“To Sustain Illusion is All
That is Necessary”: The
Authenticity of Song
Performance in Early
American Sound Cinema

“To Sus tain Illusion is All That is Necessary”

Katherine Spring

I

n the second half of Radio Rhythm, a one-reel film
released by Paramount Pictures on 10 August
1929, popular radio crooner Rudy Vallee and his
band, the Connecticut Yankees, launch into a

performance of the number “You’re Just Another
Memory”. The popular song had been recorded by
Vallee and released on the Victor record label in June
of that year, and, as in that recording, Vallee in Radio
Rhythm takes up a solo on the clarinet. His onscreen
solo performance is rendered through formal strate-
gies typical of musical shorts produced during the
coming of sound: frontal staging, medium-to-long
shot scales, and minimal editing strengthen our un-
derstanding of this passage as an unadulterated
presentation of a musical performance, a direct tran-
scription of a pro-filmic event that transpired and was
recorded on a sound stage with minimal mediation.

1

Yet toward the end of the solo, our impression of an
unmodified performance comes under threat when,
after Vallee removes the instrument from his lips and
places it on the ground beside him, the sounds of the
clarinet continue on the soundtrack for nearly five
seconds.

This brief fissure in the synchronization of im-

age and audio nonetheless eludes the attention of
most audience members and in so doing fulfills the
objectives of what likely is playback, a technique
employed during the transition to sound that entailed
the prerecording of a song to which a performer
lip-synched during shooting. Playback had been

used

in

the

production

of

cinema’s

earliest

(pre–1920) sound films, but its introduction to Holly-
wood studios can be dated to October 1928, when
Douglas Shearer, a sound engineer at MGM, was
given producer Irving Thalberg’s blessing to institute
the practice during production of The Broadway
Melody.

2

Although playback became the conven-

tional method for recording song numbers in musical
films of subsequent decades, it was only one of three
practices of non-synchronous recording that Ameri-
can studios employed during the coming of sound.
The other two were dubbing, in which a voice was
added in post-production to a recorded image, and
voice-doubling, in which an actor lip-synched to a
song sung simultaneously by a vocalist located be-
yond the purview of the camera’s viewfinder but
within earshot of the microphone. Although each of
these three practices entails a different temporal
relationship between the recording of sound and that
of the image, they all produce the same pair of
illusions: embodiment, the illusion that the body seen
onscreen is the source of the voice emanating from

Katherine Spring is an Assistant Professor in the
Department of English and Film Studies at Wilfrid
Laurier University in Waterloo, Canada. Her book,
Saying it with Songs: Popular Music and the Coming
of Sound to Hollywood Cinema, is forthcoming through
Oxford University Press.
Correspondence to [e-mail]

Film History, Volume 23, pp. 285–299, 2011. Copyright © 2011 Indiana University Press

ISSN: 0892-2160. Printed in United States of America

FILM HISTORY: Volume 23, Number 3, 2011 – p. 285

the soundtrack, and simultaneity, the illusion that the
filming of the image and the recording of the voice
occurred concomitantly and in the same space.

3

The

combined effect of these illusions is an impression
of a performance’s authenticity; the final film is
deemed to have respected the spatial and temporal
integrity of a preexisting, pro-filmic event.

4

Notwithstanding the importance of the theo-

retical questions generated by practices of non-syn-
chronous recording (exemplified by the writings of
Michel Chion and Mary Ann Doane), the illusion of
authenticity served an important commercial func-
tion during the transition to sound, when the avowed
capacity for sound film to represent a preexisting
event constituted a vital component of marketing the
new medium.

5

Studio publicity and the films them-

selves positioned and promoted the singing star’s
voice as an attraction unto itself, a strategy especially
pronounced in musical shorts, whose commercial
appeal and aesthetic form derived from the onscreen
presentation of songs performed by established vo-
calists and musicians. The allure of what Donald
Crafton has dubbed “virtual Broadway” – the “simu-
lacrum of an in-person appearance” – yielded a set
of formal strategies for the translation of stage per-
formances into onscreen filmic renditions that could
circulate across the country.

6

While others have articulated these strategies

especially as they differ from the feature-length film’s
tendency to “harness” song performances to narra-
tive fiction, I am interested in considering more spe-
cifically how they position the onscreen presentation
of songs as authentic transcriptions of celebrity vo-
calists and musicians, and how the stylistic strate-
gies developed for the short film were modified to suit
the demands of feature-length narrative films that did
not necessarily showcase musical talent. Two ques-
tions guide this study. First, what formal techniques
of the medium were deployed in the service of signi-
fying the authenticity of musical performances? To
answer this question, I have selected for study a
sample of musical shorts produced by Paramount
Pictures between 1929 and 1931, primarily shot at
the company’s studio in Astoria, Queens. The prox-
imity of the studio to Broadway facilitated the casting
of musical personalities in early sound shorts, and
the resulting corpus of films suggests that the formal
devices described by Charles Wolfe in his study of
Vitaphone shorts were characteristic industry-wide
norms.

7

Like the Vitaphone shorts, the Paramount

films were organized around discrete, self-contained

song performances, what I call “star-song attrac-
tions”. This aesthetic, which I have elsewhere argued
profited the film and popular music industries alike,
was imported by feature-length films belonging to
musical and non-musical genres.

8

The translation of the star-song attraction to

non-musical features was not unproblematic, how-
ever, for the dramatic actors who starred in these
films did not perforce possess musical talents. In
order to accommodate the musical shortcomings of
dramatic actors, feature films necessarily violated the
conditions of performative authenticity deemed so
essential to song performances in short subjects.
With this in mind, my second question considers how
audiences responded to violations of authenticity
normally attributed to the star-song attraction. Here
the focus is not the failure of the illusion of simulta-
neity, as is the case in the aforementioned passage
of Radio Rhythm, but rather that of embodiment. If,
as scholars have claimed, the suppression of audi-
ence knowledge of technology is essential to main-
taining the illusions of embodiment and simultaneity,
one would expect a scandal of sorts to emerge
following the revelation of that technology to the
public.

9

Historians who describe the public’s anxiety

over practices of lip-synching and voice-doubling
during the transition to sound often cite a pseudo-ex-
posé titled “The Truth about Voice Doubling”, pub-
lished in the July 1929 issue of Photoplay and
revealing a handful of cases in which doublers were
used on Hollywood sets.

10

However, a case study of

a broader sample of public discourse, represented
by newspaper clippings in the Richard Barthelmess
scrapbook collection at the Margaret Herrick Library
in Los Angeles, illuminates a more heterogeneous
set of responses offered up by reviewers and fans.
Clippings extracted from local and syndicated news-
papers ranging from the Portland Oregonian to North
Carolina’s Greensboro Record demonstrate not only
filmgoers’ acquiescence to the practice of voice-
doubling but also their conviction in the value of
narrative plausibility and coherence over the discrete
star-song attraction, at least where non-musical fea-
ture-length films were concerned.

11

A number in narrative: the star-song
attraction

Following the release of Radio Rhythm, a reviewer for
Variety wondered “why Rudy Vallee dropped [his
trademark] megaphone to sing straight” in the film.
The answer is obvious: the film’s appeal lies in its

FILM HISTORY: Volume 23, Number 3, 2011 – p. 286

286

FILM HISTORY Vol. 23 Issue 3 (2011)

Katherine Spring

alleged transmission of the star’s image and voice,
the latter seemingly unhampered by visible appara-
tuses such as the megaphone.

12

Heard, ostensibly,

is Vallee’s “true” voice, a cultural ideal that, as Neepa
Majumdar points out, derives from modern concep-
tions of the human voice as the “key to the interiority
or to the ‘true’ self of the speaker”.

13

The transmission

of a celebrity’s authentic voice possessed commer-
cial value as well, demonstrated by the culture of the
star vocalist cultivated by media industries during the
early twentieth century across a variety of platforms,
notably Broadway, vaudeville, phonograph record-
ings, and radio. By the late 1920s, the proven appeal
of the singing star meant that, as Richard Koszarski
notes, celebrity vocalists were “better prospects for
screen stardom” than were other kinds of musicians,
like bandleaders.

14

Paramount’s musical shorts capitalized on the

established appeal of the singing star by utilizing a
range of formal devices to highlight the star-song
attraction, a self-contained unit of performance that
translated into material formats for sale (sheet music
and phonograph records) beyond the movie theater.
The prototypical approach for rendering the star-
song attraction preserved visual and sonic continuity
by way of long takes, medium-long shots, and frontal
staging.

15

A fastidious example is the one-shot film,

Favorite Melodies Featuring Ruth Etting, released 16
March 1929.

16

At the time of its release Etting was

known as the “Sweetheart of Columbia Records”,
having recorded numerous top-selling songs for the
label. Since November 1928 she had been starring
alongside Eddie Cantor in the Broadway hit,
Whoopee!, at the 1800-seat New Amsterdam Thea-
tre, where she introduced the torch song, “Love Me
or Leave Me”.

17

The filmic presentation of Etting’s

vocal performance seems to be the sole purpose of
the six-minute Favorite Melodies, which consists of a
single take that frames the star in a doorway of a
sparsely decorated set. The camera tracks briefly
toward her before settling on a medium-long shot,
where it rests for the short’s remainder. Singing in
direct address to the camera, and with minimal ges-
tures of her hands and arms, Etting performs two
songs: a ballad titled “My Mother’s Eyes” (which
audiences likely recognized as vaudevillian George
Jessel’s theme song) and a peppy dance tune,
“That’s Him Now”. During a brief pause between the
songs, Etting clears her throat, a gesture that draws
attention to the source of the film’s sound at the same
time that, in its capacity as a sonic “blemish” on the

soundtrack, it augments the impression of authentic-
ity inherent to the star-song attraction. While the
visual austerity with which the star-song attraction is
presented in Favorite Melodies might have been
owed in part to the studio’s attempt to prevent Etting
– a musical performer who never promoted herself
as an actress – from dialogue-based acting, it also
demonstrated a tendency to promote popular songs
as self-contained, purchasable commodities, a point
underscored by a reviewer for Film Daily who noted
that Etting “knows how to sell the numbers”.

18

Although the absence of even a token narrative

framework in Favorite Melodies defines the film as a
model star-song attraction, the integration of song
numbers into narrative contexts rarely threatened
their articulation of authenticity. For example, in Ethel
Merman in Her Future (released 6 September 1930),
self-contained song performances are flimsily moti-
vated by a fiction that casts Merman as a nameless
defendant in a courtroom occupied by a prosecutor,
defense attorney, and a judge whose bench towers
impossibly high above Merman’s character. The
courtroom setting gives pretext to the introduction of
two songs. The first is prompted when the defense
attorney requests of the judge, “May I ask [the de-
fendant] to tell the court in her own way just why she
is here?” The judge turns to Merman and states,
“Defendant, proceed”, and Merman launches into
“My Future Just Passed”, the lyrics of which provide
at best a feeble explanation for her ending up in a
courtroom. When she finishes singing, the judge
orders Merman’s release and adds, “Before you go,
I want you to tell me how you intend facing life. What

Fig. 1. The
camera in the
one-shot Favorite
Melodies
Featuring Ruth
Etting
(Paramount,
1929) records
the star from a
single vantage
point.

FILM HISTORY: Volume 23, Number 3, 2011 – p. 287

“To Sustain Illusion is All That is Necessary”

FILM HISTORY Vol. 23 Issue 3 (2011)

287

will you do? Where will you go?” Merman answers in
song with the lyrics, “Gonna take a train …” – an
improvised introduction to her performance of “Sing,
You Sinners”.

19

Throughout both song perform-

ances, any semblance of a subjectified narrative
space is attenuated by a static camera set-up that
frames Merman in a medium-long shot and planar
staging. Half of the film’s eight-minute duration is
rendered through this single set-up. During “My Fu-
ture Just Passed”, the film cuts to a slightly longer
shot for twelve seconds before it reverts to the origi-
nal camera position, and no editing is used through-
out the entirety of “Sing, You Sinners”. Notably, the
film also avoids cutting to images of the three char-
acters whom we assume are witnessing Merman’s
performance, resulting in the further weakening of a
narrative space that was tenuous to begin with. Fol-
lowing the performance of “Sing, You Sinners” the
short ends abruptly with a fade to black and the
appearance of the end title card without ever return-
ing to an establishing shot of the courtroom. In these
ways, the form of Ethel Merman in Her Future clearly
demarcates the boundary between the discrete star-
song performance and its narrative container.

In films that attempted to better integrate song

performances within narrative frameworks, the use of
different stylistic strategies contrasted those frame-
works with the star-song attraction as a discrete unit
of performance imbued with authenticity (and laid the
groundwork for the formal distinction between narra-
tive and numbers in musical films of ensuing dec-
ades). An illustrative example is Office Blues
(released 22 November 1930), in which Ginger Ro-
gers, then the nineteen-year-old star of the Broadway
musical comedy Girl Crazy, plays a stenographer
pining for the affections of her boss. The 29-shot film
alternates two stylistic patterns: an edited construc-
tion of narrative space, motivated mostly by the pres-
ence of an office coworker, and static long takes and
planar staging that are reserved for song perform-
ances (even when these performances are seated).
The film’s first minute, wherein Miss Gravis (Rogers)
is accosted by her coworker, Gregory (E.R. Rogers),
comprises eight shots with an average shot duration
of seven seconds and editing that is consistent with
norms of analytical and continuity editing. When Gre-
gory invites Gravis to lunch, for example, the film cuts
to her reaction in a more closely scaled medium shot.
A brief appearance by Gravis’s boss, Jimmy (Clair-
borne Bryson), gives the stenographer sufficient rea-
son to launch into “We Can’t Get Along”.

20

With the

exception of a three-second cutaway to an image of
Gregory gazing at Gravis as she sings, the entirety
of the song’s duration (2:09) is presented through a
single medium shot of Gravis.

The second song sequence removes us fur-

ther from the narrative space established in the film’s
opening minutes. Alone in the office after Gregory
leaves for lunch, Gravis begins to sing a fantasy
letter, “Dear Sir”, and the film dissolves to a surreal
set displaying a chorus line of women standing on a
stage built to resemble a notepad. In medium-long
shot, Gravis sings while standing in front of the
women; Jimmy then enters from stage left, singing
“Dear Miss, I just received your note”, and joins in the
fantasy song that codes romance as a business
transaction (“I’d like to take your tip and form a
partnership”). At the end of the song, the couple
kisses and the film dissolves back to Gravis seated
at her desk. As though to suggest that her fantasy
was shared with the object of her affection, Jimmy
appears with a letter in hand and, after some confu-
sion on Gravis’s part, announces, “Perhaps we’d
better discuss this in my office”. The couple exit and
– in a move that betrays the film’s pre-Code vintage
– we cut to the film’s final image, a door sign that
reads, “Busy Dictating”. The resolution provided by
this final event, in which Gravis’s wishes are fulfilled,
imparts a degree of dramatic unity that resonates
with the classical, unified construction of feature-
length films. But couched within this unified frame-
work resides a stylistic template for maintaining the
illusion of authenticity of song performance.

Signature instruments: musical
tools of authenticity

The early soundtrack comprised not just singing, of
course, but also the sounds of musical instruments,
and those sounds aided in the construction of
authenticity. An oft-cited account penned by Max
Steiner recalls the necessary onscreen depiction of
musical instruments whose sounds were heard on
the soundtrack:

Music for dramatic pictures was only used
when it was actually required by the script. A
constant fear prevailed among producers, di-
rectors and musicians, that they would be
asked: Where does the music come from?
Therefore they never used music unless it
could be explained by the presence of a
source like an orchestra, piano player, phono-

FILM HISTORY: Volume 23, Number 3, 2011 – p. 288

288

FILM HISTORY Vol. 23 Issue 3 (2011)

Katherine Spring

graph or radio, which was specified in the
script.

21

Steiner’s recollection was shared by film histo-

rian Irene Kahn Atkins:

[The coming of sound marked] the beginning
of the era of a shot that was to be repeated in
countless films: when an excuse for music was
needed, someone was seen turning on a pho-
nograph, putting the needle on the record, and
listening as the appropriate song or instrumen-
tal was heard on the soundtrack.

22

Perhaps because they did not operate under

the demands of classical narration, musical shorts
seem to have been exempted from the requisite
onscreen depiction of instruments. For example, al-
though the sounds of a piano and strings are heard
clearly on the soundtrack of Favorite Melodies Fea-
turing Ruth Etting, nary an instrument is seen; the
same observation holds for Office Blues and Ethel
Merman in Her Future, even when a relatively full
orchestration appears on the soundtrack. Instru-
ments also are absent from the images of Insurance
(August 1930), an Eddie Cantor sketch set in an
insurance office, although the star’s presentation of
“Now That the Girls Are Wearing Long Dresses” is
accompanied by the sounds of a jazz orchestra
including a string and horn section.

Since they did not necessarily supply a visual

pretext for music heard, what functions did musical
instruments serve when they appeared onscreen in
musical shorts? I suggest two, each of which en-
hanced the authenticity of song performance. First,
the display of instruments on screen facilitated audi-
ence identification with celebrities whose star perso-
nas were predicated at least in part on their use of
instruments. Like the voice, the musical instrument
could serve to signify a star’s authentic presence.
Radio Rhythm is a case in point. By the time the film
was released in August 1929, Vallee had become a
radio sensation, having signed earlier that year an
exclusive contract with NBC that resulted in weekly
national broadcasts sponsored by Fleischmann’s
Yeast. On the Fleischmann’s Yeast Hour program,

Fig. 2(upper). Rudy Vallee poses with saxophone for a news
photographer, ca. late 1920s/early 1930s. [Courtesy Library of
Congress.]

Fig. 3(lower). Rudy Vallee faces the camera in direct address
in the third shot of Radio Rhythm (Paramount, 1929).

FILM HISTORY: Volume 23, Number 3, 2011 – p. 289

“To Sustain Illusion is All That is Necessary”

FILM HISTORY Vol. 23 Issue 3 (2011)

289

Vallee distinguished himself not only by emphasizing
vocal solos more than any other bandleader but also
by developing a breathy, soothing style of singing –
crooning – that fans quickly interpreted as his expres-
sion of sincere emotion.

23

Vallee’s saxophone was a

defining element of his public persona, as evidenced
by its frequent citation (sometimes by Vallee himself)
in accounts of his musical education prior to star-
dom.

24

In addition, Vallee’s distinctive style of singing

owed to his emulation of techniques for playing the
saxophone – specifically, the technique of vibrato,
which he discovered was “particularly effective in
conveying emotional involvement”.

25

Vallee’s voice,

coupled with his adept handling of the saxophone
(and to a lesser degree, the clarinet), served as
immediate signifiers of his “true” presence.

The authenticity of this presence in Radio

Rhythm is strengthened by formal means. Vallee
appears first when a large circular panel, designed
to resemble the radio speaker seen in the previous
shot, parts down the center and recedes to reveal the
star standing and smiling, one hand in his jacket
pocket and the other cradling an alto saxophone that

hangs from his shoulder. A cut to a medium shot
briefly highlights the star before the film returns to the
initial camera set-up. After minimal editing around
three vantage points, Vallee introduces the song
“Honey”, a hit ballad that he and his orchestra had
recorded for Victor six months prior to the film’s
release. Vallee’s singing is rendered through a static
medium shot that ends only when “Honey” enters
into an instrumental break, followed by a seamless
segue into a second song, “You’re Just Another
Memory”. Now Vallee picks up his clarinet and starts
to play, and the film remains in the medium-long shot
set-up until Vallee begins warbling a verse. The pat-
tern established so far – a medium-long shot for
instrumental breaks, medium shots for vocal pas-
sages – changes only during the third song, the
peppy dance number titled “You’ll Do It Someday,
So Why Not Now?” With band members singing in
chorus and effecting synchronized hand gestures,
the film cuts to a medium shot when Vallee begins a
solo on the alto saxophone. That the film denies a
medium shot of a subsequent soloist, the tenor saxo-
phonist, and instead remains fixed on an establishing

Fig. 4. A

publicity photo

snapped during

production of the

Broadway show

Present Arms

(1928) captures

Richard Rodgers

on his usual

perch at the

piano. Shown

from left, back

row: Flora Le

Breton, Charles

King, Joyce

Barbour; front

row: Richard

Rodgers, Lorenz

Hart.

[Courtesy

Photofest.]

FILM HISTORY: Volume 23, Number 3, 2011 – p. 290

290

FILM HISTORY Vol. 23 Issue 3 (2011)

Katherine Spring

shot is indicative of a formal pattern: Radio Rhythm
reserves medium shots exclusively for the attraction
constituted by Vallee singing and performing solos
on his signature alto saxophone. This shot scale,
combined with long takes and frontal staging, main-
tains audience attention on the star’s performing
body.

The onscreen appearance of hallmark instru-

ments as a means of conveying the authenticity of
embodiment was not exclusive to films starring vo-
calists. For example, in Makers of Melody (June 1929)
– a short in which famed songwriters Richard Rodg-
ers and Lorenz Hart offer a fictionalized account of
their songwriting inspirations – Rodgers first appears
seated at a piano, a position he tended to occupy in
publicity photographs. Nor was the practice confined
to Paramount’s earliest sound films. In A Rhapsody
in Black and Blue, released in September 1932, Louis
Armstrong is introduced singing and playing his sig-
nature muted trumpet (albeit in an utterly surrealist
fantasy setting), and the second shot of Bundle of
Blues (September 1933) begins with a close-up of a
pianist’s hands and tracks back until Duke Ellington
is revealed to be conducting his orchestra from his
customary perch on the piano bench.

All of these examples suggest that just like the

human voice, musical instruments functioned as ex-
ternal signifiers of a musician’s interior state, a point
underlined by philosopher Lydia Goehr in The Quest
for Voice. “At the moment of musical playing”, she
writes, “[the instrument’s] physicality, we could say,
becomes inseparable from the pure expressiveness
of our ‘souls’”.

26

Indeed, the attention accorded by

film form to the depiction of instruments in musical
shorts seems to function less to preclude the sort of
audience confusion that Steiner (and later Atkins)
identified and more to buttress the illusion of embodi-
ment. By showing stars playing their instruments, the
films foreclose the threat of “the mysterious”, as
Michel Chion has described it: “The face of a musi-
cian in the process of playing a piano, violin, harp or
any other instrument that does not involve the mouth
or breath is a mystery. It’s like a closed box, and we
cannot know if the hands are moving with or without
the participation of the mind and emotions.”

27

The

foregrounding of a star’s non-vocal musical talents
facilitated the illusion of embodiment.

As to their second function, musical instru-

ments also served to establish a diegetic setting, but
they frequently did so by exploiting ethnic and racial
stereotypes. Because the use of iconic musical in-

struments to code diegetic space in ethnic or racial
terms would become a conventional technique of
musical scoring in later feature-length films, it de-
serves brief consideration here. A good example is
found in Ol’ King Cotton, a baldly racist short that
starred among its African American cast the bass-
baritone

vocalist

George

Dewey

Washington

(dubbed the “colored Al Jolson”).

28

Released 27

December 1930, the short opens on a long shot of
what is presumed to be a Southern plantation; in the
background can be seen an expansive cotton field.
In the mid-foreground, Washington slouches at the
base of a sycamore tree, flanked on the right by a
banjo player and on the left by a man sitting with his
back to the camera and playing what appears to be
either a tenor guitar or banjo. Remaining in this
position, Washington sings the title song, a folk blues
number whose lyrics lament the hardship of picking
cotton (“I get so weary/ Of pickin’ and a-pickin’/ That
ol’ king cotton / Ain’t nothin’ but the weary blue”).
While he sings, the sounds of a jazz orchestra can
be heard on the soundtrack (including a violin, a
double-bass, a muted trumpet, and drums), but the
only instruments that appear in the image are the
banjo and tenor guitar (or tenor banjo) – two folk
instruments common enough in jazz bands of the
1920s and 1930s but also generally associated with
African American music thanks to the banjo’s appro-
priation by minstrel shows of prior decades and the
tenor guitar’s role in blues music.

29

The only other

instrument that appears in the film is a harmonica,
which a boy plays in accompaniment to Washing-
ton’s reprisal of “Ol’ King Cotton”, sung in the back-
room of a warehouse where Washington has taken
a job after traveling to a city “in the North”. Now the
song has acquired connotations of the character’s
nostalgia for home “down South”, and these conno-
tations find appropriate musical underscoring by the
harmonica, an instrument that Harold Courlander
called in his 1963 work, Negro Folk Music, U.S.A.,
“probably the most ubiquitous of Negro folk instru-
ments”.

30

The presence of instruments commonly

identified with African American culture of the early
twentieth century here helps to signify the authenticity
of Washington’s performance.

Not quite a scandal: responses to
the revelation of voice-doubling

The star-song attraction was valuable in both aes-
thetic and commercial terms. It packaged song per-
formances as discrete, authentic transcriptions

FILM HISTORY: Volume 23, Number 3, 2011 – p. 291

“To Sustain Illusion is All That is Necessary”

FILM HISTORY Vol. 23 Issue 3 (2011)

291

addressed to audiences who were given occasion
not only to enjoy celebrity performances through
filmic “virtual Broadway” but also to purchase sheet
music, piano player rolls, and phonograph records
in order to sustain the pleasure of the star-song
attraction beyond the movie theater. The perceived
significance of the star-song attraction in musical
shorts helps to explain its frequent incorporation into
feature-length, non-musical films of the transitional
era, including melodramas, westerns, and come-
dies.

31

However, whereas the established talents

and personas of the stars of musical shorts guaran-
teed audience confidence in the authenticity of song
performances, feature-length films conscripted dra-
matic actors into roles that required singing and/or
musical aptitudes – talents that did not generally
constitute the actors’ public personas and therefore
gave audiences reason to doubt the authenticity of
the song presentations. There was also the problem
of song motivation. The classical norms of narration
that the feature-length sound film had inherited from
the period of late silent cinema meant that, unlike in
musical shorts, the integration of self-contained song
performances required plausible motivation. One so-
lution came by way of writing musical tendencies into
a character, a technique that was easily assimilated
when a silent film star demonstrated an ability to sing.
In Sunny Side Up (1929), as Richard Maltby reminds
us, Janet Gaynor’s character is shown early on to be
an individual “given to expressing her innermost feel-
ings lyrically”, which justifies her character’s pen-
chant for breaking into song throughout the film.
Fortunately, Gaynor was deemed to possess at least

a tolerable singing voice.

32

But whereas numerous

silent stars succeeded in making the transition from
silence to speech, fewer proved capable of shifting
to the far more specialized act of singing. Their
deficits inspired the practices of playback, dubbing,
and voice-doubling.

When voice-doubling was used for dialogue,

as Donald Crafton points out, many fans accepted
the practice “simply as a component of Hollywood’s
trickery, not as false advertising or malicious deceit”,
but rather analogous to sound effects or make-up.

33

But song performances raised the stakes of authen-
ticity because the act required specialized skills and
also capitalized on the culture of the singing star.
How did audiences of early sound films respond to
the revelation of musical voice-doubling?

The most frequently cited article on this subject

is Mark Larkin’s aforementioned “The Truth about
Voice Doubling”, published in Photoplay in July 1929.
The four-page piece revealed what many filmgoers
already suspected: motion picture producers were
using voice-doublers (or “dubbers”) who sang off-
screen as the camera recorded the actors lip-
synching to songs – and they were doing the same
with off-screen instrumental musicians. Larkin’s
piece identified several actors who used doubles but
dedicated the most print to Richard Barthelmess,
star of the 1929 crime drama Weary River, about a
prisoner who warbles his way out the jailhouse and
into the limelight of the vaudeville stage. Elsewhere I
have discussed the commercial and narrative impor-
tance of popular songs in Weary River; Barthelmess’s
character sings the titular theme song four times over

Fig. 5. An

excerpt from the

article “The Truth

about Voice

Doubling’,

published in the

July 1929 issue

of Photoplay.

FILM HISTORY: Volume 23, Number 3, 2011 – p. 292

292

FILM HISTORY Vol. 23 Issue 3 (2011)

Katherine Spring

the course of the film, and each performance is
treated as a star-song attraction, though the individ-
ual responsible for singing in Barthelmess’s place
was Johnny Murray, a cornetist and vocalist who
worked at Hollywood’s Cocoanut Grove in the
1920s.

34

Although the Photoplay article poses as an

educational piece about details of studio practice
(“Here are all the facts about sound doubling, and
how it is done”, reads a subheading), the likelihood
that the magazine’s editors were banking on the
commercial value of letting the public in on a secret
is evidenced by the strategic timing of its appear-
ance: the article was published after several months
of speculation in the press and fan magazines about
the authenticity of Barthelmess’s singing voice.

The silent film star’s first sound film, Weary

River premiered in New York on 10 February 1929,
on the heels of a publicity campaign that promoted
the vocal talents of its star. An advertisement in Film
Daily situated the stern face of Barthelmess along-
side the caption, “You always knew he was the
screen’s greatest fighting lover …. You always knew
he was the greatest male star in pictures … but YOU
DON’T KNOW NOTHIN’ YET”, a bald reference to Al
Jolson’s legendary phrase from The Jazz Singer. In
a preview trailer, Barthelmess had even invited audi-
ences to hear him sing in his new production.

35

Suspicions that fakery was responsible for the star’s
singing voice surfaced in the weeks following the
film’s release, but the reviews were not all that dis-
paraging. In the 18 February edition of the Dallas
News, an author who signed off as “J.R. Jr.” claimed
that Barthelmess did not do his own singing, but that
the “Song Shenanigan [is] Really an Asset to ‘Weary
River’”.

36

The review instigated a swift defense from

Joseph E. Luckett, the Dallas branch manager of
First National, and James O. Cherry, owner of the
local Publix theater where the film screened. A com-
mittee of local musicians was appointed to analyze
and assess the matching of Barthelmess’s physical
performance with the sonic qualities of the voice on
the soundtrack. Meanwhile, the film’s release at the
Metropolitan Theatre in Houston generated similar
rumors, and another committee of musicians was
coordinated

there.

The

committees’

“studies”

yielded different opinions. Some insisted it was the
star’s voice they heard on the soundtrack, but others
were more skeptical. One member quipped, “If
Barthelmess actually does sing he should be in the
Smithsonian Institute. … The intensity of breathing so
essential to singing is entirely lacking”, and another

noted, “Barthelmess moves his lips with no move-
ment at all in his throat”.

37

But even committee mem-

bers who declared the singing fake observed that it
was so “well done that it does not detract from the
picture’s appeal”.

38

J.R. Jr. continued to defend the

practice of doubling on the grounds that it synthe-
sized for audiences two filmic diversions: a pleasant
voice and a “gorgeous face”.

39

The types of responses printed in the Dallas

News and Houston Chronicle were characteristic of
those that emerged in newspapers across the United
States. Various attitudes developed, only one of
which was admonishing of Barthelmess and/or First
National. A columnist for the Louisville Times warned,
“If (Barthelmess) doesn’t sing ‘Weary River’ then I,
for one, will never trust the screen again unless it
features someone like Al Jolson, who has already
been proved and found ‘not wanting’”.

40

A number

of fans were equally perturbed. Jean Betty Huber, of
Morris Plains, New Jersey, submitted this letter to the
editor of Picture Play magazine:

Please listen, you movie folk. We bow in hom-
age to your beauty and acting ability. We pay
hard-earned money to see your pictures – and
now to hear you speak as well – you who have
our complete admiration and support. Where
would you be without it? Now tell me that. If
these little Dick Barthelmess stunts are to con-
tinue, then here is one fan who is giving you a
fond farewell.

41

Later that year, Pearl H. McLaughlin, of Ham-

ilton, Ontario, wrote in to complain that the use of a
double in Weary River “entirely spoiled the picture”
for her. “My collecting days are over”, she an-
nounced, and “I think there is something wrong
somewhere and it does not lie with the star, but
perhaps in the studio system there is an error”.

42

When Bob Allen of Waynesburg, Pennsylvania, no-
ticed poor attendance at Barthelmess’s subsequent
film, Drag (1929), he submitted to Picture Play his
prediction that “the public is not going to put up with
Dick Barthelmess going through a picture on some-
body else’s shoulders”.

43

The complaints of a few disgruntled fans not-

withstanding, the majority of responses were gener-
ally supportive of the practice of voice-doubling –
though for different reasons. Numerous film review-
ers extolled the producers of Weary River for finding
a talented singer to substitute for Barthelmess’s
voice on the soundtrack. A columnist for the Hudson

FILM HISTORY: Volume 23, Number 3, 2011 – p. 293

“To Sustain Illusion is All That is Necessary”

FILM HISTORY Vol. 23 Issue 3 (2011)

293

Dispatch of Union City, New Jersey, confessed to not
knowing whether Barthelmess played the piano and
sang for the recording of the film, but added, “Who
cares. It is also well known that many of the most
charming foreign players in the talkies do not speak
the beautiful English that issues from the screen, and
it’s well for the ear that they don’t.”

44

Along similar

lines, in a piece titled “The Vocal Double”, an uniden-
tified author wrote, “That tenor [who sang for Barthel-
mess] must be safeguarded and kept on the job – it
would be disastrous if Mr. Barthelmess were to have
a different voice in each picture”.

45

These responses point up the doubled voice

as a feature that functioned at once as a source of
pleasure and a form of combined exchange value,
since profits could be derived from the commoditi-
zation of both star and song.

46

Similarly, the merged

value of star and voice was captured by extra-textual
consumer commodities generated by Weary River.
Sheet music for the theme song featured on its cover

an image of Barthelmess’s face, and phonograph
recordings were issued by no fewer than twenty
record labels in 1929 alone (including a Gene Austin
pressing for Victor).

47

The majority of recordings were

made on 1 February and served as well-timed adver-
tisements for the film set for premiere nine days later.

Another

category

of

responses

praised

Barthelmess’s and the film’s accurate synchroniza-
tion for their capacity to conceal the process of
doubling. A month following the printing of Pearl
McLaughlin’s grievance against the studio system,
Picture Play ran a letter from a reader who came to
Barthelmess’s defense. “Only a Girl of Fifteen”, from
Branford, Connecticut, stated, “I think that he ought
to be given credit for fooling every one as he did in
‘Weary River’. In my opinion, it would be harder to
have a double than to do the work itself.”

48

Similar

appraisals emerged in Los Angeles papers when the
film premiered on the West Coast on 1 March. Col-
umnist Marquis Busby of the Los Angeles Times
wrote, “Whether or not Barthelmess actually sings
the song I do not know. If he does not the voice
doubling is amazingly perfect.”

49

The Hollywood

News printed, “Had it not been for the rumor that he
didn’t [sing the theme song], no one would have
known the difference so perfect is the synchroniza-
tion”.

50

An article titled “Film’s Theme Song Stirs

Controversy” appeared one month later in the Los
Angeles Examiner, insisting similarly that if Barthel-
mess did not sing, “the synchronization is unbeliev-
ably accurate”.

51

It may be tempting to attribute these echoed

statements to the press’s coordinated promotion of
the film industry, given that Warner Bros. took out
paid advertisements regularly in these same Los
Angeles papers. But emphatic praise for the film’s
accurate synchronization appeared in other newspa-
pers, too. The Cleveland Plain Dealer noted that if
Barthelmess is not singing, then “there is the most
nearly perfect job of film doubling the screen has ever
heard”, while a columnist with the Greensboro Re-
cord of North Carolina proclaimed, “All we’ve got to
say is that it was darn clever faking and we’ll never
be quite convinced of it until Dick himself tells us he
wasn’t warbling the melody of the popular dance
tune”.

52

In the Louisville Courier of Kentucky, a re-

viewer noted that Barthelmess “gives the illusion that
he sings admirably and that is all that is necessary.
To sustain illusion is all that is necessary”.

53

The

attitude exemplified by these reviews attests to what
Crafton has described as the public’s fascination

Fig. 6. Sheet

music of the title

song of Weary

River (1929)

features the

portrait of star

Richard

Barthelmess.

[Author’s

collection.]

FILM HISTORY: Volume 23, Number 3, 2011 – p. 294

294

FILM HISTORY Vol. 23 Issue 3 (2011)

Katherine Spring

with the new medium of synchronized sound: clearly
the marvel of electrical synchronization prevailed
over the potential threat posed by the practice of
doubling.

Yet another attitude resonates strongly with

Crafton’s observation of the public’s acceptance of
voice doubling as yet another element in the cin-
ema’s system of artifice. In the 12 May issue of the
Portland Oregonian, reviewer Stanley Orne asserted,
“In the matter of voice-doubling in the movies …
there can be only one crime, the crime of being
detected in your voice-doubling. The creed of the
dumb movie player might well be, Let not thy public
know what thy voice-double doeth”. Orne then ques-
tioned why audiences would complain about dou-
blers when they absolve filmmakers of other trickery:

It is well known that someone as obscure as
Tillie Plutz has such beautiful feet that hers are
invariably substituted in close-ups for face;
that the clothes which seem white in the pic-
tures actually are yellow in the Hollywood stu-
dios; that daredevils risk their lives to spare the
necks of handsome men who get screen credit
for bravery; that the movie magazines are
stuffed with stories written by journalists but
signed with the names of actors and actresses
who can’t write ten consecutive words sensi-
bly. Why, then, should we be squeamish over
the voice-doubling fraud? Frankly, I’m not.

54

Similarly, a reporter for the Cleveland News

wrote:

The great ocean liners we at times see being
tossed about by the terrific mid-Atlantic waves
are but toys bobbing about in a room about as
large as your home’s smallest bedroom.
Knowing this we still visualize the craft as being
an ocean liner on which maybe our hero and
heroine had gone aboard a short time before
as the great ship actually stood at its dock. …
Maybe if we did hear [Barthelmess’s] very best
vocalizing we’d be in something of a panic to
get the double back on the job.

55

The December 1929 issue of Picture Play

printed a letter from an individual residing in England
who pointed out that in Weary River, “the singing of
the convict had to be of such a quality as to cause a
sensation; a merely pleasant voice would have weak-
ened the conviction of the story”.

56

Voice-doubling of the star-song attraction was

understood by these filmgoers as an example of the
medium’s requisite dependency on artifice in the
name of narrative plausibility. In these terms, public
and press discourse mirrored a key debate about
sound recording that was transpiring among Holly-
wood studio technicians during this period. The de-
bate centered on two approaches to sound
recording: fidelity set as its goal “the perfectly faithful
reproduction of a spatiotemporally specific musical
performance (as if heard from the best seat in the
house)”, whereas intelligibility sought “legibility at the
expense of material specificity, if necessary”.

57

Fidel-

ity has been identified by scholars as the predomi-
nant goal during the initial phase of the transition to
sound, with intelligibility finally prevailing in the early
1930s. But, as James Lastra points out, the two
approaches actually worked in tandem with one an-
other; he cites Carl Dreher, a sound engineer at RKO
who maintained that, “since the reproduction of
sound is an artificial process, it is necessary to use
artificial devices in order to obtain the most desirable
effects. … This may entail a compromise between
intelligibility and strict fidelity.”

58

Although intelligibil-

ity was here conceived of as audibility of dialogue,
and not necessarily narrative plausibility, the ration-
ale behind Dreher’s insistence mirrored that which
underpinned some of the audience responses to
voice-doubling – namely, what one sacrificed in pure
fidelity (or authenticity) was gained in intelligibility (or
narrative plausibility).

In 1929, therefore, while some sound engi-

neers aspired to the representation of pure percep-
tual fidelity, audience members embraced film as a
medium subject to illusory codes both visual, as in
the use of daredevil stuntmen, and sonic, as in the
use of vocal-doubling. Their embrace was constitu-
tive of what Crafton has shown was the public’s faith
in the new technology of electric sound itself, which
the public made clear “would be acceptable in fea-
tures only if it did not interfere too much with the
traditional storytelling movie”.

59

The purchase on

authenticity made by sound in general, and the star-
song attraction in particular, could be sacrificed for
the sake of the preservation of classical narrative
coherence.

The fate of voice-doubling and
illusions of authenticity

In contrast with the feature film’s emphasis on narra-
tive plausibility, musical shorts authenticated star-

FILM HISTORY: Volume 23, Number 3, 2011 – p. 295

“To Sustain Illusion is All That is Necessary”

FILM HISTORY Vol. 23 Issue 3 (2011)

295

song attractions by using a set of stylistic techniques
that included long takes, frontal staging, and me-
dium-long shots. In so doing, the shorts facilitated
the illusions of embodiment and simultaneity, the two
conditions essential to creating the impression of
performative authenticity in sound cinema. When
star-song attractions translated to feature-length
films that showcased performances by dramatic
(and not necessarily musical) actors, the appear-
ance of authenticity was attained through practices
of voice-doubling and, with increasing frequency as
the sound era took hold, playback and post-dub-
bing. Indeed, by 1931, voice-doubling was all but
abandoned by Hollywood’s major studios, a shift that
might easily be attributed to a prudent reaction on
the part of producers toward disapproving fan mail
and journalistic assessments.

60

There was also the

problem of the cultural value associated with dou-
bling, articulated best by “sophisticate” filmmaker
Ernst Lubitsch in a syndicated article from Septem-
ber 1929. Lubitsch told the journalist:

No matter if the synchronization with the
player’s lips is mechanically perfect the effect
is bad. The feeling behind the words does not
coincide with the expression on the actor’s
face. Besides being artistically bad, I believe
that pictures in which doubles are used arouse
ill feeling on the part of audiences. If they know
that a double is speaking for an actor they
believe that they have been cheated. And they
have.

61

As the foregoing discussion has suggested,

however, these assessments do not accurately rep-
resent the scope of responses engendered by the
unveiling of voice-doubling to the public. Because
many filmgoers considered doubling a pleasurable
and even essential technique for some of Holly-
wood’s earliest sound films, it is perhaps more judi-
cious to attribute the eventual demise of the practice
not to negative reactions on the part of select fans or
filmmakers (like Lubitsch) but rather to institutional
practices, such as the studios’ cultivation of stables
of singing actors in the early 1930s, or their use of
new technologies that facilitated the much easier
process of post-production dubbing. That dubbing
quickly replaced voice-doubling as a norm of pro-
duction in Hollywood musicals thus underlines a
discrepancy that existed between the interests of
studio producers, whose goal of maintaining the

star-song attraction and the illusion of authenticity
through ever-efficient means precipitated the demise
of doubling and the adoption of post-production
dubbing, and the various expectations among audi-
ences.

In fact, varied responses and expectations with

respect to performative authenticity have resurfaced
in the recent history of popular music. Writing in 1992,
Steve Wurtzler observed the “scandals” associated
with contemporaneous revelations of lip-synching
among popular vocalists. In 1990, pop music duo
Milli Vanilli were found to have employed doubles for
the recording of their multi-million-selling album, Girl
You Know It’s True, and two years later, Luciano
Pavarotti “pulled a Milli Vanilli” in a performance in
Italy.

62

More recently still, audiences booed pop star

Ashlee Simpson off the stage of Saturday Night Live
when her lip-synched performance was laid bare to
the live studio audience (2004), and they chastised
Chinese organizers of the Summer Olympics for their
coordination of the lip-synching of a nine-year-old
performer (2008). Clearly, authenticity remains in
some domains a benchmark of value. On the other
hand, popular musicology has taught us that, from
the early 1990s through to today, many audience
members have adopted an ironic and knowing
stance (or “hipness”, as Lawrence Grossberg has
put it) with respect to the practice of lip-synching;
even the fans of Milli Vanilli remained unfazed while
around them buzzed music critics who cried foul.

63

Lip-synched musical performances at the Super
Bowl are now casually dismissed as convention;
previously unknown individuals attain celebrity stat-
ure on YouTube for their achievements in accurate
lip-synching to songs; and the use of pre-recorded
songs and post-production dubbing in film are widely
acknowledged and accepted practices. Considered
from this perspective, an account of the discourses
of authenticity of song performances in early sound
films provides us with a more accurate historical
context in which to assess claims about the novelty
of today’s so-called hip or ironic audiences.

Acknowledgements: The author wishes to thank Vincent
Bohlinger, Jinhee Choi, Lea Jacobs, Rob King, and Jeff
Smith for their insightful comments on earlier versions
of this paper. The author gratefully acknowledges that
financial support for this research was received from a
grant funded by Laurier Operating funds and the SSHRC
Institutional Grant award to Wilfrid Laurier University.

FILM HISTORY: Volume 23, Number 3, 2011 – p. 296

296

FILM HISTORY Vol. 23 Issue 3 (2011)

Katherine Spring

1.

In his seminal essay on Vitaphone shorts and The
Jazz Singer, Charles Wolfe distinguishes between
two aesthetic tropes situated at polar ends of a
spectrum of strategies used for the rendering of
musical performances in early sound films: the
“(filmic) presentation of a musical performance”,
which reiterated the techniques of contemporaneous
musical stage presentations (e.g. through direct
address); and the “harnessing of that performance
for a narrative fiction”, which entailed the repre-
sentation of a character’s (or characters’) subjecti-
fied space, an effect often achieved through the
codes of continuity editing. Charles Wolfe, “Vita-
phone Shorts and The Jazz Singer”, Wide Angle 12.3
(July 1990): 71.

2.

Richard Barrios, A Song in the Dark: The Birth of the
Musical Film (New York: Oxford University Press,
1995), 64–67; Mark A. Vieira, Irving Thalberg: Boy
Wonder to Producer Prince (Berkeley: University of
California Press, 2010), 92.

3.

Michel Chion dubs this combined effect “synchre-
sis”, which he defines as “the spontaneous and
irresistible mental fusion, completely free of any logic,
that happens between a sound and a visual when
these occur at exactly the same time”. For the pur-
poses of this paper, I find it useful to separate the
two misperceptions: one relating to the body that is
assumed to produce the sound, and another relating
to the temporal synchronization of sound and image.
Michel Chion, Audio-Vision: Sound on Screen
Claudia Gorbman (ed. and trans.) (New York:
Columbia University Press, 1994), 63.

4.

I hasten to qualify my use of the term authenticity.
First, I do not intend to evoke the ways in which the
word is used in scholarship of popular music and
music criticism, wherein it refers to “a socially
agreed-upon construct in which the past is to a
degree misremembered”, as Richard A. Peterson
has put it. Neither is my use of the term meant to
serve as a synonym for “fidelity”, a word that in
historical studies of film sound has come to represent
one of two approaches to sound recording – the
other being “intelligibility” – debated among sound
technicians working in Hollywood in the late 1920s
and early 1930s. For critical accounts of the models
of fidelity and intelligibility, see Rick Altman, “Sound
Space”, in Rick Altman (ed.), Sound Theory/Sound
Practice (New York: Routledge, 1992), 46–64; James
Lastra, Sound Technology and the American Cin-
ema: Perception, Representation, Modernity (New
York: Columbia University Press, 2000), 138–153;
Steve J. Wurtzler, Electric Sounds: Technological
Change and the Rise of Corporate Mass Media (New
York: Columbia University Press, 2007), 229–278.
See also Richard A. Peterson, Creating Country Mu-

sic: Fabricating Authenticity (Chicago: University of
Chicago Press, 1997), 5.

5.

Chion, Audio-Vision, 1–65 and passim; Mary Ann
Doane, “The Voice in the Cinema: The Articulation
of Body and Space”, Yale French Studies 60 (1980):
33–50. The marketing of sound film as a novelty is a
thread that weaves through Donald Crafton, The
Talkies: American Cinema’s Transition to Sound,
1926–1931 (New York: Charles Scribner’s Sons,
1997), and is especially pronounced in chapter 12,
“The New Entertainment Vitamin: 1928–1929”,
271–312.

6.

Crafton, The Talkies, 11, 63–88, 313–15.

7.

Wolfe, “Vitaphone Shorts and The Jazz Singer”.

8.

Katherine Spring, “Pop Go the Warner Bros., et al.:
Marketing Film Songs during the Coming of Sound”,
Cinema Journal 48.1 (Fall 2008): 68–89.

9.

For examples of this claim, see Michel Chion, The
Voice in the Cinema, Claudia Gorbman (trans.) (New
York: Columbia University Press, 1999), 128–136;
Doane, “The Voice in the Cinema”; Mary Ann Doane,
“Ideology and the Practice of Sound Editing and
Mixing”, in Teresa de Lauretis and Stephen Heath
(eds.), The Cinematic Apparatus (New York: St. Mar-
tin’s Press, 1980), 47–56; Rick Altman, “Moving Lips:
Cinema as Ventriloquism”, Yale French Studies 60
(1980): 67–79; Alan Williams, “Is Sound Recording
Like a Language?” Yale French Studies 60 (1980):
51–66.

10.

Wurtzler, Electric Sounds, 272–274; Neepa Majum-
dar, Wanted Cultured Ladies Only! Female Stardom
and Cinema in India, 1930s–1950s (Champaign, Ill.:
University of Illinois Press, 2009), 178. The Photoplay
piece has become a principal, if not exclusive, source
for scholars presumably owing to its accessibility in
Miles Krueger’s anthology of Photoplay documents
from the period. Mark Larkin, “The Truth About Voice
Doubling”, Photoplay (July 1929): 32–33, 108–110,
reprinted in Miles Kreuger (ed.) The Movie Musical
from Vitaphone to 42nd Street: As Reported in a Great
Fan Magazine (New York: Dover Publications, 1975),
34–37.

11.

One of the first entries into the Special Collections
at the Margaret Herrick Library, the Richard Barthel-
mess scrapbooks (hereafter RBS) were allegedly
compiled by Barthelmess’s mother, actress Caroline
Harris Barthelmess. Given the extraordinary quantity
of reviews and reports contained in the fifty-volume
set, it is more likely that the content was supplied by
a press clipping service working under the mother’s
employ.

12.

Quoted in Edwin M. Bradley, The First Hollywood
Sound Shorts (Jefferson, North Carolina: McFarland
& Co., 2005), 231.

Notes

FILM HISTORY: Volume 23, Number 3, 2011 – p. 297

“To Sustain Illusion is All That is Necessary”

FILM HISTORY Vol. 23 Issue 3 (2011)

297

13.

In European modernity, Majumdar argues, “the
human voice is associated with interiority, truth, and
authenticity. This philosophical position is articulated
forcefully by Hannah Arendt, who asserts that
through the act of speech, individuals “reveal actively
their unique personal identities”. Majumdar, Wanted
Cultured Ladies, 173, 178; Hanna Arendt, The Human
Condition (Chicago: University of Chicago Press,
1998), 179. See also Charles Taylor, Sources of the
Self: The Making of the Modern Identity (Cambridge,
Mass.: Harvard University Press, 1989), cited in
Amanda J. Weidman, Singing the Classical, Voicing
the Modern: The Postcolonial Politics of Music in
South India (Durham, NC: Duke University Press,
2006), 145–146.

14.

Richard Koszarski, Hollywood on the Hudson: Film
and Television in New York from Griffith to Sarnoff
(New Brunswick, NJ: Rutgers University Press,
1998), 202.

15.

Rick Altman attributes the “tyranny” of the continu-
ous, uninterrupted sound track to the reigning model
of perceptual fidelity. Altman, “Sound Space”, 256n4.

16.

The title of this film has elsewhere been cited as Ruth
Etting Sings Favorite Melodies and Favorite Melodies.
I use the title entered into the Library of Congress’s
Catalogue of Copyright Entries.

17.

“Love Me or Leave Me” was recorded by every major
record label company in the United States around
this period. Whoopee! ran through December 1929,
and Samuel Goldwyn’s film adaptation of the musical
comedy was released by United Artists in September
1930.

18.

Bradley, The First Hollywood Sound Shorts, 53, 230.

19.

The remainder of Merman’s performance follows
more or less the lyrics of the song co-written by Sam
Coslow and W. Franke Harling (1930).

20.

The song was co-authored by Vernon Duke and
Edgar Y. (“Yip”) Harburg, regular songwriters for the
productions at Paramount’s Astoria studio.

21.

Max Steiner, “Scoring the Film”, in Nancy Naumberg
(ed.), We Make the Movies (New York: W.W. Norton
and Co., 1937), 218.

22.

Irene Kahn Atkins, Source Music in Motion Pictures
(East Brunswick, NJ: Associated University Press,
1983), 31.

23.

Part of this effect owed to Vallee’s use of a “close-up”
technique of singing, for which he brought his lips
very near to the microphone. Allison McCracken,
“‘God’s Gift to Us Girls’: Crooning, Gender, and the
Re-Creation of American Popular Song, 1928–1933”,
American Music 17.4 (Winter 1999): 373–378.

24.

For example, see “Soulful Mr. Vallee”, New York
Times (4 August 1929): X5; “Rudy Vallee Life Marked
by Successes”, Los Angeles Times (9 February

1930): B15; and Rudy Vallee, “Vagabond Dreams
Come True”, Milwaukee Journal (28 March 1930): 3.

25.

McCracken, “God’s Gift”, 376.

26.

Lydia Goehr, The Quest for Voice: On Music, Politics,
and the Limits of Philosophy (Berkeley: University of
California Press, 1998), 122.

27.

Michel Chion, “Mute Music: Polanski’s The Pianist
and Campion’s The Piano”, in Daniel Goldmark,
Lawrence Kramer, and Richard D. Leppert (eds),
Beyond the Soundtrack (Berkeley: University of Cali-
fornia Press, 2007), 94.

28.

Crafton, The Talkies, 590n11.

29.

Jeffrey J. Noonan, The Guitar in America: Victorian
Era to Jazz Age (Jackson, Miss.: University Press of
Mississippi, 2008), 15–20.

30.

Harold Courlander, Negro Folk Music, U.S.A. (New
York: Columbia University Press, 1992 [1963]), 216.

31.

During the 1929–1930 season, for example, popular
songs appeared in vocal or instrumental form in more
than half of all feature films produced by the major
motion picture companies. Katherine Spring, “Say It
with Songs: Popular Music in Hollywood Cinema
during the Transition to Sound, 1927–1931” (Ph.D.
diss., University of Wisconsin-Madison, 2007), 250.

32.

Film reviewer Mordaunt Hall reported in the New York
Times, “Miss Gaynor’s voice may not be especially
clear, but the sincerity with which she renders at least
two of her songs is most appealing”. Mordaunt Hall,
“The Screen”, New York Times (4 October 1929): 32.
For a more general discussion of silent film stars who
succeeded in early sound films, see Crafton, The
Talkies, 490–495.

33.

Crafton, The Talkies, 510–511.

34.

Spring, “Pop Go the Warner Bros.”, 76–81. In some
newspapers, the voice double was misidentified as,
variously, Grant Withers, Frank Withers, and Frank
Rivers. See, for example, Franc N. Dillon, “‘Voice
Doublers’ Reap Coin Talking for Movie Stars”, New-
ark Star-Eagle (16 February 1929): n.p., RBS; Jerome
A. Lischkoff, “Richard Sings in Jail”, Cincinnati Post
(8 April 1929): n.p., RBS; J.R. Jr., “Barthelmess Fails
to Sing Again Here”, Dallas News (30 March 1929):
n.p., RBS; untitled clipping, Charleston News (4 April
1929): n.p., RBS.

35.

Advertisements for Weary River, Film Daily (8 January
1929): 4; Film Daily (30 January 1929): 10–11;
Crafton, The Talkies, 509.

36.

J.R. Jr., “Song Shenanigan Really an Asset To ‘Weary
River’”, Dallas News (18 February 1929): n.p., RBS
(emphasis added).

37.

“Musicians Committee Declares Singing in ‘Weary
River’ Faked”, Houston Post (3 March 1929): n.p.,
RBS.

38.

Ibid.

FILM HISTORY: Volume 23, Number 3, 2011 – p. 298

298

FILM HISTORY Vol. 23 Issue 3 (2011)

Katherine Spring

39.

J.R. Jr., “Notes on the Passing Show”, Dallas News,
n.d., n.p., RBS.

40.

Untitled clipping, Louisville Courier (30 April 1929):
n.p., RBS.

41.

Jean Betty Huber, “Richard Barthelmess’s Decep-
tion”, Picture Play, clipping dated January 1929 (?),
RBS.

42.

“That ‘Weary River’ Double”, Picture Play (November
1929): n.p., RBS.

43.

“A Fan Loses Faith”, Picture Play (December 1929):
n.p., RBS.

44.

“Theme Song Effective In Next Stanley Film”, Union
City Hudson Dispatch (11 April 1929): n.p., RBS.

45.

Unidentified clipping, RBS.

46.

In her examination of dubbing practices in Hollywood
musicals and the soundtrack albums they spawned,
Marsha Siefert notes, “Dubbing just any image with
a beautiful voice would not do. It is the combination
of the exchange value of star and sound that made
the eventual industrialization and merchandizing of
song dubbing through soundtrack albums a suc-
cess”. Marsha Siefert, “Image/Music/Voice: Song
Dubbing in Hollywood Musical”, Journal of Commu-
nication 45.2 (Spring 1995): 46–47.

47.

The most accessible discography of recordings from
this period can be found at Tyrone Settlemier, The
Online Discographical Project, http://www.78dis-
cography.com/ (accessed 10 January 2011) and its
linked, searchable database at Honking Duck 78s,
http://honkingduck.com/mc/discography/

(ac-

cessed 10 January 2011).

48.

“Only a Girl of Fifteen”, Picture Play (December 1929):
n.p., RBS.

49.

Marquis Busby, “Film Based on Melody Appealing”,
Los Angeles Times (2 March 1929): 7.

50.

“Theme Song is Melody Hit”, Hollywood News (1
March 1929): n.p., RBS.

51.

“Film’s Theme Song Stirs Controversy”, Los Angeles
Examiner (7 April 1929): n.p., RBS.

52.

W. Ward Marsh, “About Girls and Curves, Barthel-
mess’ [sic] Voice and a Letter From John Royal”,
Cleveland Plain Dealer (24 February 1929): n.p., RBS;
“Did Dick Sing?” Greensboro Record (27 April 1929):
n.p., RBS.

53.

Untitled clipping, Louisville Courier (29 April 1929):
n.p., RBS.

54.

Stanley Orne, “Critic Makes Light of Voice Doubling”,
Portland Oregonian (12 May 1929): n.p., RBS.

55.

“Dick’s Voice? – No”, Cleveland News (1 March
1929): n.p., RBS.

56.

F.J. Raleigh, letter, Picture Play (December 1929):
n.p., RBS.

57.

Lastra, Sound Technology, 139.

58.

Ibid., 142–143, 189.

59.

Crafton, The Talkies, 226.

60.

Steve Wurtzler supplies evidence for this position in
a quote from a report by the Progress Committee of
the Society of Motion Picture Engineers, dated Feb-
ruary 1930, in which the authors cite a “public reac-
tion against” voice-doubling. Wurtzler, Electric
Sounds, 272.

61.

Dan Thomas, “Voice ‘Double’ Will Vanish Soon,
Lubitsch Predicts”, Pittsburgh Press (22 September
1929): 71.

62.

Doug Camilli, “Showcase”, Montreal Gazette (25
October 1992): D9, quoted in Christopher Martin,
“Traditional Criticism of Popular Music and the Mak-
ing of a Lip-synching Scandal”, Popular Music and
Society 17.4 (Winter 1993): 76–77. Martin provides
an excellent account of the construction of the Milli
Vanilli “scandal” and the discourse that helped to
define it.

63.

Martin, “Traditional Criticism”, 70–77. On the “hip-
ness” of contemporary music audiences, see
Lawrence Grossberg, We Gotta Get Out of This Place:
Popular Conservatism and Postmodern Culture (New
York: Routledge, 1992), 225–227.

Abstract: “To Sustain Illusion is All That is Necessary”: The Authenticity of
Song Performance in Early American Sound Cinema,
by Katherine Spring

In musical shorts, song performances were rendered through a set of stylistic techniques emphasizing
performative authenticity and synchronicity. By contrast, the integration of song performances into
feature-length films of the same period led to the use of non-synchronous recording practices, such as
voice-doubling, for dramatic actors who did not possess musical talent. Newspaper clippings in the
Richard Barthelmess scrapbooks (Special Collections, Margaret Herrick Library) indicate that film audi-
ences generally accepted voice-doubling as a necessary element of cinematic artifice.

Key words: Authenticity, Richard Barthelmess, Weary River, dubbing, Paramount Pictures, playback, voice
double.

FILM HISTORY: Volume 23, Number 3, 2011 – p. 299

“To Sustain Illusion is All That is Necessary”

FILM HISTORY Vol. 23 Issue 3 (2011)

299