1955 Masks—
Props—Mobiles made another appearance, this time without “Etudes
II” attached to the title and with slightly different punctuation. It also
carried a declaration of intent: “An experimental program in which the
dancers are depersonalized or in which their motions are extended
into external materials.”^® The dances included Noumenom. Mobilis, Web,
Aqueouscape, and Tournament by Nikolais, along with White Figure and
Red Hoop by Beverly Schmidt, Belonging to the Night and Polychrome by
Murray Louis, and Paraphrenalia by Dorothy Vislocky. Nikolais often
changed or developed dances and ideas from concert to concert, which
is evident here. However, Masks—Props—Mobiles shows a gradual de¬
velopment of his philosophy and his dance. The 1953 incarnation was
simply a group of dances with little visible connection, whereas the 1955
version had a clear objective and was more integrated in conception
while still comprising dances by both Nikolais and the company. Niko-
lais’s notion of depersonalization and of extending motion into external
materials would receive significant amplification the following year in
From left, Murray Louis, Gladys Bailin, and Coral Martindale in “Straps” from Alwin
Nikolais’s Kaleidoscope (1956). Photo: David Berlin. Courtesy of the Nikolais/Lotiis Fotmdatimi for Dance.

Kaleidoscope, his first major experimental work. Masks—Props—Mobiles


also included Tournament, which would become “Capes” in Kaleidoscope.
Kaleidoscope was a breakthrough for Nikolais, as the company was
invited to perform it in August 1956 at the ninth American Dance Fes¬
tival in New London, Connecticut. The American Dance Festival had
been started in 1948 as a showcase for modern dance. The Jose Limon
Dance Company was in residence each year along with guest artists.
The latter were usually drawn from the modern dance mainstream, al¬
though Cunningham had appeared in 1950. Nikolais was the first ob-
jectivist since that time to be invited.
Kaleidoscope consisted of eight dances, all by Nikolais, that were re¬
lated to each other primarily in their use of props. Otherwise each
dance was conceived as an independent entity. Nikolais considered this
kind of arrangement to be a collage of unrelated material whose jux¬
taposition nonetheless produced a totality (1961b; 31). The dances in
192 A Game for Dancers

Kaleidoscope consisted of “Discs,” “Pole,” “Box” (soon to be replaced by


“Paddles”), “Skirts,” “Bird,” “Hoop,” “Straps,” and “Capes.” Kaleido¬
scope featured Nikolais’s lighting, but he had not yet reached the point of
composing the entire sound score himself. Compositions by John Cage,
Edgard Varese, and Carlos Chavez were included, in addition to Niko¬
lais’s own work. The dancers wore unitards that were painted in varie¬
gated colors extending over the face, hands, and feet. They also wore
headpieces that curved upward in a tail behind the head. The effect of
the costumes and makeup was to integrate the group and depersonalize
it. However, although the costumes depersonalized the dancers in the
sense of muting their individuality, they did not disguise sexual differ¬
ences. These were made amply visible by the body-hugging unitards. I
would suggest that, in keeping with his philosophical aims, Nikolais was
not so much interested in suppressing sexuality as in emphasizing the
dancers’ relationship to a larger whole. Nikolais made his position clear
when he told an interviewer that his intention was not to erase sex from
his work. He said he wished to put sex in a neutral position instead of
making it a center of focus, as expressional dancers had done (Nikolais
1973-1974: 41).
In Kaleidoscope the dancers used props to extend or add elements to
their movement.For example in “Discs,” referred to earlier, the seven
dancers wore a metal disc attached to one foot. The dancers employed
the discs in a variety of ways, sometimes even balancing on the edges.
But they also used the discs as percussive instruments to punctuate their
movement and to act in concert with the percussion score composed by
Nikolais. In “Pole,” a duet for Murray Louis and Gladys Bailin set to
Japanese koto music, the two dancers balanced a pole between them,
moving it from shoulders, to arms, to feet in numerous ingenious ways
while rarely ceasing their slow, dreamlike dance. Although one of them
would occasionally take over the pole while the other moved out
independently, for the most part the pole acted as an object that bound
them together and which they manipulated with graceful cooperation.
“Straps” had three dancers with loops of elastic around their waists that
extended into the wings where they were attached to supports. The
dancers were able to produce movements, such as slow falls and can¬
tilevered stretches, that would have been impossible had they been un¬
aided by the straps. In these various ways, Nikolais enlarged the move¬
ment of the dance beyond the individual body to create what he called
an “abstract protagonist” (1961b: 31).
Objectivism’s Consonance 193

From left, Murray Louis, Phyllis Lamhut, Gladys Bailin, Dorothy Vislocky, Beverly
Schmidt, Bill Frank, and Coral Martindale in “Discs” from Alwin Nikolais’s Kaleidoscope
(195 6). Photo: David Berlin. Courtesy of the Nikolais/Louis Foundation for Dance.

However, this protagonist was not reduced to an abstract shape; the


human body was everywhere in evidence. In “Pole,” for example, Bailin
and Louis were perfectly recognizable as a man and woman, but their
movement vocabulary gave none of the usual cues that in expressional
modern dance would have conveyed ideas of a gendered relationship
between them. This absence allowed attention to be focused on their
general “humanness.” What was emphasized was the balance the dancers
attained between themselves and the object they held, and the relation¬
ship of the shapes they made in collaboration with that object to the
surrounding space. Throughout the work Nikolais stressed the notion
of human integration with the surrounding environment and how that
integration could enlarge human capabilities. Kaleidoscope reinforced
Nikolais’s contention that new concepts of motion, shape, time, light,
space, color, and sound would not dehumanize “man,” but rather, ac¬
cording to Nikolais, “en-humanize” him. “His power to ‘identify-with’
is his most vital human facility and to deny it would deny him his great¬
est scope of love, the essence of art and life itself” (Nikolais 1957: 31).
Kaleidoscope was an early work and was still crude in certain technical
194 A Game for Dancers

aspects and in the simplicity of much of its dance vocabulary and struc¬
ture. Nikolais also did not altogether give up narrative. For example,
Kaleidoscope contained two dances that were thematic. “Skirts” was a
quixotic comment on fashion, while “Bird” depicted two little girls who,
while playing, find a bird. They try to capture it and in the process kill
it. With the unconscious cruelty of children, they then resume their
play. Despite these lapses from his stated aims, there is no doubt that
Kaleidoscope moved Nikolais closer to realizing his philosophy through
a unified theater. This was apparent not only in his more accomplished
melding of sound, light, and movement, but in the work’s greater unity
of concept and in Nikolais’s increasing control of all the major elements
of the production.
Although critics recognized Nikolais’s work of the 1950s as exper¬
imental and opposed to expressional modern dance, they did not ac¬
knowledge his philosophical ideas. Instead, they spoke of dehumaniza¬
tion and lack of emotion. Some reviews were positive, others negative.
In negative reviews critics found fault both with Nikolais’s objectivist
aesthetic and with the fact that he gave equal weight to light, sound, and
movement. It should be noted though, that as much as critics com¬
mented on the experimental nature of his work and complained about
the equality of elements in it, they accepted it as modern dance. Niko¬
lais’s years with Holm and the many aspects of his dance that were based
in her teaching assured him of a place in the modern dance fold. No one
accused Nikolais of breaking with expressional models so completely
that he had created a new genre, as early modern dancers had done in
their rejection of ballet.
Even when reviews of Nikolais’s work were positive, they were much
the same as negative ones in their focus on aesthetic elements. Doris
Rudko of Dance Observer described Masks—Props—Mobiles as “a re¬
freshing, highly theatrical evening of dehumanized dance” (1956: 28),
while Walter Sorell wrote of Nikolais in Dajice Magazine:

It seems that, in his opinion, die expressiveness of the modern dance has
gone too much in the direction of soul-searching and caught itself in die
complex snares of psychological problems. He offers no problem themes.
He dehumanizes his dancers and makes them part of the external mate¬
rial he chooses for the development of an idea. His ideas are those of a
sculptor or painter, not primarily of a dancer. And he achieves stunning
images and unconventional designs with the help of props, masks, cos-
Objectivism’s Consonance 195

tumes and lighting. He seems to visualize an image, conceive and circum¬


scribe a basic design for it. Then he proceeds painstakingly to exhaust its
movement possibilities. Though sometimes the impression is that he
uses movement sparingly, there is not one static moment. (1956: 70 )

In stressing experimentation, dehumanization, and lack of emotion in


Nikolais’s work, critics placed it in opposition to the old expressional
modern dance and in this way positioned it to assume the mantle of a
new vanguard. However, in failing to recognize its philosophical base,
they robbed the dance of its critical power. In this way, Nikolais’s and
Cunningham’s work met a similar fate.
In probing issues in these artists’ work, it is necessary to mention one
other point. It has been suggested that their attraction to “pure” dance
emanated from a desire to hide or encode or otherwise deal with their
homosexuality at a time of intense homophobia. Pure dance, so the ar¬
gument goes, allowed them to suppress romantic relationships in male-
female partnering while at the same time closeting their own sexual ori¬
entation by continuing to make dances in which couples were inevitably
paired as male and female (Burt 1995: 144; Kowal 1999: 179-180; Fos¬
ter 2001: 175-178; Manning 2004; 209-210). Although I don’t disagree
that homosexuality may well have played a role in Nikolais’s and Cun¬
ningham’s muting of relationships, I have tried to show that there were
more compelling reasons for their turning away from expressional mod¬
els. On the one hand, objectivism provided a means of fostering a new
vanguard, which was the path to success in the field; and on the other,
it opened ways of challenging means-ends logic and of finding alterna¬
tives to it, which were modernist imperatives. It should also be pointed
out that all objectivists, no matter what their gender or sexual orienta¬
tion, eschewed overt meaning in their work, usually with the aid of
depersonalization, while a number of homosexual choreographers con¬
tinued to embrace expressional modern dance, even in the homophobic
postwar years.

If commentators focused attention on those aspects of Cunningham’s


and Nikolais’s work that linked it formally to a new vanguard while ig¬
noring its critical aspects, the fact remains that the two choreographers
offered particularly compelling reasons for calling attention to aes¬
thetic elements in their work. The most obvious of these was provided
196 A Game for Dancers

by the dances themselves, in which the choreographers did not just


abandon themes and plot lines, but turned their backs on the need to
communicate emotion. In addition, both Cunningham and Nikolais
spoke out verbally and in writing against the old modern dance, placing
themselves in opposition to it. Sometimes this was done obliquely,
sometimes with a direct attack. In stating their views the two men not
only questioned an expressional aesthetic, they went further, couching
their opposition in gendered terms that also challenged women’s lead¬
ership of the field, a leadership which at that time was imbedded in ex¬
pressional dance. They did this by associating expressional modernism
with emotion (the realm of women) and objectivism with action (the
realm of men). Feminist scholars have frequently pointed out, as Simon
Williams and Gillian Bendelow note, that “emotions have tended to be
dismissed as private, ‘irrational,’ inner sensations that have been tied,
historically to women’s ‘dangerous desires’ and ‘hysterial bodies’” (Wil¬
liams and Bendelow 1998: 131; see also Sydie 1987: 3; Laqueur 1987:
1-41), while the sphere of action, according to Sherry Ortner (1974), is
seen as a male preserve that goes on outside the domestic sphere. Niko¬
lais alluded to this idea when he wrote that, “the previous period of
dance dealt with strong emotionalism and was dominandy in the hands
of the female. This new dance level of abstraction seems to find the male
choreographer in control” (1961a: 324; see also 1965: 65).
Nikolais sometimes compared the directions of the new dance as a
response to scientific discoveries, thus tying it, however vaguely, to sci¬
ence itself. Cunningham described his working methods in the language
of the skilled artisan. No longer did the choreographer have to agonize
through a painful process of creation. He could act decisively, without
indulging in overwrought feeling. Although both Cunningham and Gra¬
ham claimed their choreography was based in movement, Cunningham
explained dance movement in a practical fashion: “In my choreographic
work, the basis for the dances is movement, that is, the human body
moving in time-space” (1980: 52), while Graham explained dance move¬
ment in emotional terms: “Each art has an instrument and a medium.
The instrument of the dance is the human body; the medium is move¬
ment. The body has always been to me a thrilling wonder, a dynamo
of energy, exciting, courageous, powerful; a delicately balanced logic
and proportion” (Graham [1941] 1980: 44-45). Similarly, Cunningham
took a workmanlike approach in describing his choreographic method:
“So in starting to choreograph, I begin with movements, steps, if you
Objectivism’s Consonance 197

like, in working by myself or with the members of my company, and


from that, the dance continues” (Cunningham 1980: 52). Graham, on
the other hand, remarked; “Iff was working on a new piece I would fix
a red ribbon to the door, which signified that no one was to enter the
studio. You did not want to he invaded when you were in the holy of
holies, which it was when you were trying to create a new dance” (Gra¬
ham 1991: 136).
It would not do to push the argument of gendered opposition too far,
as both Nikolais and Cunningham viewed their dance as intuitive and
not intellectual. Also, expressional dancers, themselves, had tried to
masculinize modern dance. As Mark Franko has pointed out, Martha
Graham identified modern dance as a virile form in some of her state¬
ments in the 1930s (Franko 1995: 38-57), while after the war Jose Limon
called for a distinctly male “virile dance” to help save the human race
from atomic extinction (Limon 1951). However, Cunningham and Niko¬
lais in their use of a language of action as opposed to emotion employed
masculine cues specifically against expressional dance and as a means
of differentiating objectivism from it. At the same time, their attack,
however indirectly, also challenged women’s domination of the modern
dance genre. The generation of modern dancers that came of age in the
1950s was the first to include a large contingent of men, and a number
of them were attracted to a nonnarrative dance, including James War¬
ing, Paul Sanasardo, David Gordon, Murray Louis, and Paul Taylor.
Also notable was Erick Hawkins, whose early work was indebted to
Graham, but who adopted an objectivist approach in the 1950s.
If Cunningham and Nikolais differentiated themselves from expres¬
sional dance, they had supporters acting on their behalf who also
stressed differences between the new and old modern dance. Nikolais
had, in particular, Murray Louis, a highly articulate spokesman, as well
as a brilliant dancer. In 1956, Louis went to John Martin and convinced
him to pay Henry Street a visit. Martin came and wrote two Sunday ar¬
ticles on Nikolais and his work, a powerful endorsement (Martin 1956a,
1956b). These lengthy pieces were unusual in Martin’s newspaper work.
They came at a crucial time in Nikolais’s career and without doubt had
an impact on it. In particular Martin extolled Nikolais’s use of im¬
provisation and the fact that he did not teach a technique. These were
in line with the theoretical basis of German modern dance that Martin
had long admired. Although he commented that Nikolais and his
dancers were in love with movement for its own sake, he assumed that
198 A Game for Dancers

this would nonetheless lead to an authentic dance. Martin’s assessment


of Nikolais’s work indicates how far he had traveled from his radically
antiobjectivist views of the 1930s. At the same time, Martin focused on
the aesthetic rather than the philosophical aspects of Nikolais’s dance.
The next year, Nikolais himself was able to contribute a Sunday
piece to the New York Times, an unusual opportunity that may well have
been influenced by Martin’s support. In the article he attacked the old
modern dance as mired in sexual psychosis, a dance in which “relations
and styles were dominantly promiscuous and the world was only a gi¬
gantic brothel. To interpret it otherwise was to deny both humanism
and dance” (Nikolais 1957: 31).^° This characterization had some of the
outrageous aggressiveness of dada and surrealist manifestos. Nikolais
then went on to describe a new dance, essentially his own, that would
“create esthetic entities out of fresh concepts of motion, shape, time,
light, space, color and sound.” This was the kind of statement, appear¬
ing as it did in the pages of the Times, that was likely to capture the
attention of the dance community, and it undoubtedly did, as Nikolais
was one of the vanguardists anointed by Dance Magazine at the end of
the year. Louis was less radical in his writing, but he explained Niko¬
lais’s work in a number of articles over the ensuing years.
Cunningham also had others who helped promote the oppositional
aspects of his work. Denby was one of the few critics to write positively
of Merce Cunningham’s dance in the 1940s, recognizing in it a rejec¬
tion of literary elements that Denby appreciated. He asked only that Cun¬
ningham be bolder, a request that Cunningham soon obliged (Denby
[1944b] 1986: 207-208; [1945] 1986; 279-280). David Vaughan, a young
transplanted English choreographer, was an eloquent defender of ob¬
jectivism who in time became Cunningham’s archivist. He organized
the symposium that was recorded in the Dance Magazine series in 1957
and that included Cunningham. He also wrote an introduction to the
symposium that plainly set out the goals of objectivism. These were
stated in positive terms, but they nonetheless placed objectivism in op¬
position to expressional modern dance and as such positioned it as a
rival to the old vanguard. Vaughan was more direct in a letter to the ed¬
itor of Dance Magazine in July 1958 in which he complained of Doris
Hering’s enthusiastic response to Martha Graham’s season at the Adel-
phi Theatre. The details of Vaughan’s argument were a summary of
objectivist criticism of the old modern dance, now voiced in specifically
oppositional terms: he found the dance less important in Graham’s work
Objectivism’s Consonance 199

than the narrative; the need to communicate led only to incompre¬


hensible symbolism that dance was not equipped to convey; the atmos¬
phere of the dances was reminiscent of what “prevails in an analyst’s
consulting-room”; Graham’s dance was not classic, which presents the
individual in relation to others and to the universe, but romantic, which
is “concerned exclusively with the individual in relation to himself”
(1958: 26-27). short, Graham’s work was not concerned with dance,
but with storytelling and self-absorbed psychology.
While both Cunningham and Nikolais could muster support in the
dance field that helped put them in leading vanguard positions, Cun¬
ningham had certain advantages because he was part of a circle of artists
that extended far beyond the dance world and whose connections put
him in touch with the newest trends in music and art. Cage seemed to
know everyone of importance in the musical world who passed through
New York, and Robert Rauschenberg, although not yet famous, intro¬
duced Cunningham to important people in the art world. Cunningham’s
audience was made up to a large degree of these artists and musicians
so that his work was known and spoken of outside the confines of the
dance world (Cvmningham 1985: 45). This gave him wider exposure and
also kept him abreast of the latest developments in the world of van¬
guard art. Nikolais, on the other hand, had isolated himself at Henry
Street, far down on the Lower East Side. However, although cut off from
the center of vanguard activity, he had the advantage of Henry Street
itself, a theater, funding, and a stable environment. This was nearly un¬
heard of in the modern dance world where choreographers (including
Cunningham) struggled to give one concert a year with a few hours of
rehearsal in the theater in which they were to perform. Nor did this
advantage go unnoticed. Dance Observer, whose writers were increas¬
ingly critical of Nikolais’s work, referred in one review to the “clique”
at Henry Street (Telberg 1959: 73). Despite such carping, there is little
doubt that the stability of Henry Street contributed significantly to the
development of Nikolais’s work.
In all then, Nikolais and Cunningham were instrumental in estab¬
lishing a new vanguard and were supported in their quest by others in
the field. This support sometimes came from surprising sources, such
as John Martin in the case of Nikolais and to a lesser extent Denby in
the case of Cunningham. But whatever the source, the overriding ele¬
ment in both the defense of objectivism and its critique was the aesthetic
aspect of its rejection of expressional models. As much as Nikolais and
200 A Game for Dancers

Cunningham might have attempted to embody alternative visions of


society in their dances, it was not acknowledged.

The success of objectivism offers a telling example of how a field re¬


fracts general social conditions tlirough its own history and needs. Ob¬
jectivism solved a number of modern dance’s problems: First, it renewed
the genre without destroying it. Lincoln Kirstein had long argued
that because modern dance was a complete break from the past and was
dependent on personalities rather than an impersonal vocabulary, it
was not viable. Objectivism proved him wrong.^* It did so by retaining
enough of the forms and structures of expressional dance to still be rec¬
ognizable as modern dance. These elements included framing devices
that ranged from the ways in which concerts were structured to where
they took place. Objectivists still appeared in the theaters that modern
dancers had long inhabited and many also shared programs with expres¬
sional dancers, as Cunningham did with Erdman in the 1940s, Nikolais
with Nina Fonaroff and others in the early 1950s, and Merle Marsicano
with Mary Anthony in the mid-1950s. As pointed out, objectivists also
preserved parts of expressional vocabulary and technique. Cunningham
maintained the sense of weight and torso flexibility that marked modern
dance, while Nikolais carried forward many aspects of Holm’s approach
to movement, including an emphasis on improvisation. Both Cunning¬
ham and Nikolais also retained the use of bare feet, one of the most
visible and characteristic elements of expressional modern dance.
Yet while some aspects of expressional dance were preserved, others
were prominently rejected, the most crucial being the communication
of essentialized emotion, which was at the heart of expressional dance.
Objectivists succeeded in attacking this central tenet without destroy¬
ing modern dance as a whole by laying down their challenge in mod¬
ernist terms. That is to say, they focused attention on movement as
dance’s acknowledged, essential element. This pushed the genre away
from narrative and illustration toward which it had been drifting and
which threatened its modernist credentials. Objectivists made it clear
that the desire to communicate meaning, abstracted though it may have
been, was a liability for modern dance, and they submitted that what
they offered was closer to modernist goals.
The second problem to be solved by objectivism was modern dance’s
need to disarm rivals in order to maintain its vanguard position in the
Objectivism’s Consonance 201

field. Objectivism thwarted competition from both commercial media


and ballet by adopting extreme forms that neither was eager to co-opt.
Although ballet and Broadway choreographers might borrow move¬
ment and even thematic material from modern dance, they had no de¬
sire to adopt radical methods like improvisation and chance procedures
or to produce the kinds of works that resulted from such methods. Nor
did they wish to create a level of depersonalization that meant aban¬
doning all romantic relationships on the stage, or to make music and
dance completely independent of one another.
Finally, objectivism could be viewed as a new vanguard without
raising the specter of critique, which helped keep modern dance from
scrutiny by more powerful social forces and at the same time served to
acquiesce to those forces. Here one might ponder how successful ob¬
jectivism would have been if modern dance’s need for a new vanguard
had not intersected with Cold War constraints; in part, it may have
been the idea of secrecy engendered by the Cold War that channeled
practice and reception in an objectivist direction. For although abstrac¬
tion gave every spectator the right to create his or her own meaning, it
also hid meaning. This seemed to make resistance possible, but secrecy
contained a contradiction because whatever resistance existed was likely
to be unrecognizable or ignored. Another contradiction existed in mod¬
ernists’ drive for autonomy. Objectiwsts sought a “pure” dance in order
to remain free of outside influence, but their art’s lack of overt signs of
protest, indeed its apparent lack of all meaning, dovetailed agreeably
with state institutional policies and desires.
Not that the muting of opposition had to be imposed from outside,
it was common enough within dance itself, whether it stemmed from a
desire to protect a vulnerable art or simply to maintain the status quo.
Frequently suppression came through an appeal to the very rules that
were supposed to keep dance free. As we have seen, choreographers who
attempted overt protest were labeled “literal” and eliminated from the
ranks of modernists, while those who opted for coded protest were dis¬
regarded. Artists who dealt with difference also encountered formidable
difficulties. At various times, modernist rules that demanded the embod¬
iment of universal experience were invoked either to emphasize or si¬
lence difference. Particularly poignant was the inability of commentators
to acknowledge the protest in African-American dance—not that of long-
accepted narratives of past iniquities but the protest against continuing
discrimination that was implicit in Dunham’s insistence on sexuality as
202 A Game for Dancers

a legitimate element of dance and in Primus’s and Beatty’s equal insis¬


tence on virtuosity as an expressive means open to black artists.
In addition to silence, commentators also used the language of aes¬
thetics throughout the postwar period as a way of deflecting critical de¬
bate, whether by directly censuring dances in which protest was clearly
present, or by reducing dances to formal concerns alone. The use of aes¬
thetics as a weapon against critique was facilitated by the very rules of
modernism, which tended to favor form both as a means of placing con¬
trol in the artist’s hands and as a means of defeating narrative and illus¬
tration. Objectivism, which had no overt subject matter except dance it¬
self, encouraged the use of a reductivist aesthetic language to describe
it and its intentions.
Was there, then, no way within the confines of modernism to resist
or protest the forces that modernism ostensibly opposed? Was mod¬
ernism, in fact, nothing more than an enforcer of the status quo dis¬
guised as an oppositional “avant-garde”? I want to return for a moment
to Clement Greenberg’s modernist theory as a somewhat circuitous way
of addressing that question. As will be recalled, Greenberg defined
modernism as “the use of characteristic methods of a discipline to
criticize the discipline itself, not in order to subvert it but in order to
entrench it more firmly in its area of competence” ([i960] 1993b; 85).
The point of this self-criticism was to avoid a “leveling down” by
“demonstrating that the kind of experience they [each art] offered was
valuable in its own right and not to be obtained from any other activ¬
ity” (ibid.: 86). For Greenberg, any resistance that adhered to mod¬
ernist art existed in its refusal to yield to leveling processes through a
constant reexamination of formal methods and materials and the elimi¬
nation of any aspect that did not reinforce a particular art’s essential ele¬
ments. The rejection of politics was a given in this dedication to art’s
purification.
Certainly modern dancers feared the culture industry. However, it
is debatable whether the fear was as much one of contamination from
below as one of seduction and co-option from more powerful forces.
There is in Greenberg a contempt for the lower orders that is seldom
felt in the theory or practice of modern dance. Dancers feared the cul¬
ture industry because, in its power, it offered economic stability and
fame. But they also feared the legitimized high-art of ballet that threat¬
ened to overwhelm and consume a weaker modern dance. Both ballet
and entertainment were viewed as aspects of entrenched power.
Objectivism’s Consonance 203

Dancers would have found other points of disagreement with Green¬


berg, the most important having to do with the role of representation
in art. Greenberg attempted to circumvent representation in painting
by allowing it to include only those elements that could be found no¬
where else. This would give painting autonomy, but the price it would
have to pay was separation from life. Dancers, however, felt that dance’s
essential element, the element that set it apart from Hterature and paint-
ing, could not be separated from life. The embodied nature of dance
movement always intimated meaning, no matter how free it was of
representation. Although it might be presumed that this attachment to
life would prevent dance from achieving independence, there was an
answer: because all humans shared embodiment, no education or con¬
ditioning was needed to respond to it. This was quite different from
Greenberg’s idea, expressed in Avant-Garde and Kitsch, that genuine art
required reflection to make its effects ([1939] 1986a: 1:16). As far as
modern dance theory was concerned, any kind of intellectual mediation
made for an adulterated response. Dance could only he “understood”
body to body through kinesthetic sympathy. In addition, any possibility
of resistance or protest in dance could only be attained through move¬
ment freed of rationalized processes, most evident in language and il¬
lustration. Therefore, the fact that resistance, per se, did not enter writ¬
ten discourse was of little consequence. Resistance lay in practice, the
activity of dancing, not in the representation of protest in dance nor in
written or spoken responses to it. Resistance lay in Dunham’s grinding
hips. Primus’s leap, Beatty’s balletic extensions, Sokolow’s gestures of
madness, in the unpredictability of Nikolais’s decentralized movement
and Cunningham’s controlled accidents. This was movement to con¬
front the accepted, to destabilize the familiar and comfortable. And it
would make itself felt through corporeal means.
This solution to the dual problem of freedom and resistance may seem
at best idealistic and at worst conveniently constructed to work both
ways. One may question how effective a concept of resistance through
bodily practice might be and think how easily it fits the role of malleable
tool for dominant ideologies. On the other hand, the social sciences
have shown how much of human behavior is dependent on unconscious
corporeal learning and teaching, and if such bodily practices are used to
reinforce the status quo, they are also employed as instruments of change.
In the final analysis, too, modernists were idealistic; they believed art
could change society.
204 A Game for Dancers

Modernist intentions are doomed to remain ambiguous and contra¬


dictory; these are accounted for in the structure and practice of mod¬
ernism itself. But if the desire for acceptance is part of modernism, so
is the desire to oppose. Opposition, though, must occur however it
can, and the postwar period was not one in which opposition was easy.
Possibilities were limited not only by modernist rules but by the pres¬
sures of the larger social world, not least of which was the increasing co¬
option of modernist form by institutions of power. The difficulties that
beset postwar modern dancers were many and serious, yet there can be
no doubt that dancers struggled to solve those problems within the
choices they thought possible. If their solutions were not free of con¬
tradiction, their faith in embodied movement nevertheless demonstrates
the power they felt existed in their medium and how that power could
be exercised even in the face of oppression.
Notes

Inti’oduction

1. Examples include Ramsay Burt’s The Male Dancer: Bodies, Spectacle, Sexual¬
ities (1995) and Sally Banes’s Dancing Women: Female Bodies on Stage (1998), which
address gender issues within a broad range of modern dance; Naomi Jackson’s
Converging Movements: Modem Dance and Jewish Culture at the pind Street Y
(2000), which focuses on the lively interaction between modern dance and Jew¬
ish concerns at the Y from the 1930s onward; and Susan Manning’s Modem
Dance, Negro Da?ice: Race in Motion (2004), which traces the convergence of mod¬
em dance and African-American concert dance from the 1930s into the 1960s.
2. These are, most importantly, Naima Prevots’s Dance for Export: Cultural
Diplomacy and the Cold Wtr(igp8), which examines the American government’s
sponsorship of dance tours abroad during the Cold War, and Rebekah Kowal’s
dissertation. Modem Dance and American Culture in the Early Cold War Years
(1999), which offers interpretations of the work of Martha Graham, Merce
Cunningham, Alwin Nikolais, Paul Taylor, and Aim(a) Halprin in light of Cold
War politics.
3. Mark Franko’s reevaluation of a modernist narrative of modern dance
history in Dancing Modemism/Perf arming Politics is among the few works to
consider modernism itself in detail, although his project centers on the politics
of modernist interpretation. Also, as in most dance research, his work reaches
across a spectrum of modem dance rather than focusing on postwar modernism
per se.
4. Since, as has often been said, there are many modernisms, I am not going
to attempt to deal with a comprehensive treatment of the subject. Rather, I
have chosen specific, accepted elements of modernism on which to focus in this
study, including issues of representation and autonomy.
5. A number of anthologies document the modernist debates within art his¬
tory, including Frascina 1985, Guilbaut 1990, Frascina and Harrison 1987, and
Harrison and Wood 1993. Nor do the issues surrounding modernism appear
2o6 Notes to Pages xv-9

to be settled; see, for example, Clark 1998 (371-404) for a recent analysis of ab¬
stract expressionism.
6. Antonin Artaud reflected this uncertainty when he wrote in The Theater
and Its Double: “If confusion is the sign of the times, I see at the root of this con¬
fusion a rupture between things and words, between things and the ideas and
signs that are their representation” ([1938] 1958: 7).
7. I use the term “objectivist” after Susan Foster, who employs it to de¬
scribe the dance of Merce Cunningham and other choreographers who favored
what she refers to as the “activity of moving” in opposition to “expressionist
dance” (1986: 167). Aldiough “formalist” is also often used to describe the work
of such choreographers, “objectivist” to my mind is less loaded. At the same
time, 1 will use Martin’s term “expressional” modern dance when differentiat¬
ing older modern dance from the new objectivism.
8. This can readily be seen simply by comparing the increasing number of
articles on dance listed in the Readers’’ Guide to Periodicals from the 1930s through
the 1950S.
9. In particular, Balanchine’s article entitled “Ballet Goes Native,” pub¬
lished in Dance in 1937 has Denby’s cadence and phrasing. It appeared in the
December issue of the magazine, eight months after Denby had written a per¬
ceptive and complimentary review of Balanchine’s choreography (Denby [1937]
1986: 44-46).
10. Hering had done freelance work for Dance Magazine beginning in 1945
(Doris Hering, personal interview with author, 17 September 2004).

Chapter i. The Trouble with Modern Dance

1. Foote included in this category not just artists such as Martha Graham
and Doris Humphrey but “modern” ballet choreographers Agnes de Mille,
Antony Tudor, and Jerome Robbins.
2. Just how porous, Foote himself could attest. In addition to being a
Dance Observer editor, Foote was an aspiring Broadway actor. Although he had
little success on the stage, he eventually won fame as a television and Holly¬
wood screenwriter (To Kill a Mockingbird, Tender Mercies) and Pulitzer Prize¬
winning playwright (The Trip to Bountiful, The Young Man from Atlanta).
3. See Jane Feuer’s The Hollywood Musical (1993) for an interesting discus¬
sion of how American ballet, in incorporating folk and modern dance elements,
made itself more accessible to a mass audience, thereby winning a prominent
place in Broadway and Holl)nvood musicals after the war.
4. See Rosenbaum 1943 and Terry 1947b for the economic plight of dancers
and the relationship of high-art dance to commercial work.
5. Regionalists included, most prominently. Grant Wood, Thomas Hart
Benton, and John Steuart Curry (Dennis 1998; Corn 1985; Doss 1991).
Notes to Pages 9-40 207

6. Xhe synthesis of idioms was institutionalized in the dance curriculum de¬


veloped by Martha Hill for the Juilliard dance program initiated in 1951.
7. Undated program for Choreographers’ Workshop, Dance Collection,
New York Public Library. In this case, the workshop presented Bettis’s As I Lay
Dy/wg with Horton Foote’s Goodbye to Richmond, which had been commissioned
by the Neighborhood Playhouse. Xhese were billed as “two dance plays.”
Choreographers’ Workshop was founded in 1946 by Xrudy Goth, Patricia
Newman, and Atty van den Berg to showcase the work of young choreogra¬
phers (Hering 1947).
8. Analysis of The Moor's Ravane is based on a performance by the Jose
Limon Dance Company (1955-1957 video) mjose Lmion: Three Modem Dance
Classics, Canada: CBC Production.
9. Limon called his characters Xhe Moor, His Friend, His Friend’s Wife,
and Xhe Moor’s Wife, which correspond to Shakespeare’s Othello, lago, Emiha,
and Desdemona. I have used Shakespeare’s designations for the sake of clarity and

Chapter 2. Ballet's Challetige

1. Although the pamphlet does not identify the writer, its style is unmistak¬
ably that of Kirstein, and his authorship is generally accepted. Xhe pamphlet,
from a private collection, was produced for Ballet Society in 1946 and was in¬
cluded in the exhibition Dance for a City: Fifty Years of the New York City Ballet,
organized by Lynn Garafola and Eric Foner at the New York Historical Society,
20 April to 15 August 1999.
2. Xhe Ballet Russe de Monte Carlo had begun its existence in Europe in
the 1930s but made the United States its home after the advent of World War
II (Anderson 1981).
3. Despite his written attacks, Kirstein’s actual relationship with the genre
was more ambivalent. He invited Iris Mabry and Merce Cunningham to show
or contribute work to Ballet Society, and he asked Xalley Beatty to dance in
Lew Christensen’s Blackface in 1947. He also knew Graham well, spending time
at her studio and writing favorably about her in an essay for Merle Armitage’s
book, Martha Graham: The Early Years (1937). In addition, Kirstein invited
Graham to choreograph half of Episodes in 1959, the other half choreographed
by Balanchine.
4. Although Kirstein here sounds much like Greenberg, the two had little
time for each other. Kirstein, an advocate of surrealism and neoromanticism,
wrote an article in 1948 condemning abstract expressionism and arguing against
Greenberg (Kirstein 1948). Greenberg, for his part, had contempt for Kirstein.
On the one occasion in which he mentioned Kirstein in writing, it was to dis¬
parage him and his taste for neoromanticism (Greenberg ([1950] 1993a: 59-62).
2o8 Notes to Pages 40-53

It is an interesting footnote to a footnote that the only time Greenberg ven¬


tured into the field of dance was to write a review defending Tudor’s Dim Lar-
trr ([1945] 1986c: 36-39). Apparently, his high-modernist purity did not extend
to ballet.
5. For an analysis of the relationship of New York City Ballet and the
Rockefeller Foundation, see Garafola 2002.
6. Kirstein founded Ballet Caravan in 1936 to provide work for the Amer¬
ican Ballet dancers during the time they were not needed by the Metropolitan
Opera. American Ballet had won the contract to become the resident company
at the Met, but it soon became apparent that it would not be allowed to develop
there. After the demise of American Ballet in 1938, Kirstein made Ballet Car¬
avan a full-time touring group. Ballet Caravan’s name was changed to Ameri¬
can Ballet Caravan in 1941 and joined with the remnants of American Ballet
for a Latin American tour that year. The entire enterprise was disbanded after
the company returned to the United States (Kirstein 1973: 43-81; Garafola
1999: 4-5)-
7. Balanchine tried to commission a score from Hindemith in 1937, but
the composer refused. He agreed in 1940 when his financial situation changed
(Buckle 1988: 162).
8. The analysis of The Four Temperaments is based on a silent Ballet Soci¬
ety rehearsal film of parts of the ballet (1946 film) with most of the original cast.
This film includes the original finale. Also used were two performances by New
York City Ballet made under Balanchine’s direction: a Radio Canada produc¬
tion (1961 film) and a Dance in America production (1977 video); also two
videos made of New York City Ballet at the New York State Theatre, Lincoln
Center (1993 video), one a close-up and the other a full-stage view of the same
performance. All are at the Dance Collection, New York Public Library for the
Performing Arts. Unless otherwise noted, I will refer to the 1977 video, since
it is commercially available and therefore the most easily accessible.
9. For a nuanced analysis of Balanchine’s use of music in his work, see
Jordan 2000.
10. Dance theorist Evan Alderson calls such movement associations “dance
metaphors” (Alderson 1983).
11. See Banes (1999) for an analysis of modern dance influences on Balan¬
chine’s choreography.
12. See also Duell 1987; Ashley 1987; and Wilde 1987, all of whom partic¬
ipated in a panel discussion that was expanded for the first in a series of articles
entitled “Celebrating The Four Temperaments" published in Ballet Review (Win¬
ter 1987, Spring 1987, and Spring 1988).
13. On surrealist aims and methods, see Breton 1972 and 1978; and Nadeau
1989. On surrealists in the United States, see Sawin 1997.
14. It should be added that by the time the surrealists reached America’s
Notes to Pages 53-76 209

shores, their own use of the methods described here had lost most of their sub¬
versive edge. By the end of the 1930s, the majority of the surrealist artists were
showing in important galleries and publishing with major houses, that is they
had already achieved consecration (see Richard Martin 1987; Sawin 1997; and
Brandon 1999).
15. In an article written in 1987, feminist dance historian Ann Daly argued
that Balanchine portrayed women as passive instruments in the hands of men,
using as her primary examples the thematic duets from The Four Te?nperaments.
Although other commentators disagreed, arguing from a wider range of Bal¬
anchine ballets, the violent manipulation of women in The Four Temperaments
had long been noted as one of the work’s most striking features (Daly 1987;
Jordan and Thomas 1994; and Siegel 1997).

Chapter Modernist Theory: John Martin, Edwin Denhy,John Cage

1. Cage is considered here only in his interaction with dance and dancers
in the 1940s and ’50s. Sally Banes and Noel Carroll have argued that Cage’s
vanguardism in his own compositions outstripped Cunningham’s. Marjorie
Perloff has also defended the radicalism of Cage’s work (Banes and Carroll
2005; Perloff 1989).
2. For biographical information on John Martin, see Sabin 1946; Hering
1952; Anderson 1989; and the unsigned New York Times obituary, “John Mar¬
tin Is Dead,” 21 May 1985.
3. Richard Boleslavsky was an alumnus of the Moscow Arts Theatre and
is credited with introducing the ideas of Constantin Stanislavski to the United
States.
4. See Damasio (2003) for a leading neurologist’s theory of how emotion
is manifested. Damasio’s theory is similar to Martin’s to the degree that Dama¬
sio posits that a physical action (for example, blinking or clenching) occurs first
and then a feeling about it is produced.
5. Martin’s criticism did not always reflect his theory. For example, he
praised narrative works by Graham, Maslow, and Limon.
6. The chronology of the Franco-Russian companies that emerged in the
wake of Diaghilev’s death in 192 9 is confusing because the company names (in¬
cluding spellings) and personnel changed frequently. Often choreographers
and dancers worked for a season or two in one troupe and then another. The
first company, Les Ballets Russes de Monte Carlo, was founded in 1932 and was
headed by Colonel W. de Basil and, until 1934, by Rene Blum. Another com¬
pany, the Ballet Russe de Monte Carlo, was founded in 1938 with Sergei Den¬
ham as director. This company based itself in the United States during the war,
while the de Basil troupe toured Australia and Latin America. The de Basil
troupe returned to Europe after the war and disbanded in 1948. Denham’s
210 Notes to Pages 76-94

Ballet Russe remained an American company until its demise in 1962 (Ander¬
son 1981; Garcia-Marquez 1990; Hurok 1947, 1953)-
7. Although Tudor was English, he moved to the United States in 1939
and lived and worked there for most of his career until his death in 1987 (Perl-
mutter 1991; Chazin-Bennahum 1994)- Denby initially responded favorably to
Tudor’s ballets, but he soon began to have reservations about them (see Denby
[1942] 1986: 93-95; [1943b] 1986: 129-131; and Jordan 2000).
8. A time art is defined as follows: “The physical conditions in which all the
arts have their existence, so far as they become communicable to the human
senses, are either space (painting, sculpture, etc.) or time, or both (drama and
in a sense all literature)” (Blom 1971: 701). Dance also would be categorized as
an art of both time and space, although Cage, as a musician, is here consider¬
ing it in terms of time.

Chapter /f.. Embodying Community

1. The analysis here is based on The Village I Knew (1977 video), performed
by the Sophie Maslow Dance Company at the Theatre of Riverside Church,
Dance Collection, New York Public Library for the Performing Arts. Maslow
originally choreographed some of the dances for The Village I Knew as Festival
in 1949. She enlarged the work in 1950.
2. The New Dance Croup had long made racial integration a policy and
Maslow embraced it too. In fact the inclusion of McKayle and Aul in Maslow’s
work was itself a political act and commitment. Maslow has said that having
a mixed company eliminated an entire area of touring for her in the South and
made it difficult in the North where blacks were not allowed to stay in most
white hotels or eat in white restaurants, among a number of forms of discrim¬
ination (Maslow 1984).
3. Although it is not clear from McKayle’s account under what circumstances
Martin made this comment, he made a similar remark in less inflammatory
language mjohn Martin's Book of the Dance (1963: 189).
4. Sokolow continued to explore Jewish themes in the 1950s and beyond,
both on the stage and for television. Her best known of these later works was
Dreams, which she created in 1961 and which dealt with the horrors of the
holocaust.
5. The analysis of Kaddish is drawn from Kaddish (1990 video), Deborah Zall
soloist. Dance Collection, New York Public Library for the Performing Arts.
The analysis of Lamentation is drawn from a film of excerpts from the dance
with Craham as soloist and from a complete version of the dance with Peggy
Lyman as soloist: Lamentation (1943 film), Martha Craham soloist and with
an introduction by John Martin, Harmon Foundation production. Dance Col¬
lection, New York Public Library for the Performing Arts; Lamentation (1976
Notes to Pages 94-120 2II

video), Peggy Lyman soloist, Martha Graham Dance Company, Dance in Amer¬
ica production. Dance Collection, New York Public Library for the Perform¬
ing Arts.
6. This analysis oiRooms is based on two video recordings, one of excerpts,
one complete: Rooms (1966 video) excerpts, Anna Sokolow Dance Company,
telecast on WNET-TV, New York, Jac Venza, producer (the dances “Dream,”
Daydream, and “The End?” are omitted); Rooms (1975 video) complete.
Contemporary Dance System, performed at American Place Theatre. Both at
the Dance Collection, New York Public Library for the Performing Arts.
7. Program, Dance Recital by Anna Sokolow and Dance Unit of the New
Dance League, 92nd Street Y, 28 Eebruary 1937, 92nd Street YW-WMHA
Archives.
8. This analysis is based on Session for Six (1964 video), Juilliard Dance
Ensemble, Dance Collection, New York Public Library.
9. The twist became popular in i960 when Chubby Checker released his
hit record “The Twist,” but the dance grew out of the jitterbug of the 1940s
and the variations that teenagers gave it in the 1950s. Sokolow’s Opus works
parallel the development of late ’50s dances into the twist.
10. The analysis of Opus ’65 is based on a performance by the Juilliard
Dance Ensemble (1963 video) and of Opus Yy on a performance by the City
Center Joffrey Ballet (1967 video), both at the Dance Collection, New York
Public Library for the Performing Arts.

Chapter 5. African-American Vanguardism: ig^os

1. See, for example, Aschenbrenner 1981, 2002; Perpener 2001; DeFrantz


2002, 2004; Gottschild 1988, 1996; and Manning 2004.
2. See Aschenbrenner 2002 for a detailed analysis of the relationship of
anthropology to Dunham’s life and work. For biographical material and infor¬
mation on Dunham’s early career, see Beckford 1979; Perpener 2001; Aschen¬
brenner 2002; Barzel 1983; and Manning 2004.
3. Adjustments were soon made in the program, perhaps because critics
found Bre’r Rabbit weak. Dunham dropped the story line of the work and ex¬
cerpted dances from it to make what she entitled Plantation and Minstrel Dances
from the Ballet BreV Rabbit. She also changed the order of the last two suites,
making the exciting Lejazz “Hot” the finale.
4. The duet from LAg’Ya is shown in Katherine Dunham Company (1941-
1944 video) [presented by] the Katherine Dunham Centers for Arts and Hu¬
manities, choreography and direction by Katherine Dunham. Dance Collec¬
tion, New York Public Library for the Performing Arts.
5. Choros #i and Rhumba Jive (usually listed on printed programs as
Rutnba with a Little Jive Mixed In) can be seen in Katherine Dunham Company
212 Notes to Pages 120-139

(1941-1944 video) [presented by] the Katherine Dunham Centers for Arts
and Humanities, choreography and direction by Katherine Dunham, as well as
in Katherine Dunham. Company Tropical Revue (1947 video): [presented by] the
Katherine Dunham Centers for Arts and Humanities, choreography and di¬
rection by Katherine Dunham. The latter film was made by Ann Barzel of a
performance at the Studebaker Theater, Chicago. Both films are in the Dance
Collection, New York Public Library for the Performing Arts.
6. The authorship of this review is not altogether clear, considering the
pseudonyms routinely used by critics writing in leftist publications. The Daily
Worker review is almost identical to parts of one written on i March 1940 in
TAC, 3. leftist magazine that ceased publication in 1942. Susan Manning states
that circumstantial evidence points to Edna Ocko as the writer of TACs un¬
signed review (Manning 2004: 258-259ni53). If so, it is likely Ocko wrote the
Daily Worker review, as well. Ocko wrote often for the Daily Worker as well as
TAC and other of the leftist publications. If Ocko (or the same author) wrote
both reviews, it would mean that essentially there was one opinion represent¬
ing leftist views of Tropics and Lejazz “Hot, ” since Owen Burke only mentioned
the Dunham season in passing in New Masses (Burke 1940: 30).
7. Printed program, Katherine Dunham and her company in a tropical
revue, 19 September 1943, Martin Beck Theatre, New York. Dance Collec¬
tion, New York Public Library for the Performing Arts.
8. This interest in psychology and sociology may have been prompted by
her friendship with Erich Eromm, whom Dunham met at the University of
Chicago and who remained a friend and mentor for many years.
9. Among those who spoke of the intrusion of e.xtraneous elements in
the program were George Freedley of the Nrw York Morning Telegraph, Fredi
Washington of the People's Voice, Edwin Denby in the New York Herald Tribune,
and John Martin in the New York Times. Although Martin mentioned intrusive
elements in Primus’s concert, he wrote a generally laudatory review under the
headline, “Brilliant Dancing by Pearl Primus” (Martin 1944b). Martin was an
admirer of Primus’s work and continued to support her through news articles
and reviews when others were less positive.
10. This work, called Block Party, was to have had music by Virgil Thomson.
However, it was never completed (Beatty 1990; Nash 1992: 13).
11 • The analysis of Southejyi Landscape is based on a performance by the
Philadelphia Dance Company (Philadanco), Joan Brown, director, in A Memo¬
rial, Tribute, Celebration of the Life and Art of Choreographer/Dancer Mr. Talley
Beatty (1995 video), Dr. Glory Van Scott, producer; and “Mourner’s Bench”
(1948 film), choreographed and performed by Talley Beatty. Both are in the
Dance Collection, New York Public Library for the Performing Arts. Accord¬
ing to Joan Brown, the Elie Siegmeister score was lost, and Beatty set Southern
Notes to Pages 139-152 213

Lmidscape to spirituals, folk, and contemporary songs when he revived it for her
company. Personal interview, 15 September 2004.
12. Beatty in The New Dance Group Gala Concert (1994 video). American
Dance Guild Production, New York.
13. Printed program, Talley Beatty and Company, 24 October 1948. The
Dance Center of the YW-WMHA, 92nd Street and Lexington Avenue, New
York. 92nd Street YW-YMHA Archives.
14. Personal interview, Joe Nash, New York City, 16 July 2002. In the last
section of the Philadelphia Dance Company production, Beatty changed not
just the choreography and music but the concept as well. He now titled the
segment Runagate Variations.” A runagate was an escaped slave. Therefore,
the new finale, with its implication of flight to the North and freedom, appears
more optimistic than the original “Settin’ Up,” which continued the work’s
overall mood of resignation and sorrow.
15. It should be noted that Smith had been a critic for the Chicago Tribune
and knew of Beatty as a young dancer. According to Beatty, Smith had seen
him perform with Dunham while he was still in middle school and provided
him with a scholarship to attend ballet classes with Edna McCrae while Dunham
was in Haiti (Beatty 1990).
16. The analysis of Ophelia is based on a performance in Dance and Myth:
The World of Jean Erdman, part i, “Early Solos” (1993 video). National Video
Industries.
17. This pose was found among Louis Horst’s “medievalist” forms in his
composition classes and was frequently used by dancers who had studied with
him. See Horst and Russell [1961] 1987: 76-86, and illustration on 80.
18. Erdman’s quote is from Dance and Myth: The World of Jean Erdman, part
I, “Early Solos” (1993 video). National Video Industries. The analysis of
Hamadryad is based on a performance in the same video.

Chapter 6. African-American Vanguardism: i95or

1. The black press seldom covered modern dance concerts in the 1950s,
so written discourse centered almost exclusively on white reception. It is worth
noting, however, that the Amsterdam News and New York Age frequently ran
announcements of concerts by both black and white choreographers at the
92nd Street Y. Although this may have been due to some extent to the Y’s up¬
town location, it also supports Naomi Jackson’s contention that the Y reached
out to black artists and audiences. It equally suggests that the city’s African-
American newspapers thought news of these concerts would be of interest to
their readers.
2. The title of the dance work came from a chain gang song and referred
214 Notes to Pages 152-179

to the arched head of a pick, one of the tools prisoners used along with sledge¬
hammers to break rock.
3. The analysis of Rainbow Round My Shoulder is based on a 1959 perform¬
ance recorded in Donald McKayle: Early Work (1999 video), John Desmond, di¬
rector, produced by John McGiffert, Creative Arts Television Archive; also
performances recorded in Revelations [and] Rainbow Round My Shoulder (1973
video), Alvin Alley American Dance Theater; and Alvin Ailey American Dance
Theatre: Three by Three (1985). All videos are in the Dance Collection, New
York Public Library for the Performing Arts.
4. Donald McKayle, personal conversation, 26 November 2004, New York.
5. Only the dance press covered the early performances of Rainbow Round
My Shoulder. Of the black press, New York Age ran an announcement of the
opening (9 May 1959: 12) and the Amsterdam News ran an announcement after
the fact (16 May 1959: 16), but neither newspaper covered the event.
6. The analysis of The Road of the Phoebe Snow is based on a rehearsal per¬
formance by the Alvin Ailey Dance Theatre (1969 video) and a performance by
the Alvin Ailey Repertory Ensemble (1983 video) presented by the Riverside
Dance Festival. Both videos are in the Dance Collection, New York Public
Library for the Performing Arts.
7. In the late 1970s, a young Bill T. Jones took note of this development and
delivered blistering attacks against the audiences who encouraged it. In a series
of solos he seduced (mostly white) spectators with beautiful movement then ag¬
gressively confronted them in words and movement (Jones 1995; Morris 2001).

Chapter 7. Objectivism's Consonance

1. It should be emphasized that in this chapter I am speaking of the work of


Nikolais and Cunningham in the 1940s and ’50s. It is reasonable to assume that
their ideas, interests, and priorities might change over the course of their careers.
2. Septet (1964 film), Cunningham Dance Foundation Archives, New York
City.
3. Years later, Cunningham was still speaking of his dance in concrete terms.
In 2005, the Cunningham company Web-site prominently featured a quote by
Cunningham stating, “there’s no thinking involved in my choreography. ... I
don’t work through images or ideas—I work through the body.”
4. Summerspace (1958 video), filmed at the American Dance Festival, Con¬
necticut College. Cunningham Dance Foundation Archives, New York City.
5. Unfortunately, this quality has been lost, as a comparison to a video
performance of Summerspace made at Berkeley, California, in 2000 attests
(Cunningham Dance Foundation Archives, New York City). The Cunningham
dancers now perform the work as an exercise in mannered virtuosity.
Notes to Pages 179-19^ 215

6. A year earlier Nikolais had written a similar but less complete explana¬
tion of his ideas in a piece for the New York Times (Nikolais 1957).
7. Murray Louis, personal interview, 16 August 2003, New York.
8. Susan Buirge, personal interview, 5 November 2003, Paris.
9. Nikolais commented, “I was always a little ashamed of dancing. Chore¬
ography removed that.” Nik and Murray (1986 video), Dance Collection, New
York Public Library for the Performing Arts.
10. Phyllis Lamhut and Susan Buirge commented on how similar these ex¬
ercises were to Pilates mat exercises. Phyllis Lamhut, personal interview, 28 July
2003, New York; Susan Buirge, personal interview, 5 November 2003, Paris.
11. “Fixation” survives in A Time to Dance: Alwin Nikolais (1959 video).
Dance Collection, New York Public Library for the Performing Arts. Louis
has said the dance’s title reflected the repetitiveness of the movement (personal
interview, 16 August 2003, New York).
12. Murray Louis, personal interview, 13 August 2003.
13. Phyllis Lamhut, personal interview, 28 July 2003.
14. On stage Nikolais used improvised movement only in limited ways.
The dancers might, for example, be asked to do darting movements for a short
period of time at a specific location on the stage; otherwise, the dances were
generally set. According to Murray Louis, the one major exception to this rule
was Mirrors (1959), a piece completely improvised on stage by the company.
Since improvisation demanded constant, instant spontaneous decisions, Louis
said that the work was so exhausting it was only repeated four times and then
dropped from the repertory. Murray Louis, personal interview, 16 August 2003.
15. Murray Louis, personal communication, 16 August 2003.
16. Ruth Grauert, transcription of interview with Claudia Gitelman, April
1999, Collection of Claudia Gitelman, New York. Murray Louis, personal in¬
terview, 16 August 2003.
17. According to several of his early dancers, Nikolais often said that the
invention of synthetic fabric tights that were stretchable but did not sag
changed dance technique.
18. Program: The [Henry Street] Playhouse Dance Company, Etudes 11:
Masks, Props, and Mobiles, 26 January 1953. Program: The [Henry Street] Play¬
house Dance Company, Masks—Props—Mobiles, 10 December 1955. Both are
in the Dance Collection, New York Public Library for the Performing Arts.
19. Analysis of Kaleidoscope is based on a performance by the Henry Street
Playhouse Dance Company (1956 film), which includes five of eight sections.
Dance Collection, New York Public Library for the Performing Arts; and on a
performance of the entire work from 1956 in the collection of Murray Louis.
20. This was not the only time Nikolais attacked expressional modern dance
for being excessively focused on sexuality (1958: 43; 1968; 1971a: 9; 1971b;
1973; 1973-1974: 41-42). His most extensive attack came in a 1961 essay he
2i6 Notes to Pages 198-200

contributed to The Encyclopedia of Sexuality (1961a) in which he accused the old


modern dance of sexual disfunction and illness in the harshest terms.
21. Although objectivist modern dance was soon overtaken by early post¬
modernism, it is debatable how much of a break with the past postmodernism
was. With the passing years, early postmodernism looks more like reformation
than revolution, despite manifestos to the contrary.
REFERENCES

Adler, Les K., and Thomas G. Paterson (1970). “Red Fascism: The Merger of
Nazi Germany and Soviet Russia in the American Image of Totalitarianism,
ig^os-ig^os” American Historical Review 75 (April): 1,046-1,064.
Adorno, Theodor, and Max Horkheimer ([1944] 1979). Dialectic of Enlighten¬
ment. Translated by John Gumming. Reprint, London: Verso.
Alderson, Evan (1983). “Metaphor in Dance: The Example of Graham.” Pro¬
ceedings, Sixth Annual Conference of Dance History Scholars: 11 i-i 18.
Anderson, Edith (1943). “Daring New Dancer Is Eirst Negro to Make YMHA
Debut.” Daily Worker February): 7.
Anderson, Jack (1981). The One and Only: The Ballet Russe de Monte Carlo. Lon¬
don: Dance Books.
-(1989). Introduction to John Martin: The Dance in Theory. Princeton,
N.J.: Dance Horizons.
Arato, Andrew, and Eike Gebhardt, eds. (1995). The Essential Frankfurt School
Reader. New York: Continuum.
Armitage, Merle ([1937] 1985). Martha Graham: The Early Years. Reprint, New
York: Da Capo Press.
Artaud, Antonin ([1938] 1958). The Theater and Its Double. Reprint, New York:
Grove Press.
Aschenbrenner, Joyce (1981). Katherine Dunham: Reflections on the Social and Po¬
litical Contexts of Afro-American Dance. Dance Research Annual XII, edited by
Patricia A. Rowe. New York: Congress on Research in Dance.
-(2002). Katherine Dunham: Dancing a Life. Urbana: University of Illi¬
nois Press.
Ashley, Merrill (1982). “Merrill Ashley.” In Striking a Balance: Dancers Talk
about Dancing, edited by Barbara Newman. Boston: Houghton Mifflin.
-(1984). Dancing for Balanchine. New York: Dutton.
Ashley, Merrill, et al. (1987). “Celebrating The Four Temperaments—1.” Ballet
Review (Winter): 12-35.
2i8 References

Balanchine, George (1937). “Ballet Goes Native.” Dance (December): 13. Re¬
printed (1944) as “The American Dancer.” Dance News (April): 3, 6.
-(1945)- “Notes on Choreography.” Dance Index 4, 2-3 (February-
March): 20-31.
_(1951). “Marginal Notes on the Dance.” In The Dance Has Many Faces,
edited by Walter Sorell. Cleveland: World Publishing.
-(1954). Balanchine’s Complete Stories of the Great Ballets. Edited by Fran¬
cis Mason. Garden City, N.Y.: Doubleday.
-(1961). “Balanchine: An Interview by Ivan Nabokov and Elizabeth
Carmichael.” Horizon (January): 44-56.
-(1965). “Now Everybody Wants to Get into the Act.” Life (ii June):
97-102.
-(1968). Balanchine’s New Complete Stories of the Great Ballets. Edited by
Francis Mason. Garden City, N.Y.: Doubleday.
-(1984). By George Balanchine. New York: San Marco Press.
Balanchine, George, and Francis Mason (1984). Balanchine’s Festival of Ballet.
2 vols. London: W. H. Allen.
Balcom, Lois (1944a). “What Chance Has the Negro Dancer?” Dance Observer
(November): iio-iii.
-(1944b). “The Negro Dances Himself.” Dance Observer (December):
122-124.
Ball, William (1984). A Sense of Direction. New York: Drama Book Publishers.
Banes, Sally (1980). Teipsichore in Sneakers: Post-Modem Dance. Boston: Hough¬
ton Mifflin.
-(1994). Writing Dancing in the Age of Postmodernism. Ahddletown, Conn.:
Wesleyan University Press.
-(1995). “Images.” In Dance Words, compiled by Valerie Preston-Dunlop.
Chur, Switzerland: Harwood Academic Publishers.
--(1998). Dancing Women: Female Bodies on Stage. London: Routledge.
-(1999). “Sibling Rivalry: The New York City Ballet and Modem Dance.”
In Dance for a City: Fifty Years of the New York City Ballet, edited by Lynn
Garafola, with Eric Foner. New York: Columbia University Press.
Banes, Sally, and Noel Carroll (2005). “Curmingham, Balanchine, and Postmod¬
ern Dance.” Paper presented at Dancing from the Center, Society of Dance His¬
tory Scholars Conference, 9-12 June. Northwestern University, Evanston, lU.
Barmerman, Henrietta (1999). “An Overview of the Development of Martha
Graham’s Movement System (1926-1991).” Dance Research 17, 2 (Winter):
9-46.
Barber, Beverly Hillsman (1992). “Pearl Primus: Rebuilding America’s Cul¬
tural Infrastmcture.” W African-American Genius in Modem Dance, Gerald E.
Myers and Stephanie Reinhart, project directors. Durham, N.C.: American
Dance Festival.
References 219

Barron, Stephanie (1991). “Degenerate Art”: The Fate of the Avant-Garde in Nazi
Geiynany. Los Angeles: Los Angeles County Museum of Art.
Barzel, Ann (1983). “The Untold Story of the Dunham/Turbyfill Alliance.”
Dance Magazme (December): 91-98.
Beatty, Talley (1990). Unpublished interview conducted by Dawn Lille. Type¬
script from audio recording. Collection of Dawn Lille, New York.
-(1993). Speaking of Dance: Talley Beatty (video). Interview with Talley
Beatty. American Dance Festival production, directed and produced by
Douglas Rosenberg. Dance Collection, New York Public Library.
Beckford, Ruth (1979). Katherine Dunham: A Biography. New York: Marcel
Dekker.
Beiswanger, George W. (1942). “Lobby Thoughts and Jottings.” Dance Ob¬
server (November)-. 116-117.
-(1943a). “Dance over the U.S.A.” Part i. Dance Observer (January): 4-6.
-(1943b). “Dance over the U.S.A.” Part 2. Dance Observer (February):
16-17.
-(1957). “New London: Residues and Reflections.” Dance Observer (Feb¬
ruary): 21-23.
Bell, Daniel ([i960] 1988). The End of Ideology: On the Exhaustion of Political Ideas
in the igyos. Reprint, with a new afterword by the author. Cambridge Mass.:
Harvard University Press.
Bentley, Eric (1971). ThifVy Years of Treason: Excerpts from. Hearings before the
House Committee on Un-American Activities, ipy8—ip68. New York: Viking
Press.
Bernstein, Harry (1958). “Alvin Alley, Ernest Parham and Companies.” Dance
Observer (June-July): 88.
-(1959)- “Kevin Carlisle, Donald McKayle and Companies.” Dance Ob¬
server (June-^uly)-. 92-93.
Berube, Allan (1990). Coming Out under Eire: The History of Gay Men and Women
in World War II. New York: Free Press.
Bettis, Valerie (1945). “Young Dancers State Their Views: As Told to David
Zellmer.” Dance Ofoeryrr (August-September): 82.
Blom, Eric (1971). Everyman's Dictionary of Music, revised by Sir Jack Westrup.
New York: New American Library.
Bourdieu, Pierre (1977). Outline of a Theory of Practice. Cambridge: Cambridge
University Press.
_(1986). “The Forms of Capital.” In Handbook of Theory and Research for
the Sociology of Education, edited by John G. Richardson. New York: Green¬
wood Press.
_(1990a). In Other Words: Essays towards a Reflexive Sociology. Palo Alto,
Calif.: Stanford University Press.
_(1990b). The Logic of Practice. Palo Alto, Calif.: Stanford University Press.
220 References

-(1993a). The Field of Cultural Production. New York: Columbia University-


Press.
-(1993b). Sociology in Question. London: Sage Publications.
-(1996). The Rides of Art. Cambridge: Polity Press.
Bourdieu, Pierre, and Loic Wacquant (1992). An Invitation to Reflexive Sociology.
Chicago: University of Chicago Press.
Brandon, Ruth (1999). Surreal Lives: The Surrealists, London:
Macmillan.
Breton, Andre ([1969] 1993). Conversations: The Autobiography of Surrealism.
New York: Paragon House.
-(1972). Manifestoes of Surrealism,. Ann Arbor: University of Michigan
Press.
-(1978). Andre Breton: What Is Surrealism,! Selected Writings. Edited by
Franklin Rosemont. New York: Monad Press.
Brinkley, Alan (1996). End of Reform: New Deal Liberalism in Recession and War.
New York: Vintage Books.
-(2001). “The Illusion of Unity in Cold War Culture.” In Rethinking
Cold War Culture, edited by Peter J. Kuznick and James Gilbert. Washing¬
ton, D.C.: Smithsonian Institution Press.
Brodie, Joan (1948). “Choreographers’ Workshop.” Dance Observer (April): 44.
-(1950). “Talley Beatty and Company.” Dance Observer (November):
138-139.
Broughton, Shirley (1957). “Close-Up of Modern Dance Today: The Non-
Objective Choreographers.” Dance Magazine (November): 20-23.
Brown, Jean Morrison, ed. (1979). The Vision of Modem Dance. Princeton,
N.J.: Princeton Book.
Buckle, Richard, in collaboration with John Taras (1988). George Balanchine,
Ballet Master. New York: Random House.
Burger, Peter (1984). Theoiy of the Avant-Garde. Minneapolis: University of
Minnesota Press.
Burke, Owen (1940). “A Review of Recent Outstanding Performances.” New
Masses (26 March): 30.
Burley, Dan (1940). “Miss Dunham Wins B’way with Dances.” New York Am¬
sterdam News (24 February): 21.
Burt, Ramsay (1995). The Male Dancer: Bodies, Spectacle, Sexualities. London:
Routledge.
-(1998). Alien Bodies: Representations of Modernity, ''Race’’’’ and Nation in
Early Modem Dance. London: Routledge.
Butler, Gervase (1937). “American Ballet.” Dance Observer
-(1943). “Five Dancers.” Dance Observer (March): 27.
C^S^iJohri (1944). “Grace and Clarity.” Dance Ofacrycr (November): 108—109.
-(^957)- This Day.” Dance Observer (January): 10.
References 221

-(^973)- Silence. Middletown, Conn.: Wesleyan University Press.


-— (1991). John Cage: An Anthology. Edited by Richard Kostelanetz. New
York: Da Capo Press.
Campbell, Joseph (1944a). “Betwixt the Cup and the Lip.” Dance Observer
(March): 30-31.
-(1944b). “Text, or Idea?” Dance Observer (June—July): 66, 75.
-(1945). “The Jubilee of Content and Form.” Dance Observer (May):
52-53-
Caute, David (1978). The Great Fear: The Anti-Communist Purge under Truman
and Eisenhower. New York: Simon and Schuster.
Chafe, MTlliam (1982). “The Civil Rights Revolution.” In Reshaping America:
Society and Institutions, ig4y-ip6o, edited by Robert Bremner and Gary
Reichard. Columbus: Ohio State University Press.
-(1999). The Unfinished Journey: America since World War II. 4th ed. Ox¬
ford: Oxford University Press.
Chambers, John Whiteclay (1992). The Tyranny of Change: America in the Pro¬
gressive Era, ipoo-ipiy. New York: Palgrave Macmillan.
Charlip, Remy (1954). “Concerning Merce Cunningham and His Choreog¬
raphy: Composing by Chance.” Dance Magazine (January): 17-19.
Chazin-Bennahum, Judith (1994). The Ballets of Antony Tudor: Studies in Psyche
and Satire. Oxford: Oxford University Press.
Chujoy, Anatole (1943). “Dance in Review.” Dance News (October): 3.
-(1949). The Dance Encyclopedia. New York: A. S. Barnes.
-(1953). The New York City Ballet. New York: Alfred A. Knopf.
Churchill, Allen (1949). “Dancer with a Dream ...” Theatre Arts (March): 34-37.
Clark, T. J. (1984). The Painting of Modem Life: Paris in the An of Manet and His
Pollowers. Princeton, N.J.: Princeton University Press.
-(1998). Earewell to an Idea: Episodes from a History of Modernism. New
Haven, Conn.: Yale University Press.
Clark, Veve A., and Margaret B. Wdkerson, eds. (1978). KAISO!Katherine Dun¬
ham, an Anthology of Writings. Berkeley: University of California, Institute
for the Study of Social Change.
Cockcroft, Eva (1974). “Abstract Expressionism: Weapon of the Cold War.”
Artforum (June): 39-41.
Cohen, Selma Jeanne (1959). “Donald McKayle, Kevin Carlisle and Cos.”
Dance Magazine (July): 17, 80.
-, ed. (1965). The Modem Dance: Seven Statements of Belief. Middletown,
Conn.: Wesleyan University Press.
Coleman, Martha (1948). “Talley Beatty, Nina Fonaroff, Yuriko.” Dance Ob-
semer (January): 7.
Collins, Patricia Hill (2004). Black Sexual Politics: African-Americans, Gender,
and the New Racism. New York: Routledge.
222 References

Cooper, John Milton (1992). Pivotal Decades: The United States, igoo-igio.
New York: W. W. Norton.
Copeland, Roger (1973-1974). “A Conversation with Alwin Nikolais.” Dance
Scope 8, I (FallAVinter): 41-46.
- (2004). Merce Cunningham: The Modernizing of Modem Dance. New
York: Routledge.
Corn, Wanda (1985). Grant Wood: The Regionalist Vision. New Haven, Conn.:
Yale University Press.
Cornfield, Robert, and William MacKay, eds. (1986). Edwin Denby: Dance
Writings. New York: Alfred A. Knopf.
Craig, Jenifer (2001). “A Redacted Past: Bella Lewitzky Encounters the U.S.
Government.” Paper presented at the twenty-fourth annual conference of
the Society of Dance History Scholars, 21-24 June. Goucher College,
Baltimore.
Croce, Arlene (1977). Afterimages. New York: Alfred A. Knopf
-(1982). Going to the Dance. New York: Alfred A. Knopf.
Cunningham, Merce (1951). “The Function of a Technique for Dance.” In
The Dance Has Many Faces, edited by Walter Sorell. New York: World
Publishing.
-([1955] (1978). “The Impermanent Art.” In Esthetics Contemporary, ed¬
ited by Richard Kostelanetz. Buffalo N.Y.: Prometheus Books. Originally
published in y Arts, 3, edited by Fernando Puma. Indian Hills, Conn.: Fal¬
con Wing’s Press.
-(1966). “Summerspace Story: How a Dance Came to Be.” Dance Mag¬
azine (June): 52-54.
—-(1968). Changes: Notes on Choreography. Edited by Frances Starr. New
York: Something Else Press.
-(1980). “Choreography and the Dance.” In The Dance Anthology, edited
by Cobbett Steinberg. New York: New American Library.
Cunningham, Merce, with Jacqueline Lesschaeve (1985). The Dancer and the
Dance. New York: Marion Boyars.
Dalfiume, Richard M. (1968). “The ‘Forgotten Years’ of the Negro Revolution.”
Journal of American History 15 (June): 90-106.
Daly, Ann (1987). “The Balanchine Woman: Of Hummingbirds and Channel
Swimmers.” The Drama Review 31,1 (Spring): 8-21.
Damasio, Antonio (2003). Looking for Spinoza: Joy, Sorrow, and the Feeling Brain.
New York: Harcourt.
Dana, Margery (1940). “The Dance.” Daily Worker (21 February): 7.
Davis, Luther, and John Cleveland (1940). “Russian Genius.” Collier’s (28 De¬
cember): 22, 36.
DeFrantz, Thomas F, ed. (2002). Dancing Many Drums: Excavations in African
American Dance. Madison: University of Wisconsin Press.
References 223

-(2004). Dancing Revelations: Alvin Ailey's Dmhodiment of African Aweri-


can Culture. Oxford: Oxford University Press.
D’Emilio, John (1998). Sexual Politics, Sexual Communities: The Making of a
Homosexual Minority in the United States, 1940-/970. 2nd ed. Chicago: Uni¬
versity of Chicago Press.
D’Emilio, John, and Estelle Ereedman (1997). Intimate Matters: A History of
Sexuality in America. 2nd ed. Chicago: University of Chicago Press.
Denhy, Edwin ([1937] 1986). “Balanchine’s American Ballet.” In Dance Writ¬
ings, edited by Robert Cornfield and William MacKay. New York: Alfred A.
Knopf: 44-46. Originally published in Modem Music (March-April).
-([1942] 1986). “Eokine’s ‘Russian Soldier’; Tudor’s ‘Pillar of Fire’; Bal¬
anchine’s Elephant Ballet.” In Dance Writings, edited by Robert Cornfield
and William MacKay. New York: Alfi'ed A. Knopf, 92-95. Originally pub¬
lished in Modef-n Music (May-June 1942).
■--—([1943a] 1986). “Katherine Dunham.” In Dance Writings, edited by
Robert Cornfield and William MacKay. New York: Alfred A. Knopf, 142-
143. Originally published in the Hew York Herald Tribune (26 September).
-([1943b] 1986). “Tudor and Pantomine.” In Dance Writings, edited by
Robert Cornfield and William MacKay. New York: Alfred A. Knopf, 129-
131. Originally published in the New York Herald Tribune (i i July).
-([1944a] 1986). “A Note on Dance Intelligence.” In Dance Writings,
edited by Robert Cornfield and William MacKay. New York: Alfred A.
Knopf, 204-205. Originally published in the New York Herald Tribune (26
March).
-([1944b] 1986). “Merce Cunningham.” In Dance Writings, edited by
Robert Cornfield and William MacKay. New York: Alfred A. K/iopf, 207-
208. Originally published in the Necw York Herald Tribune (6 April).
-([19ZI4C] 1986). “Pearl Primus on Broadway.” In Dance Writings, edited
by Robert Cornfield and William MacKay. New York: Alfred A. Knopf,
247-248. Originally published in the New York Herald Tribune (5 October).
-(i944d). “The Dance: Dunham in Full Bloom.” New York Herald Tri¬
bune (27 December): 10.
-([1945] 1986). “Cunningham Solo.” In Dance Writings, edited by Robert
Cornfield and William MacKay. New York: Alfred A. Knopf, 279—280.
Originally published in the New York Herald Tribune (10 January).
-([1946] 1986). “The Four Temperaments.” In Dance Writings, edited by
Robert Cornfield and William MacKay. New York: Alfred A. Knopf, 414-
415. Originally published in Dance News (December 1946).
_([1947] 1986). “Ballet: The American Position.” In Dance Writings, ed¬
ited by Robert Cornfield and William MacKay. New York: Alfred A. Knopf,
507-517. Originally published in Town and Country (April 1947).
_([1948] 1986). “A Briefing in American Ballet.” In Dance Writings, edited
2 24 References

by Robert Cornfield and William MacKay. New York: Alfred A. Knopf,


^ly—527. Originally published in Kenyon Review (Autumn 1948).
-([1949a] 1986). “Against Meaning in Ballet.” In Dance Writings, edited
by Robert Cornfield and William MacKay. New York: Alfred A. Knopf,
527-531. Originally published in Ballet (March 1949)-
-(1949b). Looking at the Dance. New York: Pellegrini & Cudahy.
-([1952] 1986). “A Letter on New York City’s Ballet.” In Dance Writings,
edited by Robert Cornfield and William MacKay. New York: Alfred A.
Knopf, 415-430. Originally published in Ballet (August 1952).
-([1953] 1986). “Some Thoughts about Classicism and George Balan¬
chine.” In Dance Writings, edited by Robert Cornfield and William MacKay.
New York: Alfred A. Knopf, 433-440. Originally published in Dance Maga¬
zine (February 1953).
-(1965). Dancers, Buildings and People in the Streets. New York: Popular
Library.
Denning, Michael (1996). The Cultural Front. London: Verso.
Dennis, James M. (1998). Renegade Regionalists: The Modem Independence of
Grant Wood, Thomas Hart Benton and John Steuart Curry. Madison: Univer¬
sity of Wisconsin Press.
Desmond, Jane, ed. (2001). Dancing Desires: Choreographing Sexualities On and
Ojf the Stage. Madison: University of Wisconsin Press.
Diggins, John Patrick (1988). The Proud Decades: America in War and Peace,
ip4i-ig6o. New York: W. W. Norton.
“Doom Eager Dance” (1944). Newsweek (17 January): 85.
Doss, Erika (1991). Benton, Pollock, and the Politics of Modernism from Regional¬
ism to Abstract Expressionism. Chicago: University of Chicago Press.
Drake, St. Clair (1981). Foreward to Katherine Dunham: Reflections on the Social
and Political Contexts of Afro-American Dance, by Joyce Aschenbrenner. Dance
Research Annual XU, edited by Patricia A. Rowe. New York: Congress on
Research in Dance.
Drown, Marilyn Vaughan (1997). “Merce Cunningham and Meaning: The
Zen Connection.” Choreography and Dance 4, 3: 17-28.
Dudziak, Mary L. (2000). Cold War Civil Rights: Race and the Image of American
Democracy. Princeton, N.J.: Princeton University Press.
Duell, Daniel, et al. (1987). “Celebrating The Four Temperaments—i.” Ballet Re¬
view (Winter): 12-35.
Dunham, Katherine ([n.d.] 1978). “Statement.” In KAISO! Katherine Dunham,
an Anthology of Writings, edited by Veve A. Clark and Margaret B. Wilker-
son. Berkeley: University of California, Institute for the Study of Social
Change.
-([1941a] 1978). “The Negro Dance.” In KAISO! Katherine Dunham, an
Anthology of Writings, edited by Veve A. Clark and Margaret B. Wilkerson.
References 225

Berkeley: University of California, Institute for the Study of Social Change,


66-74. Originally published in The Negro Caravan, edited by Sterling A.
Brown, Arthur P. Davis, and Ulysses Lee. New York: Dryden Press.
---— ([1941b] 1978). “Thesis Turned Broadway.” In KAISO! Katherine Dun¬
ham, an Anthology of Writings, edited by Veve A. Clark and Margaret B.
Wilkerson. Berkeley: University of California, Institute for the Study of So¬
cial Change, 55-57. Originally published in California Arts and Architecture
(August).
-([1941c] 1978). “Form and Function in Primitive Dance.” In KAISO!
Katherine Dunham, an Anthology of Writings, edited by Veve A. Clark and
Margaret B. Wilkerson. Berkeley: University of California, Institute for the
Study of Social Change, 192-196. Originally published in Educational Dance
(October).
-([1951] 1978). “Program: Southland." In KAISO! Katherine Dunham, an
Anthology of Writings, edited by Veve A. Clark and Margaret B. Wilkerson.
Berkeley: University of California Press, 117-120.
-(1994). A Touch of Innocence. Chicago: University of Chicago Press.
Dunning, Jennifer (1985). But First a School: The First Fifty Years of the School of
American Ballet. New York: Viking Press.
-(1996). Alvifi Alley: A Life in Dance. New York: Addison-Wesley.
Emery, Lynne Fauley (1988). Black Dance from i6ig to Today. 2nd ed. Penning¬
ton, N.J.: Princeton Book.
Erdman, Jean (1948). “Young Dancers State Their Views: As Told to Joseph
Campbell.” Dance Observer (April): 40-41.
-(1994)- The New Dance Group Gala Concert (video). Interview with Jean
Erdman. American Dance Guild Production.
Fanon, Frantz (1967). Black Skin, White Masks. New York: Grove Press.
Feuer, Jane (1993). The Hollywood Musical. 2nd ed. Basingstoke, U.K.: Macmil¬
lan Press.
Fonaroff, Nina (1948). “Young Dancers State Their Views: As Told to Mary
Phelps.” Dance Observer (May): 52, 63.
Foner, Eric (1998). The Story of American Freedom-. New York: W. W. Norton.
Eoote, Horton (1944). “The Long, Long Trek.” Dance Observer (October):
98-99.
-(1945a). “Pearl Primus and Group.” Dance Observer (April): 43-44.
-(1945b). “Notes for the Euture.” Dance Observer (May): 56.
Eoster, Susan Leigh (1986). Reading Dancing: Bodies and Subjects in Contemporary
American Dance. Berkeley: University of California Press.
-, ed. (1995). Choreographing History. Bloomington: University of Indiana
Press.
_(2001). “Closets Pull of Dances: Modern Dance’s Performance of Mas¬
culinity and Sexuality.” In Dancing Desires: Choreographing Sexualities On and
226 References

Off the Stage, edited by Jane Desmond. Madison: University of Wisconsin


Press.
Foulkes, Julia L. (2002). Modem Bodies: Dance and American Modernism from
Martha Graham to Alvin Ailey. Chapel Hill: Uruversity of North Carolina
Press.
Franko, Mark (1995). DancingModemism/Petforming Politics. Bloomington: Uni¬
versity of Indiana Press.
-(2002). The Work of Dance: Labor, Movement, and Identity in the igyos.
Middletown, Conn.: Wesleyan University Press.
Frascina, Francis, ed. (1985). Pollock and After: The Critical Debate. London: Paul
Chapman Publishing.
Frascina, Francis, and Charles Harrison, eds. (1987). Modem Art and Mod¬
ernism: A Critical Anthology. New York: Harper and Row.
Fraser, Steve, and Gary Gerstle, eds. (1990). The Rise and Fall of the New Deal
Order, ipyo-1980. Princeton, N.J.: Princeton University Press.
Freedley, George (1944). “Pearl Primus—in Two Parts.” New York Morning Tele¬
graph (10 October): 2.
Fried, Richard (1990). Nightmare in Red: The McCarthy Era in Perspective. New
York: Oxford University Press.
Fromm, Erich (1947). Man for Himself: An Inquiry Into the Psychology of Ethics.
New York: Rinehart.
-(1950)- Psychoanalysis and Religion. New Haven: Yale University Press.
-(1955). The Sane Society. New York: Rinehart & Co.
Gaddis, John (1972). The United States and the Origins of the Cold War, 1941-
ip4'j. New York: Columbia University Press.
-(1997). We Now Know: Rethinking Cold War History. Oxford: Oxford
University Press.
Garafola, Lynn (1989). Diaghilev’s Ballets Russes. Oxford: Oxford University Press.
-(2002). “Dollars for Dance: Lincoln Kirstein, City Center, and the
Rockefeller Foundation.” Dance Chronicle 25, i: 101-114.
Garafola, Lynn, with Eric Eoner, eds. (1999). Dance for a City: Fifty Years of the
New York City Ballet. New York: Columbia University Press.
Garcia-Marquez, Vicente (1990). The Ballets Russes: Colonel de Basil's Ballets
Russes de Monte Carlo, ipyi-ipyz. New York: Alfred A. Knopf.
Gellhorn, Martha (1947). “Cry Shame .. . !” New Republic (6 October): 20-21.
Giddings, Paula (1992). “The Last Taboo.” In Race-ing Justice, En-gendering
Power: Essays on Anita Hill, Clarence Thomas and the Construction of Social Re¬
ality, edited by Toni Morrison. New York: Pantheon Books.
Gifford, Joseph (1945). “Smoke Gets in Our Eyes.” Dance Observer (June-
July): 65-66.
Gilbert, James (1986). A Cycle of Outrage: America's Reaction to the Juvenile
Delinquent in the ipyos. New York: Oxford University Press.
References 227

Gillmor, Daniel S. (1948). “Guilt by Gossip.” New Republic (21 May): 15-19.
Gilman, Sander L. (1985). Dijference and Pathology: Stereotypes of Sexuality, Race
and Madness. Ithaca, N.Y.: Cornell University Press.
Gitelman, Claudia (2000). “From Bauhaus to Playhouse: Tracing the Aesthetic
of Alwin Nikolais.” Conference proceedings. Dancing in the Millennium
(19-23 July), Washington, D.C.
Goff, Eleanor Anne (1946). “Pearl Primus.” Dance Observer (June-July): 76-77.
Goffinan, Erving (1959). The Presentation of Self in Everyday Life. New York:
Doubleday.
-— (1963). Behavior in Public: Notes on the Social Organization of Gatherings.
New York: Eree Press.
-(1979). Gender Advertisements. Cambridge, Mass.: Elarvard University
Press.
Goodman, Paul (i960). Growing Up Absurd: Problems of Youth in the Organized
Society. New York: Random House.
Gottschild, Brenda Dixon [Stowell] (1988). “Black Dance and Dancers and the
White Public: A Prolegomenon to Problems of Definition.” In The Black
Tradition in American Modern Dance, Gerald E. Myers and Stephanie Rein¬
hart, project directors. Durham, N.C.: American Dance Festival.
-— (1996). Digging the Africanist Presence in American Performance: Dance and
Other Contexts. Westport, Conn.: Praeger.
Goulden, Joseph (1975). The Best Years: 1945-7950. New York: Scribner.
Graff, Ellen (1985). “The Four Temperaments and Orpheus: Models of a Modern
Classical Tradition.” Ballet Review 3 (Eall): 54-59.
-(1997). Stepping Left: Dance and Politics in New York City, ipiS-ip^z.
Durham, N.C.: Duke University Press.
Graham, Martha ([1937] 1985). “Graham 1937” and “Affirmations 1926-1937.”
In Manha Graham: The Early Years, edited by Merle Armitage. Reprint,
New York: Da Capo Press.
-([1941] 1980). “A Modem Dancer’s Primer for Action.” In The Dance
Anthology, edited by Cobbett Steinberg. New York: New American Library.
Originally published in Dance: A Basic Educational Technique, edited by Ered-
erick R. Rogers. New York: Macmillan.
-(1973)- The Notebooks of Martha Graham. New York: Harcourt, Brace
Jovanovich.
-(1991)- Blood Memory: An Autobiography. New York: Simon and Schuster.
Gramsci, Antonio (2000). The Antonio Gramsci Reader, edited by David Eorgacs.
New York: New York University Press.
Grauert, Ruth (1999). Unpublished interview conducted by Claudia Gitelman.
Typescript from audio recording. New Yirk: Collection of Claudia Gitelman.
Green, Richard C. (2002). “(Up)Staging the Primitive: Pearl Primus and ‘the
Negro Problem’ in American Dance.” In Dancing Many Drums: Excavations
228 References

in African American Dance, edited by Thomas F. DeFrantz. Madison: Uni¬


versity of Wisconsin Press.
Greenberg, Clement ([1939] 1986a). “Avant-Garde and Kitsch.” In Clement
Greenberg: The Collected Essays and Criticism, vol. i, edited by John O’Brian.
Chicago: University of Chicago Press. Originally published in Partisan Re¬
view (Fall 1939).
—-([1940] 1986b). “Towards a Newer Laocoon.” In Clement Greenberg:
The Collected Essays and Criticism, vol. i, edited by John O’Brian. Chicago:
University of Chicago Press. Originally published in Partisan Review (July-
August 1940).
-([1945] 1986c). “Review of the Ballet Dim Luster by Antony Tudor.” In
Clement Greenberg: The Collected Essays and Criticism, vol. 2, edited by John
O’Brian. Chicago: University of Chicago Press. Originally published in
the Nation (3 November 1945).
-([1950] 1993a). “The European View of American Art.” In Clement
Greenberg: The Collected Essays and Criticism, vol. 3, edited by John O’Brian.
Chicago: University of Chicago Press. Originally published in the Nation
(25 November 1950).
-([i960] 1993b). “Modernist Painting.” In Clement Greenberg: The Col¬
lected Essays and Criticism, vol. 4, edited by John O’Brian. Chicago: Univer¬
sity of Chicago Press. Originally published as a pamphlet in Forum Lectures.
Washington, D.C.: Voice of America.
Greene, Jonnie (1997). “Classic Black.” Dance Magazine (February): 86-88.
Griffith, Robert (1981). “Harry S Truman and the Burden of Modernity.” Re¬
views in American History 9 (September): 295-306.
Guilbaut, Serge (1983). How New York Stole the Idea of Modem Art. Chicago:
University of Chicago Press.
-, ed. (1990). Reconstructing Modernism: Art in New York, Paris, and Mon¬
treal, iy4y~ig6^. Cambridge, Mass.: MIT Press.
Hamby, Alonzo (1976). Beyond the New Deal: Harry S Truman and American Lib¬
eralism. New York: Columbia University Press.
Hammonds, Evelynn M. “Toward a Genealogy of Black Female Sexuality: The
Problematic of Silence.” In Feminist Theory and the Body: A Reader, edited by
Janet Price and Margrit Shildrick. New York: Routledge.
Harrison, Charles, and Paul Wood (1993). Art in Theory: ipoo-iggo. Oxford:
Blackwell.
Haskins, James (1982). Katherine Dunham. New York: Coward, McCann and
Geoghegan.
-(1990). Black Dance in America: A History through Its People. New York:
HarperCollins.
Helpern, Alice (1991). “The Technique of Martha Graham.” Studies in Dance
History 2,2: 1-59.
References 229

Hering, Doris (1947). “The Workshop, A New Perennial.” Dance Magazine


(July): 19-21.
-(1948). “Reviewers’ Stand.” Dance Magazine (December): 10, 35.
-(1949). “The Season in Review: ist Season: New York City Ballet Com¬
pany.” Dance Magazine (January): 12-13, 32-34-
(1950)- The New York City Ballet Company Season.” Dance Magazine
January): 12, 47-48.
(1952). “Meet John Martin.” Dance Magazine (November): 20—23, 3^—
39. 45-
(i953^)' '"Jose Limon and Company.” Da?tce Magazine (February): 53—
54. 56, 58-
(1953^*)- “New York City Ballet.” Dance Magazine (March): 10, 60.
-(1955)- “An Evening of Dance Works by Anna Sokolow.” Dance Maga¬
zine (April): 75-77.
-(1956). “Geoffrey Holder and Louis Johnson.” Dance Magazine (Janu¬
ary): 76-78.
-(1958). “Anna Sokolow Dance Company.” Dance Magazine (April):
26-27.
-(1959). “Anna Sokolow Dance Company.” Dance Magazine (Febru¬
ary): 28.
-(i960). “Talley Beatty and Guest Performers.” Dance Magazine (Janu¬
ary): 80-81.
Higham, John (1989). “The Collapse of Consensus YUstoTy.” Journal of Amer¬
ica?! History (September): 460-466.
Hill, Constance Valis (1994). “Katherine Dunham’s Southland: Protest in the
Face of Repression.” Dance Research Journal 26, 2 (Fall): i-io.
Hill, Martha (1965). Martha Hill. Audio interview conducted by Billy Nichols
for National Education Network. Dance Collection, New York Public
Library.
-(1990). Martha Hill Video Project (video). Dance Collection, New York
Public Library.
Hinz, Berthold (1979). An in the Third Reich. Translated by Robert and Rita
Kimber. New York: Random House.
Hodson, Millicent (1978). “How She Began Her Beguine: Dunham’s Dance
Literacy.” In KAISO! Katherine Dunham, an Anthology of Writings, edited by
Veve A. Clark and Margaret B. WiLkerson. Berkeley: University of Califor¬
nia, Institute for the Study of Social Change, 196-198.
Hoelterhoff, Manuela (1975). “Art of the Third Reich: Documents of Oppres¬
sion.” Hrl/oram (December): 55-62.
Horosko, Marian (1991). Manha Graham: The Evolution of Her Dance Theory
and Training, ipzh-ippi. Chicago: A Cappella Books.
Horst, Louis (1940). Pre-Classic Dance Forms. New York: The Dance Observer.
230 References

-(1943)- “Katherine Dunham and Company.” Dance Observer (Octo¬


ber): 90.
-(1946). “Choreographers’ Workshop.” Dance Observer (D&cemhtr)\ 124.
-(1947)- “Choreographers’ Workshop.” Dance Observer (F&hraurf)-. 20.
-(1958a). “Anna Sokolow Dance Company.” Dance Observer (April): 55.
-(i958h). “Juilliard Dance Theater.” Dance Observer (June-July): 86.
-(1959)- “Juilliard Dance Theater, Geoffrey Holder & Co., and Midi
Garth.” Dance Ofoeryer (March): 39.
Horst, Louis, and Carroll Russell ([1961] 1987). Modem Dance Forms in Rela¬
tion to the Other Modem Arts. Reprint, Princeton, N.J.: Princeton Book Co.
Hurok, Sol, with Ruth Goode (1947). Impresario, a Memoir. London: Macdonald
and Co.
-(1953). S. Hurok Presents: A Memoir of the Dance World. New York: Her¬
mitage House.
Jackson, Naomi (2000). Converging Movements: Modem Dance and Jewish Cul¬
ture at the gind Street Y. Middletown, Conn.: Wesleyan University Press.
Jacoby, Russell (1987). The Last Intellectuals: American Culture in the Age of Acad¬
eme. New York: Basic Books.
Janson, H. W. (1946). “Benton and Wood, Champions of Regionalism.” Mag¬
azine of Art 39, 5 (May): 184-186, 198-200.
Jay, Martin (1996). The Dialectical Imagination: A History of the Frankfurt School
and the Institute of Social Research. Berkeley: University of California Press.
Jenkins, Nicholas (1998). “The Great Impresario.” Nm) Yorker (13 April): 48-61.
Jezer, Marty (1982). The Dark Ages: Life in the United States, ip^y-ip6o. Boston:
South End Press.
Joel, Lydia (1957). “Close-Up of Modern Dance Today: The Non-Objective
Choreographers.” Dance Magazine (November): 20-23.
“John Martin Is Dead at 91; Times Critic 35 Years” (1985). New York Times
(21 May): B6.
Johnson, Harriet (1948). “Talley Beatty and Company.” Dance Observer (No¬
vember): 122.
Johnston, Jill (1955). “Thoughts on the Present and Future Directions of
Modern Dance.” Dance Ofceryer (August-September): 101-102.
-(1957a). “The Modern Dance—Directions and Criticisms.” Dance Ob¬
server (April): 55-56.
-(1957b). “Abstraction in Dance.” Dance Observer (December): 151—152.
-(i960). “Talley Beatty, Louis Johnson, Ernest Parham.” Dance Observer
(June-July): 89.
-(1968). “Not in Broad Daylight.” Village Voice (21 November): 32-33.
Jones, Bill T, with Peggy Gillespie (1995). Last Night on Earth. New York: Pan¬
theon.
References 231

Jordan, Stephanie (2000). Moving Music: Dialogues with Music in Twentieth-


Century Ballet. London: Dance Books.
Jordan, Stephanie, and Helen Thomas (1994). “Dance and Gender: Formalism
and Semiotics Reconsidered.” Dance Research 12, 2 (Autumn): 3-14.
Jowitt, Dehorah (2004). Jerome Robbins: His Life, His Theater, His Dance. New
York: Simon and Schuster.
Jung, Carl (i9'7i). The Portable Jung. Edited, with an introduction hy Joseph
Campbell. New York: Viking Press.
Kinsey, Alfred, Warded B. Pomeroy, and Clyde E. Martin, (1948). Sexual Be¬
havior of the Human Male. Philadelphia: W. B. Saunders.
Kinsey, Alfred, et al. (1953). Sexual Behavior of the Human Female. Philadelphia:
W. B. Saunders.
Kirstein, Lincoln (1934a). “Revolutionary Ballet Forms.” New Theatre (Octo¬
ber): 12-14.
-([1934b] 1983). “Prejudice Purely.” In Ballet: Bias & Belief; “Three Pam¬
phlets Collected” and Other Dance Writings of Lincoln Kirstein. Introduction
and comments by Nancy Reynolds. New York: Dance Horizons. Originally
published in New Republic (it April 1934).
-([1937] 1983)- “Blast at Ballet.” In Ballet: Bias Belief “Three Pamphlets
Collected” and Other Dance Writings of Lincoln Kirstein. Introduction and
comments by Nancy Reynolds. Reprint, New York: Dance Horizons.
-(1939). Ballet Alphabet: A Primer for the Layman. New York: Kamin
Publishers.
-(1945)- “Art in the Third Reich—Survey 1945.” Magazine of An 28, 6
(October): 223-242.
-(1947a). “Balanchine Musagete.” Theatre Arts (November): 37-41.
-(1947b). “Balanchine and the Classic Revival.” Theatre Arts (Decem¬
ber): 37-41.
-(1948). “The State ofModern Painting.” Harper’s (October): 47-53.
-([1953] 1983). “Alec: Or the Future of Choreography.” In Ballet: Bias &
Belief; “Three Pamphlets Collected” and Other Dance Writings of Lincoln Kirstein.
Introduction and comments by Nancy Reynolds. New York: Dance Hori¬
zons. Originally published in Dance News Annual, vol. i, edited by Winthrop
Palmer and Anatole Chujoy. New York: Alfred A. Knopf.
-([1959] 1983). “VVfrat Ballet Is About: An American Glossary.” In Bal¬
let: Bias & Belief; “Three Pamphlets Collected” and Other Dance Writings of Lin¬
coln Kirstein. Introduction and comments by Nancy Reynolds. Reprint, New
York: Dance Horizons.
-(1970). Movement and Metaphor: Four Centuries of Ballet. New York:
Praeger.
-(1973)- The New York City Ballet. New York: Alfred A. Knopf.
232 References

-(1983). Ballet: Bias it Belief; “Three Pamphlets Collected” and Other Dance
Writings of Lincoln Kirstein. Introduction and comments by Nancy Reynolds.
Reprint, New York: Dance Horizons.
-(1991). By With To it From: A Lincoln Kirstein Reader, edited by Nicholas
Jenkins. New York: Farrar, Straus and Giroux.
-(1994). Mosaic: Memoirs. New York: Farrar, Straus and Giroux.
Klosty, James (1975). Merce Cunningham. New York: Saturday Review Press.
Kochno, Boris (1954). Le Ballet. Paris: Hachette.
Korff, William (1993). “History of the New Dance Group from the 1930s to
the 1970s.” In printed program for the video recording of The New Dance
Group Gala Concert. American Dance Guild Production. New York: Ameri¬
can Dance Guild.
Kostelanetz, Richard, ed. (1978). Esthetics Contemporary. Buffalo, N.Y.: Pro¬
metheus Books.
--—, ed. {\()gi).John Cage: An Anthology. New York: Da Capo Press.
-, ed. (1992). Merce Cunningham: Dancing in Space and Time. Chicago:
A Cappella Books.
Kowal, Rebekah (1999). Modem Dance and American Culture in the Early Cold
War Years. Ph.D. dissertation. New York University.
Krevitsky, Nik (1948). “Choreographers’ Workshop.” Dance Observer (May):
56-57-
--— (1949). “American Dance Festival.” Dance Observer (August-September):
98.
-(1951a). “Merce Cunningham and Company.” Dance Observer (bAurch):
41-42.
-(1951b). “Jean Erdman and Company.” Dance Observer (March): 42.
-(1951c). “Jane Dudley, Sophie Maslow, William Bales and Company.”
Dance Observer (March): 42-43.
Kriegsman, Sali Ann (1981). Modem Dance in America: The Bennington Years.
Boston: G. K. Hall.
Kuznick, Peter J., and James Gilbert, eds. (2001). Rethinking Cold War Cidture.
Washington, D.C.: Smithsonian Institution Press.
Laban, Juana de (1945). “What Tomorrow?” Dance Observer (May): 55-56.
Lacey, Michael, ed. (1990). The Truman Presidency. Cambridge: Cambridge
University Press.
Langer, Susanne (1953). Peeling and Form: A Theory of Art. New York: Scribner.
Laqueur, Thomas (1987). “The Politics of Reproductive Biology.” In The
Making of the Modern Body: Sexuality and Society in the Nineteenth Century,
edited by Thomas Laqueur and Catherine Gallagher. Berkeley: University
of California Press.
Larkey, Joan (1949). Jean Erdman and Company.” Dance Observer (December):
151-152.
References 233

Lash, Scott, and Sam Whimster, eds. (1987). Max Weber, Rationality andModer-
nity. London: Allen and Unwin.
Lawson, Steven F. (1976). Black Ballots: Voting Rights in the South, 1944-1969.
New York: Columbia University Press.
Lears, Jackson (1989). “A Matter of Taste: Corporate Cultural Hegemony in
a Mass-Consumption Society.” In Recasting America: Culture ajid Politics in
the Age of Cold War, edited by Lary May. Chicago: University of Chicago
Press.
LeClercq, Tanaquil (1982). “Tanaquil LeClercq.” In Striking a Balance: Dancers
Talk about Dancing, edited by Barbara Newman. Boston: Houghton Mifflin.
Lehmann-Haupt, Hellmut (1954).^^ Under a Dictatorship. New York: Oxford
University Press.
Leuchtenburg, William (1973). A Troubled Feast: American Society since 1949.
Boston: Little, Brown.
Limon, Jose (1951). “The Virile Dance.” In The Dance Has Many Faces, edited
by Walter Sorell. New York: World Publishing.
-(1999)- Jose Limon: An Unfinished Life. Middletown, Conn.: Wesleyan
University Press.
Lippincott, Gertrude (1946). “Pilgrim’s Way.” Dance Observer (August-
September): 84-85.
-(1948). “Freedom and the Arts.” Dance Observer (April): 44-45.
Lipsitz, George (1981). Class and Culture in Post War America: A Rainbow at
Midnight. New York: Praeger.
-(1989). “Land of a Thousand Dances: Youth, Minorities, and the Rise
of Rock and Roll.” In Recasting America: Culture and Politics in the Age of Cold
War, edited by Lary May. Chicago: University of Chicago Press.
Litz, Katherine (1949). “Young Dancers State Their Views: As Told to Nik
Krevitsky.” Dance Observer (May): 67.
Lloyd, Margaret (1949). The Borzoi Book of Modem Dance. New York: Alfred A.
Knopf.
Louis, Murray (i960). “The Contemporarry Dance Theatre of Alwin Nikolais.”
Dance Observer (January): 5-6.
-(1973). Audio interview conducted by Tobi Tobias. Dance Collection,
New York Public Library.
-(1980). Inside Dance: Essays. New York: St. Martin’s Press.
Lowe, Roberta (1943). “Dunham Dancers Fvoke Applause in B’way Opening.”
Amsterdam News (25 September): 9-B.
Macdonald, Dwight (1962). Against the American Grain. New York: Random
House.
MacKay, William (1986). “Edwin Denby, 1903-1983.” In Edwin Denby: Dance
Writings, edited by Robert Cornfield and William MacKay. New York: Al¬
fred A. Knopf.
234 References

Manchester, P. W. (1958). “Anna Sokolow Dance Company.” Dance News


(April): 9.
-(1959a). “Juilliard Dance Theatre, Midi Garth, Geoffrey Holder &
Co.” Dance News (March): 7.
-(1959b). “Donald McKayle, Kevin Carlisle and Dance Companies at
YM-YWHA.” Dance News (June): 14.
-(1960a). “Talley Beatty.” Dance News (January): 7.
-(1960b). “Talley Beatty, Louis Johnson, Earnest Parham.” Dance News
(March): lo-ii.
Manning, Susan (1993). Ecstasy and the Demon: Feminism and Nationalism in the
Dances of Mary Wigman. Berkeley: University of California Press.
-(2004). Modem Dance, Negro Dance: Race in Motion. Minneapolis: Uni¬
versity of Minnesota Press.
Marcuse, Herbert ([1955] 1974)- Eros and Civilization: A Philosophical Inquiry
into Freud. Reprint, Boston: Beacon Press.
-([1964] 1991). One-Dimensional Man. Reprint, Boston: Beacon Press.
-(1968). Negations: Essays in Critical Theory. Boston: Beacon Press.
-(1978). The Aesthetic Dimension: Toward a Critique of Marxist Aesthetics.
Boston: Beacon Press.
Martin, John ([1933] 1972). The Modem Dance. Reprint, Princeton N.J.: Dance
Horizons.
-(1936). America Dancing: The Background and Personalities of the Modem
Dance. New York: Dodge Publishing Co.
-(1937). “The Dance: New Ballets.” New York Times (2 May): X 7.
-([1939] 1965). Introduction to the Dance. Reprint, Princeton N.J.: Dance
Horizons.
-(1940a). “Negro Dance Art Shown in Recital.” New York Times (19 Feb¬
ruary): 23.
-(1940b). “The Dance: A Negro Art.” New York Times (25 February):
114.
-(1943^)- *”The Dance: Five Artists” New York Times (21 February): X 5.
-(1943b). “Katherine Dunham Gives Dance Revue.” New York Times
(20 September): 24.
-(1944^)- “New Voodoo Dance in Dunham Revue.” New York Times
(27 December): 15.
-(1944b). “Brilliant Dancing by Pearl Primus.” New York Times (5 Octo¬
ber): 18.
-(1945)- The Dance. New York: Tudor Publishing Co.
-(1949)- “The Dance: Newcomer.” New York Times (27 February): X 9.
(1950). “Maslow’s Dances Given at Festival.” New York Times (19 Au¬
gust): 8.
(1952)- The World Book of Modem Ballet. Cleveland: World Publishing.
References 235

(1953)- “The Dance; Diagnosis, A Few Notes Concerning the Strength


as Well as Weakness in Modern Field.” New York Times (June 28): X 3.
-(19553)- “Dance: Study in Despair.” New York Times (16 May): 26.
-(1955b)- “The Dance: Modern, Time for Serious Study of Basic The¬
ories.” New York Times (5 June ): X 8.
-(1956a). “The Dance: New Life.” New York Times (19 February): X 9.
-(1956b). “Fhe Dance: Bright Augury.” New York Times (26 February):
X6.
-(1959). “Dance: Good Job.” New York Times (6 December): X 18.
-John Martin’s Book of the Dance. New York: Tudor Publishing.
-(1989). John Martin: The Dance in Theory. Princeton, N.J.: Dance
Horizons.
Martin, Richard (1987). Fashion, and Surrealism. New York: Rizzoli.
Maslow, Sophie (1984). Dance and Social Consciousness in the lyyos and fos
(video). Interview with Sophie Maslow and Anna Sokolow. Produced by
ARC Videodance as part of the television series Eye on Dance, WNYC, New
York. Dance Collection, New York Public Library.
Mauss, Marcel ([1934] 1973). “Techniques of the Body.” Translated by Ben
Brewster. Economy and Society, 2, i (February): 70-88.
May, Elaine Tyler (1988). Homeward Bound: American Families in the Cold War
Era. New York: Basic Books.
May, Lary, ed. (1989). Recasting America: Culture and Politics in the Age of Cold
War. Chicago: University of Chicago Press.
McDonagh, Don (1970). The Rise and Tail and Rise of Modem Dance. New York:
New American Library.
-(1973). Martha Gi'aham, a Biography. New York: Praeger.
McKayle, Donald (n.d.). Draft for an autobiography, I Will Dance with You.
Donald McKayle Papers, Special Collections and Archives, UCI Libraries,
Irvine, California.
-(1965). “Donald McKayle.” In The Modem Dance: Seven Statements of
Belief, edited by Sehna Jeanne Cohen. Adiddletown, Comi.: Wesleyan Uni¬
versity Press.
-(1966). “The Negro Dancer in Our Time.” In The Dance Has Many
Eaces, edited by Walter Sorell. 2nd ed. New York: Columbia University Press.
-(1993)- Speaking of Dance: Donald McKayle (video). Interview with Don¬
ald McKayle. American Dance Festival production, directed and produced
by Douglas Rosenberg. Dance Collection, New York Public Library.
_(2000). Images and Reflections: Celebration of a Masterpiece, Donald Mc¬
Kay le’s Rainbow Round My Shoulder (video). American Dance Legacy In¬
stitute Production, American Dance Legacy Institute.
-(2002). Transcending Boundaries: My Dancing Life. London: Routledge.
Meldon, John (1943)- “The Dance.” Daily Worker September): 7.
236 References

Mercer, Kobena (1994). Welmne to the Jungle: New Positions in Black Cultural
Studies. London: Roudedge.
Mills, C. Wright (1951). White Collar: The American Middle Classes. London: Ox¬
ford University Press.
-(1956). The Power Elite. London: Oxford University Press.
-(1959)- The Sociological Imagination. London: Oxford University Press.
Morris, Gay, ed. (1996). Moving Words: Re-writing Dance. London: Roudedge.
-(2001). “What He Called Himself: Issues of Identity in Early Dances
by Bill T. Jones.” In Dancing Desires: Choreographing Sexualities On and Off
the Stage, edited byJane Desmond. Madison: University of Wisconsin Press.
Moulton, Robert D. (1957). Choreography in Musical Comedy and Revue on the
New York Stage from 1925 to ipyo. Alicrofilm, New York Public Library.
Myers, Gerald E. (1988). “Ethnic and Modern Dance.” In The Black Tradition
in American Modem Dance, Gerald E. Myers and Stephanie Reinhart, project
directors. Durham, N.C.: American Dance Eestival.
Myers, Gerald E., and Stephanie Reinhart, project directors (1988). The Black
Tradition in American Modern Dance. Durham, N.C.: American Dance
Eestival.
-African American Genius in Modem Dance. Durham, N.C.: Amer¬
ican Dance Eestival.
Myrdal, Gunnar ([1944] 1999). An American Dilemma: The Negro Problem and
Modem Democracy. Reprint, London: Transaction Publishers.
Nabokov, Ivan, and Elizabeth Carmichael (1961). “Balanchine: An Interview
by Ivan Nabokov and Elizabeth Carmichael.” Horizon (January): 44-56.
Nadeau, Maurice (1989). The History of Surrealism. Cambridge, Mass.: Harvard
University Press.
Nash, Joe (1992). “Talley Beatty.” \n African American Genius in Modem Dance,
Gerald E. Myers and Stephanie Reinhart, project directors. Durham, N.C.:
American Dance Eestival.
Navasky, Victor (1980). Naming Names. New York: Viking Press.
Nettleton, Sarah, and Jonathan Watson, eds. (1998). The Body in Everyday Life.
London: Routledge.
Newman, Barbara (1982). Striking a Balance: Dancers Talk about Dancing. Boston:
Houghton Mifflin.
Nikolais, Alwin (1957). “Dance: Semantics.” New York Times (18 August): 31.
-(1958). “The New Dimension of Dance.” Impulse: 43-46.
-(1961a). “Dance, Sexual Dynamics in Contemporary.” In The Encyclo¬
pedia of Sexual Behavior, vol. i. New York: Hawthorn Books.
(1961b). Growth of a Theme.” Dance Magazine (Eebruary): 30—34.
(1965). No Man from Mars.” The Modem Dance: Seven Statements of
Belief, edited by Selma Jeamie Cohen. Aliddleton, Conn.: Wesleyan Univer¬
sity Press.
References 237

(1968). Lecture on Avant-Garde Dance.” Aural Recording. Dance


Collection, New York Public Library of the Performing Arts.
-(1971a). Nik~A Docu?nentary. Edited, with an introduction by Marcia B.
Siegel. Dance Perspectives: 48.
(1971b). Interview with Alwin Nikolais.” Elinor Rogosin, interviewer.
Aural recording. Dance Focus, Radio Station WBAI, New York. Dance Col¬
lection, New York Public Library of the Performing Arts.
-(^973)- “Interview with Alwin Nikolais,” James Day, interviewer. Audio
portion of television broadcast Day at Night. WNET/Channel 13, New
York. Dance Collection, New York Public Library of the Performing Arts.
-(1973-74). “A Conversation with Alwin Nikolais,” by Roger Copeland.
Dance Scope, 8, i (Pall-Winter): 41-46.
Oakley, J. Ronald (1986). God’s Country: America in the Fifties. New York: Dem-
bner Brooks.
Ocko, Edna (1948). “Dance: Anna Sokolow.” Masses & Mainstream i, i (March):
95-96.
O’Donnell, Mary P. (1940). “Katherine Dunham and Group.” Dance Observer
(December): 148.
O’Neill, William (1986). American High: The Year's of Confidence, ip4y-ip6o.
New York: Eree Press.
Orme, Prederick L. (1938). “The Negro in the Dance: As Katherine Dunham
Sees Him.” The American Daticer (March) 10, 46.
Ortner, Sherry (1974). “Is Pemale to Male as Nature Is to Culture?” In Woman,
Culture & Society, edited by Michelle Rosaldo and Louise Lamphere. Palo
Alto, Calif.: Stanford University Press.
Palatsky, Eugene (1958). “Backstage at New York City Ballet.” Dance Magazine
(December): 64-67, loi.
Palmer, Winthrop (1949). “Structure and Porms of Writing for the Theatre.”
Dance Observer {Octohe.v)\ 112.
Patterson, James T, ed. (1975). Paths to the Present. Minneapolis: Burgess
Publishing.
-(1996). Grand Expectations: The United States, 1945—1974. New York:
Oxford University Press.
Pells, Richard H. (1989). The Liberal Mind in a Conservative Age: American In¬
tellectuals in the ip^os and igyos. Middletovm, Conn.: Wesleyan University
Press.
Perlmutter, Donna (1991). Shadowplay: The Life of Antony Tudor. New York:
Viking Press.
Perloff, Majorie (1989). Postmodern Genres. Norman: University of Oklahoma
Press.
Perpener, John O. (2001). African American Concert Dance: The Harlem Renais¬
sance and Beyond. Urbana: University of Illinois Press.
238 References

Phelps, Mary (1946a). “Pearl Primus.” Dance Observer (February): 24.


-(1946b). “Dancers or Donkeys?” Dance Observer (November): i lo-i 11.
--— (1947). “Choreographers’ Workshop.” Dance Observer (June-July): 65.
-(1948). “Valerie Bettis and Company.” Dance Observer (February): 19.
Pierre, Dorathi Bock (1941). “A Talk with Katherine Dunham.” Educational
Dance (August-September): 7-8.
-(1947)- “Katherine Dunham: Cool Scientist or Sultry Performer?”
Dance Magazine (May): i i-i 3.
Polenberg, Richard (1991). One Nation Divisible: Class, Race, and Ethnicity in the
United States since 1938. New York: Viking Press.
Preston-Dunlop, Valerie (1995). Dance Words. Chur, Switzerland: Harwood
Academic Publishers.
Prevots, Naima (1985). “Zurich Dada and Dance: Formative Ferment.” Dance
Research Journal 17, i (Spring-Summer): 3-8.
-(1998). Dance for Export: Cultural Diplomacy and the Cold War Middle-
town, Conn.: Wesleyan University Press.
Prickett, Stacey Lee (1992). Marxism, Modernism and Realism: Politics and Aes¬
thetics in the Rise of the American Modem Dance. Ph.D. dissertation. The
Laban Centre for Movement and Dance, London.
Revill, David (1992). The Roaring Silence: John Cage; A Life. New York: Arcade
Publishing.
Reynolds, Nancy (1977). Repertory in Review: 40 Years of New York City Ballet.
New York: Dial Press.
-(1983). Introduction and comments. In Ballet: Bias & Belief; ‘’"Three
Pamphlets Collected” and Other Dance Writings of Lincoln Kirstein. New York:
Dance Horizons.
Richter, Hans (1965). Dada: Art and Anti-Art. New York: McGraw-Hill.
Riesman, David (1950). The Lonely Crowd. New Haven, Conn.: Yale University
Press.
Robinson, Harlow (1994). The Last Impresario: The Life, Times, and Legacy of Sol
Hurok. New York: Viking Press.
Rosaldo, Michelle, and Louise Lamphere, eds. (1974). Woman, Culture & So¬
ciety. Palo Alto, Calif.: Stanford University Press.
Rosenbaum, Ruth (1943). “Dancers’ Debit.” Dance Observer (August): 78.
Rouzeau, Edgar T. (1940). “A Word about Katherine Dunham.” Norfolk Jour¬
nal and Guide (9 March): 17.
Rubin, William (1968). Dada, Surrealism, and Their Heritage. New York: Mu¬
seum of Modern Art.
Rudko, Doris (1956). “The Playhouse Dance Company.” Dance Observer (Feb¬
ruary): 28.
Runciman,W. G., ed. (1978). Weber: Selections in Translation. Cambridge: Cam¬
bridge University Press.
References 239

Sabin, Robert (1944). Ballet Russe de Monte Carlo.” Dance Observer (Octo¬
ber): lOI.
(1945^)- '^Valerie Bettis and Group.” Dance Ofoeryer (June—July): 70.
(^945^*)- ^'Ballet Russe de Monte Carlo.” Dance Observer (October):
97-98.
-(1946). “Dance Critics in America: John Martin of The New York Times. ”
Dance Ofoeryer (June-July): 72—73.
-(1950). “Choreographers’ Workshop.” Dance Observer (Msy): 71.
-(1952). “Janet Collins.” Dance (January): lo-ii.
Saunders, Frances Stonor (1999). The Cultural Cold War: The CIA and the World
of Arts and Letters. New York: New Press.
Sawin, Martica (1997). Surrealism in Exile, and the Beginning of the New York
School. Cambridge, Mass.: MIT Press.
Schlesinger, Arthur M., Jr. ([1949] 1998). The Vital Center: The Politics of Free¬
dom. With a new introduction by the author. New Brunswick, N.J.: Trans¬
action Publishers.
Severin, Reed (1948). “The Ballet Society.” Dance Magazine (March): 7, 42-46.
“The Shape of Things” (1947). Nation (29 March): 345-347.
“She is Priestess of Intellectual Ballet” (1947). Life (17 March): 101-104.
Siegel, Frederick (1984). Troubled Journey: From Pearl Harbor to Ronald Reagan.
New York: Flill and Wang.
Siegel, Marcia B., ed. (1971). Nik—A Documentaty. With an introductory essay.
Dance Perspectives-. 48.
-(1979). The Shapes of Change: Images of American Dance. Boston: Hough¬
ton Mifflin.
-(1987). Days on Earth: The Dance of Doris Humphrey. New Haven, Corm.:
Yale University Press.
-(1997). “Using Lexicons for Performance Research: Three Duets.” New
Approaches to Theatre Studies and Performance Analysis: Papers Presented at the
Colston Symposium, Bristol, England, ii-zy March. Tubingen: MaxNiemeyer
Verlag.
Silk, Mark (1988). Spiritual Politics: Religion and America since World War 11. New
York: Simon and Schuster.
Simon, Henry (1943). “The Tropics Invade Broadway.” PM (20 September):
22.
Sitkoff, Harvard (1993). The Struggle for Black Equality, ipyq-igpz. New York:
Hill and Wang.
Skolnick, Arlene (1991). Embattled Paradise: The American Family in an Age of
Uncertainty. New York: HarperCollins.
Smith, Cecil (1947). “The Maze of the Heart.” Theatre Arts (May): 29-32.
-(1948). “Talley Beatty and Company.” Musical America (15 Novem¬
ber): 10.
240 References

Soares, Janet (1992). Louis Horst: Musician in a Dancer’s World. Durham, N.C.:
Duke University Press.
Sokolow, Anna (1965). “I Hate Academies ...” Dance Magazine (July): 38-39.
-(1974)- Unpublished interview with Barbara Newman. T34)escript from
audio recording. Oral History Archives, Dance Collection, New York Pub¬
lic Library.
Sorell, Walter, ed. (1951). The Dance Has Many Faces. New York: World
Publishing.
-(1956). “The Playhouse Dance Company.” Dance Magazine (July): 70.
-(1958). “Close-Up of Modern Dance Today: The Henry Street Play¬
house Produces a Major Dance School, Workshops and Company.” Dance
Magazine (January): 49-52.
-(1962). “Donald McKayle and Company.” Dance Observe?- (June-July): 91.
-(1966). The Dance Has Many Faces. 2nd ed. New York: Columbia Uni¬
versity Press.
Stahl, Norma Gengal (1954). “The First Lady of the Metropolitan Opera Bal¬
let: An Interview with Janet Collins.” Dance Magazine (February): 27-29.
Staples, Robert (1982). Black Masculinity: The Black Man’s Role in American So¬
ciety. San Francisco: Black Scholar Press.
Steinberg, Cobbett, ed. (1980). The Dance Anthology. New York: New American
Library.
Steuben, Francis [Edna Ocko] (1943). “Katherine Dunham’s Revue.” New Masses
(2 November): 26.
-(1944)- “Pearl Primus Dance.” New Masses (31 October): 28.
Steyn, Mark (1997). Broadway Babies Say Goodnight: Musicals Then and Now.
London: Faber and Faber.
Stich, Sidra (1990). Anxious Visions: Surrealist Art. New York: Abbeville Press.
Stuckey, P. Sterling (2002). “Christian Conversion and the Challenge of Dance.”
In Dancing Many Drums: Excavations in African American Dance, edited by
Thomas F. DeFrantz. Madison: University of Wisconsin Press.
Suskind, Peter (1943). “A Plug for One of‘Little Folks.’” Norfolk Journal and
Guide (27 February): 14
Suzuki, Daisetz, T. ([1934] 1964). An Introduction to Zen Budhism. Reprint, New
York, Grove Press.
Sydie, R. A. (1987). Natural Women, Cultured Men: A Feminist Perspective on
Sociological Theory. New York: New York University Press.
Tallchief, Maria (1997). Maria Tallchief: America’s Prima Ballerina. New York:
Henry Holt.
Taper, Bernard (1984). Balanchine: A Biography. New York: Times Books.
Taylor, Paul (1987). Private Domain. New York: Alfred A. Knopf.
Telberg, Lelia K. (1959). “Murray Louis and Group.” Dance Observer (May): 73.
References 241

Terry, Walter (1940a). “Dunham Dancers.” New York Herald Tribune (19 Feb¬
ruary); 13.
(1940b). “The Negro Dances.” Nczr York Herald Tribune (28 April): VI10.
(1941)- 'Talanchine and Graham.” New York Herald Tribune (26 Janu¬
ary): VI 10.
-(1947a). “Initial Ballet Society Works Stress Variety of Form, Style.”
New York Herald Tribune (19 January): V 4.
-(1947b). “Lo! The Poor Dancer.” Dance Magazine (March): 25—28.
-(1951a). “The Dance World: A Jewish Community in Dance.” New
York Herald Tribune (4 February): IV 7.
-(1951b). “The Ballet: Absorbing Revival.” New York Herald Tribune (23
November): 17.
-(1955). “American Dance.” New York Herald Tribune (16 May): 12.
-([1956] 1971)- The Dance in America. Reprint, New York: Harper and Row.
-(1978). I Was There: Selected Dance Reviews and Articles ip^6-ip'j6. New
York: Marcel Dekker.
Thomas, Helen (1995). Dance, Modernity and Culture: Explorations in the Soci¬
ology of Dance. London: Routledge.
-(1996). “Do You Want to Join the Dance? Postmodernism, Poststruc-
trualism, the Body, and Dance.” In Mow’/zg Words: Rewriting Dance, edited by
Gay Morris. London: Routledge.
Tomkins, Calvin (1976). The Bride and the Bachelors: Five Masters of the Avant-
Garde; Ducha?np, Tinguely, Cage, Rauschenberg, Cunningham. New York: Pen¬
guin Books.
-(1980). Ojfthe Wall: Robert Rauschenberg and the Art World of Our Time.
New York: Doubleday.
Vaughan, David (1957). “Close-Up of Modern Dance Today: The Non-
Objective Choreographers.” Dance Magazine (November); 20-23.
-(1958). “Dissenting View on Graham.” Dance Magazine (July): 26-27.
-(1997a). Merce Cunningham: Fifiy Years. New York: Aperture.
-, ed. (1997b). Merce Cunningham: Creative Elements. Choreography and
Dance 4: 3.
Volkov, Solomon (1985). Balanchine's Tchaikovsky: Conversations with Balanchine
on His Fife, Ballet and Music. New York: Doubleday.
Warren, Larry (1998). Anna Sokolow: The Rebellious Spirit. Amsterdam: Har¬
wood Academic Publishers.
Washington, Fredi (1944). “Pearl Primus Captures Broadway; Josh White
Added Attraction.” People's Voice (21 October): 29.
Watts, Alan (1957). The Way of Zen. New York: Mentor Books.
Weber, Max ([1904-1905] 1999). The Protestant Ethic and the Spirit of Capital¬
ism. With an introduction by Anthony Giddens. London: Routledge.
242 References

-(1946). From Max Weber: Essays in Sociology. Edited by H. H. Gerth and


C. Wright Mills. New York; Oxford University Press.
-(1978). Weber: Selections in Translation. Edited by W. G. Runciman.
Cambridge: Cambridge University Press.
Weber, Nicholas Fox (1992). Patron Saints: Five Rebels Who Opened America to a
New Art, ipzS-ip^y. New York: Alfred A. Knopf.
Werth, Alexander (1948a). “The ‘Reform’ of Soviet Music.” Nation (10 April):
393-395-
-——— (1948b). “Culture and Commissars: The Dilemma of Soviet Artists.”
Nation (21 August): 207-209.
Whitfield, Stephen (1991). The Culture of the Cold War. 2nd ed. Baltimore:
Johns Hopkins University Press.
Whyte, William H., Jr. (1956). The Organization Man. Garden City, N.Y.:
Doubleday.
Wiebe, Robert H. (1990). The Search for Order: iSyy-ipzo. New York: Farrar,
Straus and Giroux.
Wiggins, Ebzabeth (1946). “The Pilgrim Again Waylaid.” Dance Observer (De-
cember): 125.
Wilde, Patricia, et al. (1987). “Celebrating The Four Temperaments—r.” Ballet
Review (Winter): 12-35.
Wilder, Lucy (1952). “Stage for Dancers.” Dance Observer (April): 58.
Wilkins, Darrell (1993). “The Four Temperaments, an Interpretation.” Ballet
Review, 21, 2 (Summer): 81-87.
Williams, Raymond (1983). Keywords: A Vocahidary of Culture and Society. New
York: Oxford University Press.
Williams, Simon J., and Gillian Bendelow (1998). The Lived Body: Sociological
Themes, Embodied Issues. London: Routledge.
Wolfson, Bernice J. (1947). “A Time to Speak and Dance.” Dance Observer
(October): 88-89.
Wollf, N. S. (1944). “Pearl Primus.” Dance News (November): 7.
Index

Page numbers in italic represent illustrations.

abstraction, 68-70, 71-72, 167, 201 stams quo, 68; emhodiment of emo¬
Actors Studio, 103 tion and, xiv, 66-67, 731 ^5
Adorno, Theodor, 3-4, 102 ;avant-garde, 3-4
Africa, 50-52, 136-137, 144, 157, 159
African Americans. See race Bailin, Gladys, 166, 185, igi, 192, 194;
Ailey, Alvin, 33-34, 149, 162-, Blues Suite, Enclave, 190; Harlequinade, 190
149 Baker, Josephine, 51, 149
Aleichem, Sholem, 89-90, 92 Balanchine, George: “Africanist” influ¬
Alexander, James, 124 ences on, 50-52, Apollon
alienation, 97-102 Musagete, 51; Ballet Imperial, 43;
American Ballet, 41, 43-44, 2o8n6 and ballet technique, 60; Le Chant du
American Ballet Caravan. See Ballet Rossignol, 60; as choreographer, 56-59,
Caravan 82, 161, 190; Concerto Barocco, 43-44;
American Ballet Theatre. See Ballet contributions to films and musicals,
Theatre 1-2; critics and, xx, xxii, 43-44; on
American Dance Festival, 105, 178, 191 dance as male profession, 34; Danses
Anthony, Mary, xxvi, 200; The Devil in Concertantes, 44, 80; The Four Tem¬
Massachusetts, 32 peraments, 43, 44-57, 46, 48, 50, yy,
anthropology, 116-117, 119-120, 170, 2o8n8; impersonahty in dances,
136-137 56, 58-59; internationalism and,
antifascism, 8-9 41-42, 61; and manipulation of
Armitage, Merle, 207n3 women, 52, 54-55, 209ni5; on mod¬
art: autonomy of, xiii, xv-xvi, 5-7, 19, ernist ballet, xxvii, 37; movement style
70-72, 82, 166, 201, 203-204; as imi¬ of, 168-169; neoclassicist, 38-39,
tation, 70; and life, xv, 19, 30, 66, 70, 43, 57, 77-78; as neoromanticist, 43,
85-86, 203-204; time art, 80, 2ion8 72-73; Orpheus, 43; photo, 49; Prodigal
Artaud, Antonin, 2o6n6 Son, 42; Serenade, 43; The Seven Deadly
Ashley, Merrill, 49 Sins, 42; Stars and Stripes, 63; surreal¬
Astaire, Fred, 2 ism and, 52-54; The Triumph of Bac¬
Aul, Rorme, xxvi, pr, 93, 115, 149, 2ion2 chus and Ariadne, 43; use of youthful
authenticity: artistic autonomy and, 82; dancers, 59-60; Western Symphony, 62
ballet and, 6, 71-72; as challenge to IBalcom, Lois, 134-136
244 Index

Bales, William, 25, Sg; Judith, 25 Beiswanger, George W, xxi, xxiii, 36, 105
Ball, William, 46-47 Belafonte, Harry, 151
ballet: balletic abstraction, 71-72; in Bell, Daniel, 87, 102, 148, 150
Broadway choreography, 1-2, 5; clas¬ Bellmer, Hans, 54
sical danse d’e'cole, 38, 43, 46, 49, 58; Bendelow, Gillian, 196
Dunham’s use of, 120, 121; Graham’s Benjamin, Walter, xviii
attitude toward, 15-16, 20; influence Berk, Ered, 88
in modern dance, 5-6, 15-16, 17-18, Berman, Eugene, 42
54-55, 79; modernist ballet, xxvii; Bernstein, Harry, 158
modernist vanguard and, 36-37, Bernstein, Leonard, 4
38-39, objectivism and, 200-201; Bettis, Valerie, 16, 24-25, 64; As I Lay
psychological ballet, 77; race and, 120, Dying, 24, 32; Daisy Lee, 24, 26
121, 139, 150, 161 Bibb, Leon,152-153, 156
Ballet Caravan, 15, 41, 2o8n6 Bird, Bonnie, 78, 168
Ballet Negre, 117 Birsh, Patsi: In the Mines of Avondale, 32
Ballet Russe de Monte Carlo, 38, 60, 76, Bloch, Ernest, 151
150, 20701, 209-21006 Blok, Alexander, 42
Ballet Society: charter of, 38, 40; Denby body. See embodiment
on, 83; internationalism and, 41; Bolender, Todd, 46, 47
Kirstein and, 38-39, 20703; neo¬ Boleslavsky, Richard, 65, 20903
romanticism and, 43; reformed as Bolshoi Ballet, 74
New York City Ballet, 56, 62, 76 Borzoi Book of Modem Dance, The (Lloyd),
Ballet Theatre, 38, 76 xxii
Ballets Russes, Les, 42, 52 Bourdieu, Pierre, xvi-xvii, xxiv-xxvi, 19,
Banes, Sally: on “Aificanist” presence in 58, 84
Balanchine, 50-52, 60; on Cage, Bowles, Paul, 74
20901; on dance images, 46 Bradley, Buddy, 51
Baronova, Irina, 60 Brancusi, Constantin, 40
Barr, Alfred H., 41 Braque, Georges, 40
Barrault, Jean-Louis, 25 Brassai (Gyula Halasz), 54
Baudelaire, Charles, xviii Brecht, Bertolt, 42
Bausch, Pina, 113 Breton, Andre, 52
Bears-Bailey, Kim, 141 Brinkley, Alan, 87
Beatty, Talley: Balanchine and, 63; Blues, Britten, Enid, 162
139; in Broadway musicals, 2; career Broadway (New York City): ballet tech¬
of, 137-146, 159-165; as Dunham nique and, 1-2, 5; as Dunham venue,
dancer, 121, 137; Kanzo, 142; Kirstein 2, 121-123; modern dance techniques
and, 20703; photo, i$8-. The Road and, 7-8, 201; musicals, xxiii, 107; as
of the Phoebe Snow, 159-165, 262; Primus venue, 13 2-134; as symbol of
Rural Dances of Cuba, 142; Saudades commercialism, i, 6, 7-8
do Brazil, 142; Southern Landscape, Brodie,Joan, 32, 142-143
139-144, 141, 213014; use of ballet Bronte, Charlotte, Emily, and Anne, 20
technique, 121, 139, 145-146, 164; Broughton, Shirley, 166
Vieux Carre, 142; virtuosity as protest Brown, Carolyn, xix, i6g, 170, 175,
in,161-163, 201-202 178, lyg
Beck, Esta and Eve, 147 Brown, Earle, 175
Beiswanger, Barbara Page, xxi Brown, Richie, 189
Index 245

Buirge, Susan, 184, 215010 147-148; House Un-American Activi¬


Burger, Peter, 173 ties Committee (HUAC), xviii-xix,
Burley, Dan, 122 xxiii, II, 33; literalism and, 32-33,
Burt, Ramsay, 125, 20501 201; objectivism and, 201; Truman
Butler, Gervase, 44, 130 policies and, 10—13. commu¬
nism; postwar era
Cabin in the Sky, xxiii, 2, 51, 123, 161 Coleman, Martha, 142
Cage, John; critique of modern dance, Collins, Georgia, 163
78-82; Cunningham and, 64, 78, 81, Collins, Janet, 63, 149-150; Spirituals,
168, 171-172, 199, 20901; impersonal 151; Three Psalms of David, 151
art practices, 80-81; Nikolais and, commercialism: black modernism attrib¬
192; photo, 79; Zen Buddhism and, uted to, 114-115, 125-128, 135-136;
81,174-175 Broadway as symbol of, i, 6, 7-8; as
Caldwell, Candace, 163 element of modernity, 3-4; Graham
Campbell, Joseph, xxi, 28, 30-31, 72 on entertainment, 13-14; influence in
capitalism, xiv-xv, xv, xviii, 66. See also modern dance, 1-4, 82, 2o6n2; objec¬
commercialism tivism and, 200-201. See also capital¬
Caribbean Islands, 117, 119-120, ism; market
126-128, 137-139, i42-i43>149 Committee for the Negro in the Arts, 151
Caron, Leslie, 2 communication. See meaning
Carroll, Noel, 20901 communism; homosexuality associated
chance procedures, 81, 83, 172-181, 201 with, 36; “Masterpieces of the Twenti¬
Charisse, Cyd, 2 eth Century” festival and, 62; modern
Charhp, Remy, xix, 147, 157,172-173, 178 dance and, 8; modernism as weapon
Chavez, Carlos, 192 against, 41; Sokolow and, 98; Soviet
Chicago Opera Ballet, 117 art, 11-12. See also Cold War; House
choreodrama, 42 Un-American Activities Committee
Choreographers’ Workshop, 24-25, (HUAC); Marxism
I39-140 community, xxviii, 5, 6-7, 88-89, 97-102
Christensen, Lew: Blackface, 139, 20703 Como, Perry, 106
Christopher, Patricia, 109-110 Congress for Cultural Freedom, 62
Chujoy, Anatole, xx, xxii, 44, 126. See connotative movement, 46-47
also Dance News Copland, Aaron, 74
Churchill, Allen, 61 Copper and Brass, 152
CIA (Central Intelhgence Agency), Crawford, Cheryl, 103
xviii-xix criticism: on Balanchine, 43-44; dance as
Clark, T. J., XV, 18 bodily practice and, xxiv; expressional
Clarke, Martha, 113 vs. objectivist dance in, xxii-xxiii,
classicism: Balanchine as neoclassicist, 194-195, 197-199; as gendered dis¬
38-39, 43, 57, 77-78; balletic abstrac¬ course, xxiii; consecration of dancers,
tion, 71-72; Denby on, 75, 77, 82-83; xxii; leftist discourse and, xxiii; post¬
as historical style, 69-70; vernacular war dance magazines, xx-xxii; as racial
themes and, 57-58. See also danse discourse, xxiii
d'kole Croce, Arlene, 56-57
Clooney, Rosemary, 106 Cronkite, Walter, 157
Cohen, Selma Jearme, xxii, 158 cubism, 40
Cold War: fears of communism and. Cullen, Countee, 151
246 Index

culture industry, 3-4. See also kitsch; mass Denby, Edwin; on American ballet style,
culture 76-78; background, xx; on Balan¬
Cummings, e. e., 40 chine, 44, 51, 54. 5*5. 57-58, 75-76,
Cunningham, Merce: anti-meaning tech¬ 83; biographical sketch, 74; on classi¬
niques of, 171-181, 2i4n3; Cage and, cism, 75, 82-83; on Cunningham,
64, 78, 81-82, 168, 171-172, 209ni; 198-199; on de Mille, 77; on Dun¬
consecration of, xvii; as dancer, xix, ham, 127, 130; on Graham, 78; photo,
15, 168, 195, 200; Denby on, xxii-xxiii; 75; on Primus, 132; on protest dances,
homosexuality and, 195; Kirstein and, 132, 146; on Robbins, 77; on Tudor,
207n3; modernist vanguard and, 77, 21007
195-199; movement theory of, Depression. See New Deal era
168-174, 19*5-197; as objectivist, Destine, Jean Leon, 157
167-181, 195-197, 20607; photo, 169; Diaghilev, Serge, 52, 72, 76, 209-21006
The Seasons, 63; Septet, i6p, 169-170; Dickinson, Emily, 20
Sixteen Dances for Soloist and Company Dolin, Anton, 76
of Three, 181; Suite by Chance, 172; Dove, Arthur, 40
Summerspace, 171-172, 175, 178-179, Dove, Ulysses, 162
179, 21405; use of African American Drake, St. Clair, 127-128
dancers, 147; use of chance proce¬ Drown, Marilyn Vaughan, 174-175
dures, 172-174, 175-176, 178-179 Dudley, Jane, 10, 13, 15, 89, 91
Dudziak, Mary L., 148
dadaism, 81, 174, 180, 188 Duncan, Jeff, 113
Dalcroze Method, 117 Dunham, Katherine: Balanchine and, 51,
Dalgleish, James: Salem Witchhunt, 32 63; Broadway as venue for, 2, 121-123;
Daly, Ann, 209015 career of, 116-130; Chorus #1, 120;
Damasio, Antonio, 20904 commercialism attributed to, 126-128;
Dana, Margery, 122-123 critics on, xxiii, 142-143; Dunham
dance drama, 25-32, 42, 69. See also technique, \20-, UAg’Ya, 115, 120;
narrative racial cross-fertilization initiative,
Dance Encyclopedia (Chujoy), xx 119-120; government harassment of,
Dance Magazine, xxi-xxii, 166-167. See 148; Rhtimbajive, 120; Rites de Pas¬
also Hering, Doris; Joel, Lydia sage, 123-128, 124; sexuality as protest
Dance News, xxi, 2. See also Chujoy, in, 126-130, 201-202-, Southland, 148;
Anatole Tropical Revue, 123-128, 124; Tropics
Dance Observer: on Balanchine, 44; on and Le Jazz “Hot”: From Haiti to
Beatty, 142; on commercialism, i; Harlem, 117, 118, 122-123, 21103;
dancers’ views represented in, xix-xxii; use of ballet, 120, 121
on Dimham, 121-122; on Limon,
27-28; on postwar modern dance, Elg, Taina, 2
167; on Primus, 134-135 Eliot, T. S., 40
Dance Theater Workshop, 113 Ellen, Vera, 2
Danieli, Fred, yo Ellington, Duke, 160, 163
danse d’e'cole, 38, 43, 46, 49, 58, 75. See also Ellis, Lucille, 724
classicism embodiment: abstraction and, 68-70,
de Kooning, Elaine and Willem, 74 71-72, 196-197; corporeal idealiza¬
de Mille, Agnes, xxvii, 1-2, 36-37, 77 tion, 71; dance as independent art
decentralization, 186-187 and, 67-72, 83-86; of emotion,
D’Emilio, John, 33 67-68; narrative and, 18; of the
Index 247

psyche, 72; surrealism and, 53-54. Four Temperaments, The, 43, 44-57,46,
See also representation 48, 50, 55, 170, 2o8n8
emotional motor memory (Martin), 67, Fowler, Jason, 55-
209n4 France, xv, xvi
Erdman, Jean; Cage and, 64; Campbell Frank, Bill, 195
and, xxi; Cunningham collaboration Frankel, Shelley, 156
with, 200; The Fair Eccentric, or The Franko, Mark, 197, 20503
Te?}tporary Belle of Hangtoum, 25; on Freud, Sigmund, 31
Graham, 15; Ha?nadryad, 145; on move¬ Fromm, Erich, 98-102, 21208
ment vocabulary, 16; Ophelia, 143-145; Frontier, 20-21, 22
Transformations of Medusa, 17; use of Funaroff, Sol, 94
African American dancers, 147
Ernst, Max, 52, 81 Garth, Midi, 166
ethnicity: ethnic dance, 88, 149-150, 164; gender; choreographic partnering and,
Maslow’s treatment of, 88-93; 195; criticism and, xxiii; dance as
and, 149-150, 164; Sokolow’s treat¬ feminine vs. masculine profession, 34,
ment of, 88-89, 94~97> 21004 197; female body manipulation, 52,
expressional modem dance: Alartin on, xix, 54-55, 209015; objectivism and,
70-71, 20607; Nikolais’s and Cunning¬ 196-197; surrealism and, 53-54
ham’s opposition to, 170-171, Germany, 12
182-183, i95~299> 215-216020 Gifford, Joseph, 7-10, 13
Gilman, Sander, 129-130
Ealco, Louis, 147 Gitelman, Claudia, 184
Earber, Viola, i6g, 170, 178, 179 Goff, Eleanor Anne, 135-136
Earrell, James, i Goffman, Erving, xxiv
fascism, 8-9, 12, 41 Gomez, Tommy, 139
Fast, Howard, 139, 143 Gordon, David, 197
Faulkner, William, 24 Gorky, Arshile, 74
FBI (Federal Bureau of Investigations), Gottschild, Brenda Dixon, 50-52, 60
xviii-xix, to Graff, Kurt and Grace, 137
Federal Theater Project, 115 Graham, Billy, 35-36
Feldman, Morton, 175 Graham, Martha; affiliations with other
film, 2, 51, 106 dancers, 15, 89, 130-131, 151, 152,
Fippinger, Lynne, iii 184; on American dance, 57-58;
Fokine, Michel, 61, 72 American Provincials, 20; attitude
Fonaroff, Nina, 25, 27, 64, 200 toward ballet, 15-16, 20; Balanchine
Foner, Eric, xviii compared to, 43; on the body as an
Foote, Horton: on crisis in modem emotional instmment, 196-197;
dance, i, 3-4, 6, 20601; as Dance Campbell and, xxi, 30-31; Cave of the
Observer editor, xxi; as librettist, 24; Heart, 20; Dark Meadow, 20; Deaths
on political engagement by artists, 5, and Entrances, 20; Denby on, 78;
6-7; on Primus, 135. See also Dance Diversion of Angels, 20; Dunham com¬

Observer pared with, 122; early “pure dance”

formalism, 11-12, 66, 68-69, 71-72, style of, 21, 24; Every Soul is a Cirms,
24; Erontier, 20-21, 22; Kirstein and,
20607
Foster, Susan, xxiv-xxvi, 85, 20607 40, 207n3; Lamentation, 94-95,96;
Letter to the World, 20; Martin on, 73;
Foucault, Michel, xxiv
on narrative, 13, 18-21, 24, 30, 199;
Foulkes, Julia, xiv
248 Index

Graham, Martha (continued) House Un-American Activities Commit¬


Night Journey, 14, 20-21, 25; Perspec¬ tee (HUAC), xviii-xix, xxiii, ii, 33.
tives, 20; Primitive Mysteries, 15; on See also postwar era
representation, 18; “theater of Martha Hoving, Lucas, 29
Graham,” 13-14, 14, 21, 24; use of Howe, Irving, to
African American dancers, 147; use of Hughes, Langston, 151
American material, 8; Vaughan on, Humphrey, Doris, 8, 17, 167, 184
198-199 Humphrey-Weidman group, 2 7
Grauert, Ruth, 185, 189 Hurok, Sol, 123, 132
Greenberg, Clement, xv-xvi, 3-4, 8, 70,
202-203, 207-2o8n4 Imperial Ballet of St. Petersburg, 43
Guggenheim, Peggy, 81 internationalism: as antifascist initiative,
Guilbaut, Serge, 9 8-9; Ejrstein and, 41; liberalism and,
Guthrie, Woody, 89 10; State Department “cultural am¬
bassador” program, 62, 113, 148-149;
habitus, xvi, xxiv surrealism and, 52-53, 2o8-209ni4.
Hadassah, 88 See also nationalism
Hammarskjold, Dag, 63
Harper, Herbie, 51 Jackson, Naomi, 88, 92, 21301
Harvard Society for Contemporary Art, 40 Janson, H. W., 9-10
Hawkins, Erick, xxvi, 2, 15, 63, 166 Jaques-Dalcroze, Emile, 74
Henry Street Playhouse, 166, 184-185, jazz, 105, 107-109, 134, 160
iSy, 189, 199 Jelin, Max, 132
Hering, Doris; on Balanchine, 55, 57-58; Joel, Lydia, 166
on Beatty, 140, 162-163; Dance Joffrey Ballet, 107, 112
Magazine writer, xxii, xxiii, 2o6nio; on Johnson, Harriet, 142
Graham, 198; on Limon, 27-28; on Johnson, Louis, xxvi, 63; Spiritual Suite, 151
Sokolow, 104, 109; treatment of racial Johnson, Philip, 41
issues, 151 Johnson, Tommy, 154, 163
Herskovits, Melville, 116 Johnston, Jill, xxii, 113, 163, 167
Hindemith, Paul, 45, 2o8n7 Jones, Bill T, 21407
Hinkson, Mary, 158-159 Jordan, Stephanie, 58-59
Hodson, Millicent, 120 Joyce, James, 40
Holder, Geoffrey, 149 Juilliard Dance Ensemble, 99, 104, iii
Holm, Hanya, 2, 184, 186, 188, 194, 200 Jung, Carl, 30-31
homosexuality, 33-36, 195
Hoover, J. Edgar, 10 Kaddish, 89, 93-95, 95, 97
Hopper, Edward, 40 Kaleidoscope, 187, 190-194, 191, 295
Horkheimer, Max, 3-4, 102 Kaschmann, Truda, 184
Horst, Louis: on Beatty, 139-140; com¬ Kazan, Elia, 103
position classes, 17; as Dance Observer Kelly, Gene, 2
founder, xxi, xxiii; on Dunham, Kertesz, Andre, 54
125-126; Erdman and, 17; Martin Khatchaturian, Aram, 11
compared with, 65; Nikolais and, 184; kinesthetic sympathy, 85, 203
Primus influenced by, 130; Sokolow Kinsey report on male sexuality, 35
and,103, 104, 109-110 Kirstein, Lincoln; background, 33-34,
Hound and Horn literary magazine, 40 40- 41, 20703; Balanchine and, xx,
House of Flowers, 152 41- 43, 61; Ballet Society charter by.
Index 249

38; on communism, 42; on The Four i86-i87;^w tech amber, 190; Belonging
Temperaments, 47, 56; Greenberg to the Night, 190; Polych7'ome, 190
compared with, 207-20804; on mod¬ Louther, William, 154
ernist ballet, xxvii, 37, 38-40; on Lowe, Robert, 125
neoclassicism, 38-39, 43; neoroman¬
ticism and, 42-43, 207-208114; oppo¬ Mabry, Iris, xxvi, 16, 63-64, 20703
sition to modern dance, 40, 200; Macero, Teo, 105, 108, no
photo, 49; on Seligmann, 56. See also Manchester, P. W, xxii, 109, 159, 163
Ballet Caravan Manning, Susan, 115, 21206
kitsch, XV, 3 Maracci, Carmelita, 150, 167
Koner, Pauline, 29 Marchowsky, Marie, 64, 166
Krevitsky, Nik, xxiii, 16, 31-32, 92, 181 Marcuse, Herbert, 102
market, xv-xviii, 4. See also commercialism
Laban, Juana de, 7-8 Markova, Alicia, 60, 76
Laban, Rudolf, 103, 117, 186, 188 Marsicano, Merle, 64, 166, 200
Laban Movement Analysis, 103, 117 Martin, Don, 754
Labor Stage, 115 Martin, John: background, 33-34, 65; on
Lamhut, Phyllis, 166, 185, 187, 195, Balanchine, 43-44, 72-73, 77-78; Bal-
215010; Annoyous Insectator, 190 com on, 134; on Beatty, 121, 137-138,
Lang, Pearl, 88 163-164; as critic, xx; critique of mod¬
language, 19-20, 179-180 ern dance, 17-18; Denby compared
LeClercq, Tanaquil, 59, 60 with, 74-75; on drama, 69; on Dun¬
Lesschaeve, Jacqueline, 168, 177 ham, 121, 125-126; on embodiment,
Lewis, Robert, 103 19, 67-70, 71-72, 83-85, 20904;
Lewitzky, Bella, 33 emotional motor memory, 67, 20904;
liberalism; Cold War and, xxiii, 11; Erich on expressional modern dance, xix,
Fromm and, 98; internationabsm and, 70-71, 20607; ittner mimicry, 67; on
10; socialism and, 8 Limon, 27-28; on Maslow, 92, 93;
Limon, Jose: at American Dance Festival, metakinesis, 67; on Nikolais,
191; £/ Grito, 32; gender and, 197; 197-199; photo, 66; Primus and,
Johnston on, 167; Martin on, 73; The 131-132, 136,21209; on Sokolow,
Moor’s Pavane, 27-28, 29, 20709; 105; theory of modern dance, xiv, xix,
postwar reputation of, 27-28; The 65-73, S3-S6; treatment of racial
Traitor, 32 issues, 93, 151, 163-164, 21003
Lippincott, Gertrude, xxi, 5-8, 12, 31 Martindale, Coral, 191, 194
Lippold, Louise, 64 Marxism, 98. See also communism
Lipset, Seymour, 102 Maslow, Sophie; background, 13, 15, 89;
literalism: criticism of, 31-32, 201; Dust Bowl Ballads, 89; Folksay, 89;
racial discourse and, 132, 158. See also McKayle and, 93, 155; Bagged, Hun¬
narrative gry Blues, 89; social protest and, 10,
Litz, Katherine, xxvi, 27, 166, 167 89; treatment of Jewish ethnicity,
Lloyd, Barbara, 170 88—93; Songs about Lenin, 89; use
Lloyd, Margaret: on Dunham, 129; on of African American dancers, 93, 147,
Graham, 15; on internationalism, 10; 21002; The Village 1 Knew, 89-93, 9^
as modern dance historian, xxii; on mass culture, xv, 3-4, 7-
Primus, 135; on Sokolow, 107 Massine, Leonide, 150; Gatte Paiisienne, 142
Louis, Murray, xix, 166, 184-193, 191, Masson, Andre, 52
194, 197-198, 215014; \n Allegory, Matisse, Henri, 40
250 Index

Matlowsky, Samuel, 89-90 movement: Bettis on, 25; in Dunham,


Mauss, Marcel, xxiv 120, 128; connotative movement,
McCarthy, Joseph, 10. See also postwar era 46-47; Cunningham’s emphasis on,
McCrae, Edna, 137, 213015 168-170; Denby’s emphasis on, 75,
McKayle, Donald, 93, 115, 147, 152-159, 77; Fonaroff on, 27; Johnston’s theo¬
21002; Games, 152, 155, is-]-,Her ries on, 167; Laban Movement Analy¬
Name Was Harriet (Her Name Was sis, 103, 117; Litz on drama and, 27;
Moses), 155; Rainbow Round My Shoul¬ modern dance vocabularies, 16, 17;
der, 152-159, 1^4, iy6, 213-21402; objectivism and, 200; political engage¬
Saturday's Child, 151, 155 ment and, 203-204. See also embodi¬
meaning: artistic communication and, ment; representation
67-69, 80-84, 167; chance procedures Moylan, Mary Ellen, 50
and, 81, 83, 172-181; connotative Museum of Modem Art, 40-41
movement, 46-47; embodied vs. rep¬ music: Balanchine and, 45-46; Cunning¬
resented, XV, 18-19, 75> 77“7^> 8o-8r, ham and, 81-82, 171-172; jazz, 105,
92-93, 203-204; inner mimicry and, 107-109, 134, 160; McKayle and,
67; kinesthetic sympathy and, 85, 203. 155-157; Nikolais and, 188-189
See also embodiment; representation Musical America, xxi, xxiii
Meldon, John, 125, 127 Myers, Gerald, 150
Melsher, Jo Anne, 269 Myrdal, Gunnar, 148
metakinesis, 67 myth, 21, 24, 30-31,43, 53
Metropolitan Opera Ballet, 150
Miller, Arthur, 3 2 Nabokov, Nicholas, 62
Mills, C. Wright, 174 Nagrin, Daniel, xxvi, 2
minstrelsy, 119, 12 2-12 3 narrative: Bettis’s use of, 24-25; Camp¬
Modem Music, xx bell on, 28, 30-31; Denby’s critique
modernism: artistic control and, xiii; of, 75; Graham and, 13, 18-21, 24, 30,
black modernism, 114-115; Cold War 199; Martin’s critique of, 69; Palmer
environment and, 36, 41; as focus of on, 31; representation and, 18; social
book, xiii, 20504; as historical style, protest and, 32-33, 163-164; Wolfson
69-70; political engagement and, 202; on, 31. See also dance drama
racial discourse and, 159, 164-165; Nash, Joe, 2, 140
relationship to the market, 3-4; nationalism: American impersonal style
representation and, xv, 18, 20606; and, 59, 76; Balanchine and, 57-63;
self-criticism in, xv, 202; Zen internationalism vs., 8-9; modern
Buddhism and, 179-180. dance as American form, 57-58; ver¬
modernist vanguard: artistic autonomy nacular themes and, 8-9, 41. See also
and, xvii-xviii, 201; Balanchine and, internationalism; regionalism
61-62; ballet and, xxiii, 36-37, 38-40; Neal, Charles, 154
critics’ influence on, 195, 197-199; Negro Dance Group, 117, 120—121, 137
dance drama and, 32; Kirstein initia¬ Neoclassicism. See classicism
tives and, 41, 62; objectivism and, neoromanticism, 42-43, 207-2o8n4
167, 195-196, 198-199; as position in New Dance Group: internationalism and,
dance field, xiii, xix, xxvii, 38-39, 8; Maslow and, 89, 93; McKayle and,
200-201; repressive postwar environ¬ 152, 155; Primus’s affiliation with,
ment and, 36. See also modernism 130-131, 157; racial integration in,
Moore, Jack, 113 115,2ion2
Moor’s Pavane, The, 27-28, 29, 20709 New Deal era, xviii, 5, 9, 115
Index 251

Newman, Patricia: Green Mansions, 31-32 Page, Ruth, 117


Newton, Frankie, 132 painting, 40-41, 70, 74
New York City Ballet, xxii, 41, 44, 56, Palatsky, Eugene, 59
62, 76 Palmer, Wmthrop, 31
Night Journey, 14, 20-21, 24 Pandor, Miriam: Salem Witch Hunt, 3 2
Nikolais, Alwin: Allegory, 186-187; back¬ pantomime: connotative movement and,
ground, 184-186; Cage and, 64, 192; 46-47; Erdman’s use of, 25; in de
critics on, 194-195, 199; dance theory Mille and Robbins, 77; Lim6n’s use
of, 181-184, 186-188; decentrahzation, of, 28; Maslow’s use of, 90, 92-93;
186-195, 215010, 215014; Etudes Com¬ narrative and, 18
posed by the Dancers, 189; Etudes II: Parham, Ernest, 163
Masks, Pivps, a?id Mobiles, 190; at Henry Paris Opera Ballet, 61
Street Playhouse, 184-185, i8y, 189, Perrottet, Suzanne, 188
igg; Kaleidoscope, 187, 190-194, ipi, Petipa, Marius, 43, 72-73, 75-76
195; Masks—Props—Mobiles, 190-191, Phelps, Mary, 7, 16, 135, 140
194; Mirrors, 215014; Noumenon (Nou- photography, xxvi-xxvii, 53-54
menom Mobilis), 189, 190; as objectivist, Pins and Needles, 117
167-168, 195-197; photo, i8y, sexual¬ Poitier, Sydney, 151
ity and, 191-192, 195, 198, political engagement: artistic autonomy
215-216020; Technical Etudes, 189; use and, xv-xvi, 5-7, 201; Cold War envi¬
of African American dancers, 147 ronment and, 87-88; consensus ideol¬
Nixon, Richard, 10 ogy and, 88, 97, 131, 147, 149; critics’
Noble, Duncan, 26 deflection of, 195, 199-200, 202;
North, Alex, 94, 98 Maslow’s background in, 89; mod¬
ernism and, 202; movement and,
objectivism: Cohen and, xxii; Cold War 203-204; New Dance Group produc¬
and, 201; Cunningham and, 167-181, tions and, 8, 89; objectivism and, 200;
195-200, 20607; Denby and, xxii; political embodiment (Bourdieu), 58;
expressional modern dance vs., xix, Sokolow’s approach to, 93-94, 9S,
xxii-xxiii, 194-199; formalism and, 106-107, 112-113; virtuosity as, 131,
20607; gender and, 196-197; John¬ 161-163, 164-165, 201-202, 2i4n7
ston and, xxii; Litz and, 167; mod¬ Porter, Katherine Anne, 4
ernism and, 200; modernist vanguard postmodernism, 2i6n2i
and, 167, 195-196, 198-199; Nikolais postwar era: antifascism and, 8-9; homo¬
and, 167-168, 181-200; political phobia and, 33-36; political consensus
engagement and, 201-204; Vaughan during, 87-89; religious revival,
and, xxii, 198 35-36; social welfare programs and,
Ocko, Edna (Francis Steuben), xxiii, 33, xviii; threat of atomic catastrophe, 56;
127,132, 21206 treatment of racial issues, 115-116,
O’Donnell, Mary P, 122 147-152, 157-158; youth rebellion
O’Donnell, May, 15, 167 and, 105-107; Zen Buddhism and,
O’Hara, Frank, 74 181. See also Cold War
O’Keeffe, Georgia, 40 Pound, Ezra, 40
Oklahoma!, 1-2 Pratt, John, 121-122, 124
Olivier, Laurence, 25 Preger, Marianne, 176-177

On the Town, 2 Presley, Elvis, 106, 147


Opus'6^, 107, 110-112, III Primus, Pearl: African Ceremonial, 131;
black press coverage of, xxiii; Broad-
Ormer, Sherry, 196
252 Index

Primus, Pearl (continued) Riabouchinska, Tatiana, 60


way debut, 132, 21209; Cage and, 64; Richter, Hans, 188
career of, 130-137, 21209; exhibition¬ Riesman, David, 174
ism attributed to, 135-136; Hard Rites de Passage, 123-128, 124
Times Blues, 131, zyy, 136; HUAC Road of the Phoebe Snow, The, 159-165,162
and, 33; ^ Man Has Just Been Lynched Rohbins, Jerome, xxiii, xxvii, 2, 33-34, 77
(Strange Fruit), 131, 134; mentioned, Robeson, Paul, 151
161; The Negro Speaks of Rivers, 132, Rockefeller, Nelson, 41, 63
136; New Dance Group affiliation, romanticism, 69-70. See also neo¬
115; in Show Boat, 2; virtuosity as romanticism
protest in, 136, 201-202 Rooms, 98-105, 99, 704, 112, 152
Prokofiev, Sergei, 11 Ross, Nancy Wilson, 174
psychology, 77, 105 Rouzeau, Edgar T., 122
Rudko, Doris, xxi, 194
race: “Africanist” influences on Balan¬ Ryder, Mark, 15
chine, 50-51, 57-58; ballet and, 120,
121, 139, 150, 161; black press cover¬ Sabin, Evelyn, xxi
age of modern dance, xxiii, 21301; Sabin, Robert, xxi, xxiii, 44, 142, 151
civil rights movement, 115, 147-148; Sanasardo, Paul, 113, 197
consensus ideology and, 147, 149; Satie, Eric, 169-170
criticism and, xxiii; ethnic dance and, Schaeffer, Louis, 115
149-150, 164; Martin’s treatment of Schlesinger, Arthur, Jr., 10, 35, 102, 148
race, 163-164; narrative and, 158; Schmidt, Beverly, 189, 795; Red Hoop,
Negro dance as type, 131, 134-135, 190; White Figure, 190
149; race in postwar dance discourse, Schonberg, Arnold, 78
147- 150-152, 157-158- 163-164; School of American Ballet, 15, 59, 63, 161
sexuality and, 129-130, 201-202; in Scott, Marion: The Afflicted Children, 3 2
The Village 1 Knew, 93; virtuosity and, Seligmann, Kurt, 52-53, 56
131, 161-163, 164-165, 201-202, Septet, i6p, 169-170
21ZP17. See also ethnicity Severin, Reed, 57
Rainbow Round My Shoulder, 152-159, sexuality, 33-36, no, 126-130,
154, 156, 213-21ZP12 135-136, 195
rationalization, 3, 82, 84 Shakespeare, William, 27-28
Rauschenberg, Robert, 172, 175, 178, 199 Shearer, Sybil, xxvi, 167
Ravel, Maurice, 94 Sheen, Eulton J., 35-36
Ray, Man, 52, 54 Shostakovich, Dimitri, 11-12
Redfield, Robert, 116 Show Boat, xxiii, 2
regionalism, 9, 41. See also nationalism show business, 1-2
religion: expressional dance and, 71; Jew¬ Shurr, Gertrude, 15
ish ethnicity, 88-97; postwar religious Siegel, Marcia B., 28
revival, 35-36; religious television Siegmeister, Elie, 139
programming, 2; Zen Buddhism, 81, Simon, Henry, 127
I74-18I Sinatra, Frank, 106
representation: embodiment vs., xv, 18-19, Smith, Cecil, 21, 24, 142-143, 2i3ni5
82, 93; language and, 19-20; mod¬ social dance: cakewalk, 51; jitterbug, 107,
ernism and, XV, 18, 2o6n6; Zen Bud¬ 122; jive, no; quadrille, 119; square
dhism and, 176-177. See also meaning dance, 119; twist, no, 2nn9
Index 253

social protest. See political engagement television, 2, 157


Sokolow, Anna: The Bride, 94; Case His¬ Terry, Walter: on Balanchine, 43, 49-50,
tory No.-, 106-107; on commu¬ 56; on Dunham, 121; on ethnic dance,
nity, xxviii, 97-105; coverage in leftist 149; on Maslow, 92; mentioned, 74
press, xxiii; D?-ea?ns, 2ion4; The Exile, Theatre Arts magazine, xxi, xxiii
94; in Graham company, 15, 98; Thompson, Claude, 154
HUAC and, 33; Images from the Old Thompson, Stith, 31
Testament, 94; Kaddish, 89, 93-95, 95, Thomson, Virgil, 74, 212010
97; leftist attachments, 98; Lyric Suite, time art, 80, 21008
103, 112; Opus’yS, 107, 109; Opus’60, Tinsley, Jennifer, 55
107; Optis ’62, 107; Optis ’65, 107, Toumanova, Tamara, 60
I lo-i 12, III-, Opus ’65, 107, 112; Truman, Harry, lo-ii, 35, 116. See also
Revelations, 94; Rooms, 98-105, 99, postwar era
10^, 112, 152; School of American Tuhman, Harriet, 155
Ballet affiliation, 63; Session ’y8, 107; Tucker, Gregory, 89-90
Session for Eight, 107, no; Session for Tudor, Antony, 36-37, 77, 207-20804
Six, 107-111; treatment of Jewish eth¬ Turhyfill, Mark, 117
nicity, 88-89, 94“97i 2ion4; use of
African American dancers, 147; use of universal experience: Jewish ethnicity
jazz, 105, 107-109, 160 and, 88-89, 97> primitive rituals and,
Sorell, Walter, 159, 166, 194-195 125; race and, xxviii, 131, 147. See also
Southern Landscape, 139-144, 141, 213014 modernism
Speranzeva, Ludmilla, 117
Stahl, Norma Gengal, 150 Varese, Edgard, 192
Stanislavski, Constantin, 103, 20903 Vaughan, David: on artistic autonomy,
State Department “cultural ambassador” 166; objectivism and, xxii, 198
program,62, 113, 148-149 Village I Knew, The, 89-93, 9^
Stein, Gertrude, 40 Vislocky, Dorothy, 189, 195; Paraphrena-
Steuhen, Francis (Edna Ocko), 127, 132 lia, 190
Stone, Cynthia, 178
Strayhorn, Billy, 160, 163 Walther, Claire, 188
Street Scene, 107 Warburg, Edward, 40
Studio for Dance, 113 Waring, James, xix, 166
Summerspace, 171-172, 175, 178-179, Watts, Alan, 174-177, 180-181
179, 21405 Weber, Max, 3
surrealism, 43, 52-54, 81, 83, Weidman, Charles, 33-34, 184
208-209014 Welles, Orson, 74
Suskind, Peter, 132 Werth, Alexander, 11-12
Suzuki, Daisetz, 180-181 West Side Story, 107, 152
White, Josh, 132, 136
Taeuher, Sophie, 188 Whitfield, Stephen, 34-35
Taft-Hartley Act, xviii Whyte, William H., Jr., 148, 174
Tallchief, Maria, 60, 61 Wiggins, Elizabeth, 7
Tamiris, Helen, 2, 8 Wigman, Mary, 117, 184, 186, 188
Taras, John: The Minotaur, 43 Wdde, Patricia, 49
Taylor, Paul, xvii, 33-34, 63, 166, 197 Wilder, Lucy, 151
Tchelitchev, Pavel, 42-43 Williams, Lavinia, 139
254 Index

Williams, Raymond, 83 Wollf, N. S., 13 2-134


Williams, Simon, 196 Wood, Marilyn, 178
Williams, Tennessee, 4
Williams, William Carlos, 40 Yuriko, 64
Winston, Morton, 154
Wolff, Christian, 172 Zall, Deborah, 94
Wolfson, Bernice J., 31 Zen Buddhism, 81, 174-181
ABOUT THE AUTHOR

Gay Morris is a dance and art critic whose work has appeared in numerous pub¬
lications, including Dance Research, Art in America, and Body and Society. For the
past three years she has been a research fellow in sociology at Goldsmiths Col¬
lege, University of London. She is the editor of the anthology Moving Words,
Rewriting Dance (1996).
\ m ■
jr -

Cfi 1
*1 1
#•
I
J

•t ’ •

\
v
■*,
1.

«»•> Vtrr-f,
s'Tjf) ■» !?d iivi .•vJie'

• r :•<. pii* .K-:- ~ .>..’ V* Y ‘-^r,hr 1

a
<■
?<

“* * /.

^ -• ■" ^
‘ ■*:. J.
■ 7:;.'’r:':'*^
r . ; '
h

^ ' •' • -i'^ * •'.
■ VV>
■*'
/. ’X*^
» / V; '■ (
y'.*'

-.'i* ••

«• « <•
If.

!■• , -(!

"V
^'4

i‘

1^.

Ik V 1.

KS jk/^

fi V
MARYGROVE COLLEGE LIBRARY

^2 P
E
DANCE / AMERICAN STUDIES / SOCIOLOGY OF ART

The first in-depth study of the modern dance world of the 1940s and 1950s

A Game for Dancers examines the difficulties American modern dancers faced as
the Cold War took hold and the genre became institutionalized after its pioneering
phase. It draws on the sociology of Pierre Bourdieu to explore the interconnections
between art and politics while paying close attention to modern dance’s ambivalent
relationship to the market. At the heart of the book is an inquiry into modernism
itself, and how dancers struggled to meet modernist demands for abstraction and au¬
tonomy. Also crucial is the issue of embodiment, which appeared to answer modern¬
ist skepticism of representation and aid modern dance’s elusive pursuit of indepen¬
dence. Subjects include modernist dance theory, the emergence of new constituencies
including African-American choreographers, and the work of Merce Cunningham
and Alwin Nikolais, whose objectivism was declared a new modern dance vanguard
in the 1950s.

“A Game for Dancers is a tour de force: a history of postwar art and art criticism, an
aesthetic theory of embodied performance, a sociology of modernism, a theory of the
human body, and a study of American culture in the postwar period. Morris writes
with the voice of authority, but she also communicates elegantly, lucidly and persua¬
sively. A work of major scholarly achievement, it should be read closely by anybody
interested in the place of modern dance in twentieth-century American society.”
—Bryan S.Turner, Professor of Sociology,
Asia Research Institute, National University of Singapore

“Morris’s elegant study of modern dance is the first to examine the richness and com¬
plexity of the postwar dance field as a whole. Her emphasis on market, class, and
institutional issues is both sophisticated and enlightening.”
—Lynn Garafola, Professor of Dance, Barnard College/Columbia University

“An excellent introduction to the shifting terrain of postwar modernism in the


United States and the role of concert dance in defining its terms.”
—-Thomas DeFrantz, author of Dancing Revelations: Alvin Alley’s
Embodiment of African American Culture

Gay Morris is a dance and art critic whose work has appeared in
numerous publications, including Dance Research, Art in Amer¬
ica, and Body and Society. Currently she is a research fellow in
sociology at Goldsmiths College, University of London. She is the
editor of the anthology Moving Words, Rewriting Dance (1996).

ISBN 0-8195-6805-8
900 DO
WESLEYAN
UNIVERSITY PRESS 9 780819 568052

COVER ILLUSTRATION: Jack Moorc in Anna Sokolow’s Rooms (1955)- Photo: W. H. Stephan (courtesy of the Jerome Robbins Dance
Division, The New York Public Library for the Performing Arts, Astor, Lenox and Tilden Foundations).

You might also like