You are on page 1of 14

Trinity University

Digital Commons @ Trinity


Philosophy Faculty Research Philosophy Department

Winter 2012

A Musical Photograph?
Richard Beaudoin

Andrew Kania
Trinity University, akania@trinity.edu

Follow this and additional works at: http://digitalcommons.trinity.edu/phil_faculty


Part of the Philosophy Commons

Repository Citation
Beaudoin, R., & Kania, A. (2012). A musical photograph? The Journal of Aesthetics and Art Criticism, 70, 115-127. doi: 10.1111/
j.1540-6245.2011.01503.x

This Article is brought to you for free and open access by the Philosophy Department at Digital Commons @ Trinity. It has been accepted for inclusion
in Philosophy Faculty Research by an authorized administrator of Digital Commons @ Trinity. For more information, please contact
jcostanz@trinity.edu.
RICHARD BEAUDOIN AND ANDREW KANIA

A Musical Photograph?

This article compares two objects: a photographic i. william henry fox talbots latticed window
negative made by William Henry Fox Talbot in (with the camera obscura), august 1835
1835 and the score of a solo piano work com-
posed by Richard Beaudoin in 2009. Talbots neg- In 1835, Talbot made numerous images of his
ative has come to be known as Latticed Win- home in Wiltshire, England, including a series de-
dow (with the Camera Obscura), August 1835, and picting the central window in its South Gallery,

Beaudoins musical composition is called Etude using a technique that he called photogenic draw-
dun prelude VIILatticed Window. As suggested ing. The process involved building a camera ob-
by their titles, the composition owes a debt to scura out of a large box, fixing a glass at one
the negative and thereby joins a long list of end, and using that glass to project an image onto
musical compositions indebted to particular vi- the opposite end. Sensitive paper mounted at the
sual images.1 However, the relationship is deeper, point of the projected image would, given time,
and by explicating their respective ontologies, we capture the light allowed it. Talbot gives a de-
hope to show that these two objects are strik- tailed account of his materials and methods in his
ingly analogous to each other across their re- privately published 1839 paper, Some Account
spective media, so much so that we suggest the of the Art of Photogenic Drawing, or, The Pro-

score of Beaudoins Etude dun prelude VII cess by Which Natural Objects May Be Made to
Latticed Window should be considered a sort of Delineate Themselves Without Aid of the Artists
musical photographa photograph of a musical Pencil, which was presented at the Royal Soci-
performance.2 ety of Great Britain on January 31, 1839.3 In his
We begin by describing the origins and char- account, Talbot also reflects on the implications
acteristics of both Talbots photographic negative of his discovery for such areas as portraiture, mi-
and Beaudoins score and offering our basic ar- croscopy, and the rendering of sculpture. The essay
gument that the latter is a musical photograph includes remarkably poetic descriptions of the act
(Sections I and II). In Section III, we compare of photography, as when Talbot describes images
Beaudoins score to other things one might con- of the exterior of his house: And this building I
sider contenders for the title of musical photo- believe to be the first that was ever yet known to
graphmusical recordings, sonic spectrographs, have drawn its own picture.4
ordinary musical transcriptions, and typical musi- As seen in Figure 1, Latticed Window is accom-
cal scoresand argue that it has at least as good panied by a text in Talbots own hand, likely pro-
a claim to the title as any of these and a better duced to accompany the image when it was exhib-
claim than most. In the final section, we consider ited by Michael Faraday at the Royal Institution
an argument based on the transparency of pho- on January 25, 1839. Beside the image, which is
tographs, which allows us to recapitulate our main roughly one inch square, Talbot wrote: Latticed
claims. Window (with the Camera Obscura), August


c 2012 The American Society for Aesthetics
116 The Media of Photography

Figure 1. William Henry Fox Talbot, Latticed Window (with the Camera Obscura) August 1835, (1835). Courtesy of the
National Media Museum/SSPL.

1835When first made, the squares of glass about rhythmic interpretation of Chopins score. Senn
200 in number could be counted, with the help of also measured the sound energy (or volume) of
a lens. Two significant facts emerge from Talbots each onset, to the level of the decibel.
concise inscription: (1) Talbot encourages viewers Collaborating with Senn in April 2009, Beau-
to marvel at the precision of the image by inspect- doin devised a method of taking LARAs nu-
ing it with a magnifying glass and (2) by the phrase merical output (the onset times in milliseconds
when first made, Talbot acknowledges that his and the sound energy in decibels) and transcrib-
technique of fixing the light patterns to paper was ing it back into standard musical notation. This
unable, over time, to preserve every detail of the discovery became the framework for a series of
actual image.5 Of course, even when first made, compositions based on the ChopinArgerich ma-
the image did not capture every detail of the scene
terial, called Etudes dun Prelude, twelve of which
photographed, a feature of photography that per- were completed in 20092010. Most of the works
sists, if on a different scale, even in our digital in the series altered the original Chopin mate-
age. rial, applying techniques of elongation or dis-
tortion, or applying techniques borrowed from
photography.7

Etude dun prelude VIILatticed Window is

ii. richard beaudoins etude
dun prelude
viilatticed window unlike any other work in the series, in that its
length in performance is the same as the length of
While only 1 51 in duration, Beaudoins Etude
Argerichs recording (1 51 ). There is a sense in
dun prelude VIILatticed Window is the result which a good performance of Beaudoins Latticed
of months of labor, not only by the composer but Window is a kind of altered re-performance of
also by a small team of acoustic researchers. The Argerichs Chopin interpretation (though the al-
piece originated from data collected using LARA terations are not insignificant). In this way, Beau-
(The Luzern Audio Recording Analyser), devel- doins score reveals itself as a graphic representa-
oped by Dr. Olivier Senn and his colleagues at tion, or visualization, in standard musical notation,
the Hochschule Lucerne in Switzerland. In 2008, not just of another score, but of a specific perfor-
Dr. Senns team spent months measuring the ex- mance of another score. (Figure 2 shows Chopins
act moments of pitch onset in Martha Argerichs original Prelude, while Figure 3 shows Beaudoins
celebrated recording of Chopins Prelude in E mi- score.) In this regard, it is, to our knowledge,
nor, op. 28/4 (Deutsche Grammophon 415 8362, unique within the field of music composition. In
recorded in Munich, October 2225, 1975).6 order to understand the way in which Chopins
These onsets (the exact moments in time when Prelude (and Argerichs performance of it) have
each note comes into being) were charted at the been transformed by Beaudoin into his Latticed
level of the millisecond. The coordinates of each Window and how that transformation is informed
onset were recorded relative to all others, allow- by Talbots image, consider the following four net-
ing for an extremely detailed picture of Argerichs works of similarity and difference.
Beaudoin and Kania A Musical Photograph? 117

Figure 2. Chopin, Prelude in E minor, op. 28/4. Public domain, accessed at http://imslp.org/wiki/Preludes,_Op.28_(Chopin,
_Fr%C3%A9d%C3%A9ric).

First, Chopins left and right hands have been work the left and right hands, melody and accom-
switched in Beaudoins work, meaning that the paniment, figure and ground, as it were, have been
melody is now below the accompaniment. This reversed.
is inspired by the fact that Talbots Latticed Win- Second, Chopins music, which was written in
dow is a photographic negative. Whereas light and the middle register of the piano, has been moved
dark are reversed in Talbots image, in Beaudoins to the instruments extreme high and low registers
118 The Media of Photography


Figure 3. Richard Beaudoin, Score of Etude dun prelude VIILatticed Window. Copyright 2009 by the composer; all rights
reserved.
Beaudoin and Kania A Musical Photograph? 119

Figure 3. Continued.
120 The Media of Photography

by Beaudoin. The shift of register is a response inal Prelude). As with most such pieces, the work
to Talbots negative being in black and white and is physically embodied in a scorea set of instruc-
follows the analogy between color and music artic- tions to the performer about what to do in order
ulated by the French composer Olivier Messiaen: to produce a performance of this work. But be-
When I move the same chord from midrange up cause of its unique origins, we argue that the score
one octave, the same color is reproduced shaded is unique in that it is a musical photograph of Arg-
toward whitewhich is to say, lighter. When I erichs performance of Chopins Prelude.10 Just as
move the same chord from midrange down an oc- Talbot (with the aid of the camera obscura) used
tave, the same chord is reproduced, toned down the light present during a span of time in 1835 to
by blackwhich is to say, darker.8 By recast- fix a two-dimensional image, mechanically coun-
ing Chopins music into the pianos extreme high terfactually dependent on and visually similar to
and low registers, Latticed Window uses only the the scene that can be recognized in it, Beaudoin
lightest and darkest sonorities available on (with the aid of LARA) used the sounds produced
the instrument.9 during Argerichs 1975 performance to fix a two-
Third, the key has been changed from Chopins dimensional image, mechanically counterfactually
original E minor into E-flat minor in Beaudoins dependent on those sounds and in which can be
composition. This change was made simply be- recognized a performance of Chopins prelude.11
cause E-flat minor is a historically (and some We thus focus on two features usually taken to be
would argue, aurally) darker key than E minor, essential to a representational photograph: (1) it
just as Talbots image is darker than the sunlit must be mechanically counterfactually dependent
window it depicts. on its target, and (2) this dependence is reflected
There exists one further cluster of relation- in visual similarity between the image and its
ships between these objects, which is alluded to target.
in the texts that Beaudoin and Talbot supply to Is Beaudoins score really a musical photo-
accompany their respective objects. In the text graph? Of course, to some extent, this is just
accompanying Latticed Window, Talbot admits semantics; the relationship between Beaudoins
that the image fails to capture all of the details score and Argerichs performance is like that be-
of the scene he photographed. Beaudoins score tween a photograph and the object or scene it
is headed by the words Latticed Window (with represents in certain ways and unlike it in other
LARA)August 2009. When first measured the ways. Whether or not we ultimately decide to call
total events, initially about 225 in number, were the score a musical photograph is not as impor-
filtered to include those >0.748 seconds (r.h.) and tant as the ways in which thinking about these
>0.540 seconds (l.h.). This highlights a fourth dif- relationships illuminates the media of both pho-
ference between Argerichs performance of the tography and music and the aesthetic implications
Chopin and Beaudoins work: any event in Arg- of those media. Nonetheless, we do believe that
erichs performance whose duration was below a the similarities are significant enough to defend
certain time threshold (separately chosen for each the claim that Beaudoins score is a musical pho-
hand) was simply left out of the transcription. tograph as more than just a thought-provoking
Just as Talbots negative loses the details of the metaphor.
light that the negative was not sensitive enough
to retain, so Beaudoins score loses the details of
Argerichs interpretation of the Chopin that fell iii. comparisons with other media
below his chosen durational threshold. This pur-
poseful imperfection in the preservation pro- i. Recordings. When one thinks of musical media
cess from Chopin to Beaudoin via Argerich is the analogous to photography, one might first think of
very thing that allows the score to be readable and sound recordings. For instance, one might say that
performable by a musician, as we discuss further if anything is the musical equivalent of a photo-
below. graph of Argerichs performance, it is the original
The primary result of Beaudoins labors is a mu- recording with which Senn and Beaudoin began.
sical work that is intended to be performed, which But there are a number of disanalogies here. Per-
in this respect is no different from a typical piece haps most striking is the fact that a musical record-
of Western classical music (such as Chopins orig- ing, appropriately experienced, has a perceptible
Beaudoin and Kania A Musical Photograph? 121

and meaningful temporal length. That is, it takes less like a musical photograph than a recording in
time to unfold, its musical events occurring in a that it is cross-modalit represents sounds visu-
predetermined order.12 In this respect, a record- ally, while a recording represents them sonically.
ing is more like a moving image, such as a film, But it is more like a musical photograph in that
than a photograph, which represents (usually) a it is a visual representation. It is no surprise that
short temporal slice of some scene in an atemporal musical events can be represented sonically, in a
way (in the sense that there is no predetermined mechanically counterfactually dependent way. It
order in which one must experience the different is perhaps surprising that musical events can be
elements of a photograph, nor any specific length represented visually in a mechanically counterfac-
of time an appropriate experience of the photo- tually dependent way and that they can be repre-
graph should take). On the other hand, while a sented using artistic techniques, such as filtering,
score is a visual representation, it is at least lin- that have analogues in photography.
ear, in that it is appropriately read in a particular
order. And it might be argued that it is temporal ii. Spectrographs. However, there are other ways
insofar as it is appropriately viewed or read at of visually representing musical performances in
the same tempo as a performance of it. (Of course two dimensions than with a musical score. One
one could look at different parts of the score in is by means of a spectrograph. Consider the data
any order one liked, but one could similarly listen delivered by LARA in analyzing Argerichs per-
to bits of a recording in different orders. The point formance. Such data can be (and initially are)
here is about the appropriate mode of experience represented numerically or linguistically; that is,
of these things.) it is an array of numbers representing the time,
A second disanalogy between the musical frequency, energy, and so on, of every sound
recording and a photograph (closely related to the event included on the recording. But these data
first) is that the recording is not a visual but a sonic can be displayed visually. Figure 4 shows a spec-
representation. That is, the sounds of a recording trograph taken by LARA of the first 21 sec-
(played in an appropriate way), including many onds of the performance, the x-axis represent-
musically important audible features, such as tim- ing time (in seconds) and the y-axis showing
bre, pitch, and rhythm, are mechanically counter- frequency (in hertz). Could we thus consider a
factually dependent on the sounds of the recorded spectrograph a musical photograph of Argerichs
performance. Moreover, the sound of the record- performance as captured on the recording?
ing being played is aurally, not visually, similar to Such a spectrograph may be considered a pho-
the sound of the original performance: we can hear tograph of a musical performance in some sense:
Argerichs performance in the played recording; it is a visual representation of a musical perfor-
we do not see it there. mance that is mechanically counterfactually de-
What is the relevance of these disanalogies, pendent on it. Even if such a spectrograph may
though? Recordings are to be heard rather than to be considered a sonic photograph, however, it
be seen and are thus different from photographs. should not be considered a musical one. A spec-
But this is because recordings aim to represent trograph represents sounds, or vibrations in the
musical events, which are audible rather than vis- air, but not music, which requires perception of
ible. Thus, it might be argued that at a deeper, a uniquely human sort. The distinction between
structural level, a recording is more like a mu- sounds and music is difficult to limn precisely,
sical photograph than Beaudoins score because but there is general agreement that human beings
sound recording bears the same relations to mu- (unlike dogs, say) hear music in certain sounds,
sical events that photography bears to visible just as we see three-dimensional objects in cer-
objects. We have no objection to the idea that tain two-dimensional arrangements of pigment.13
musical recordings could be considered musical Illustrative differences between sounds and music
photographs for the above reasons. What is in- (at least Western tonal music) include the follow-
teresting about Beaudoins score is that it shows ing: (1) When we hear sounds as music, we hear the
that there are other possible kinds of represen- sonic spectrum as divided into octavesnotes that
tations of musical events that could equally well repeat at different heights. This phenomenon is
be called musical photographskinds that have not an intrinsic feature of the sonic spectrum. We
heretofore gone undiscussed. Beaudoins score is hear octaves in sounds with frequencies separated
122 The Media of Photography

Figure 4. Spectrograph of Chopins op. 28/4 in Martha Argerichs 1975 interpretation (seconds 021), created with the
Lucerne Audio Recording Analyzer (LARA).

by a factor of two, but pairs of sounds whose fre- be seen or recognized in the spectrograph; only the
quencies are related by other factors (three, five, sonic features can. In some sense, of course, the
and so on) are just as objectively present in the spectrograph contains musical information. Thus,
sound spectrum. The repetition of notes at the oc- if we understand visual representation, in a thin
tave is a response-dependent feature of sounds sense, to mean something like carrying informa-
and seems to be culturally universal.14 (2) When tion in a visual form, then the spectrograph could
we hear sounds as music, our perception is cate- be said to represent Argerichs performance. But
gorical, as it is when we hear sounds as language. when we talk of photographs and other pictures
Just as we hear any of a wide range of sounds being representational, we typically mean some-
as a certain phoneme (for example, /pa/) when we thing more than this: we mean that features of
hear it as language, but past a certain threshold we what is represented can be seen in the visual
hear intrinsically very similar sounds as a different array. (Compare a JPEG file on your hard drive.
phoneme (for example, /ba/), we hear all sounds Theres visual information there, but it cannot be
within a certain range as a certain note (for ex- seen until interpreted by a piece of software that
ample, C#) when we hear it as music, but past a projects or prints an image.)
certain threshold we hear intrinsically very similar There is an interesting distinction to be made
sounds as a different note (for example, D).15 This here between the score and an ordinary photo-
general phenomenon also seems to be culturally graph, however. Musical notation is largely con-
universal, though of course not all cultures divide ventional in a way that ordinary depictions (such
the octave into the same pitch classes. (We also as typical photographs) are not. (For example,
do not mean to imply that all musical sounds are untutored children can understand the content
actually, or perceived as, pitched.) Though these of many photographs, but not of any scores.)
two features relate to pitch, similar points could This might lead us to say that the disanalogy be-
be made about rhythm and other musical features tween the spectrograph and a photograph is lo-
of sound.16 cated more in our perceptual systems than in the
A spectrograph of Argerichs performance, representations.17 Whatever its source, however,
then, is not a musical photograph of it because the disanalogy is there, supporting our argument
the musical features of the performance cannot in this section that Beaudoins score is a better
Beaudoin and Kania A Musical Photograph? 123

candidate than a spectrograph for being a mu- terfactual dependence between Argerichs per-
sical photograph of Argerichs performance. On formance and Beaudoins score is continuously
the other hand, to the extent that musical nota- mechanical.
tion is linguistic, it detracts from our claim that
the score is a musical photograph. There is some- iii. Ordinary transcriptions. Beaudoins Latticed
thing to the notion that we see the music in the Window, by contrast with a spectrograph, employs
notation, as we see an object in an image. Musical musical notation and thus visually represents the
notation is not wholly linguistic; it is partly iconic. musical features of the performance, albeit in a
Nonetheless, the analogy is imperfect. Part of the distorted way. It is thereby, in part, a musical tran-
problem is the cross-modal nature of the represen- scription of Argerichs performance. Considering
tation of music in notation, a point we return to the reasons for the distortions can help us de-
below. velop the analogy with photography in more de-
Is Beaudoins score mechanically counterfactu- tail. Given the technology available to Beaudoin,
ally dependent on Argerichs performance, as the one might ask why he did not produce a more ac-
spectrograph is? As discussed above, Beaudoin curate, less distorted transcription of Argerichs
made many decisions in the production of his score performance. Part of the answer, obviously, is that
that had implications for how Argerichs record- he was engaged in an artistic, not a reportorial
ing was represented in it, such as, for example, the project. But another part of the answer is that
decision to represent the sounds Argerich pro- there is no way to represent the musical details of
duced with her right hand, higher in the musical such a performance in a way that would be useful
register, with notes lower in the musical register, for Beaudoins purposes. Western musical nota-
which the performer of Beaudoins work would tion has been developed to represent something
play with his left hand. It would be a mistake to like musical works for performance, works that
think that this detracts from the mechanical coun- require performative interpretation.18 It has also
terfactual dependence of the score on its target, been put to other uses, such as transcribing impro-
though, just as it would be a mistake to think that vised performances, but because of the purposes
a photographers choice of lens or printing tech- for which it was designed, no doubt in combination
nique detracts from the mechanical counterfactual with basic perceptual and cognitive limitations,
dependence of her photograph on its target. Such such transcriptions tend to represent something
decisions result in less visual similarity between like the work of which the improvisation would
the resulting object and its representational target, have been a performance, were it a performance
but visual similarity is distinct from counterfactual of a work.19
dependence. Given the remarkable precision of LARAs
However, there is an intentional connection in rhythmic measurements (pinpointing the onset of
the chain of counterfactual dependence linking each note at the level of the millisecond), tran-
Beaudoins score with Argerichs performance: scribing Argerichs performance into standard no-
Beaudoin transferred by hand the information tation involves a complicated balance between the
from LARA into his score. He did so mechan- richness of the data and the limits of human per-
ically in a loose sense, that is, in strict accor- ception. It is possible, say, to create a mapping
dance with predefined, unambiguous rules; but such that each millisecond in Argerichs record-
this part of the process was mediated by beliefs, ing is represented by one 512th note in the tran-
and thus was not mechanical in the sense required scription. (A 512th note is one with seven flags at-
for a strict analogy with photography. We will ig- tached to its stem.) Such a transcription could be
nore this break in the chain, however. Because helpful for those engaged in performance studies,
both the LARA data and the resulting score are since it would represent musical details in a form
made up of discrete, discontinuous bits of infor- any trained musician could understand, allowing
mation, it is easy to imagine a mechanical pro- for the comparison, for instance, of the rubato
cess that does what Beaudoin in fact did inten- applied in different measures. However, although
tionally. For the purposes of this article, then, one could study such a score, one could not take
we consider a close counterpart of Beaudoins it in in the ordinary way, due to its overwhelming
actual scoreone in which the chain of coun- detail.
124 The Media of Photography

For the same reason, such a transcription could sider the history of artistic photography a history
not be performed. At the tempo of Argerichs per- of the invention and use of such limitations.
formance of the Chopin, this transcription would
confound almost any human pianist; even if a pi- iv. Musical scores. Since we have argued that
anist could handle such a notation cognitively, Beaudoins score for Latticed Window could be
it would be nearly impossible to obtain the mo- considered a musical photograph, it is worth ask-
tor control to execute it. This presents a problem ing whether traditional musical scores could be
for a composer wanting to use such data to pro- too. For instance, why not say that Chopins score
duce a new piece that is realistically performable. for his Prelude in E minor, op. 28/4, is a musi-
Beaudoins solution, in Latticed Window, is to use cal photograph of Argerichs performance? There
a remarkably fast (but humanly possible) tempo are two main reasons why we should not say this.
alongside an unusually large time signature. This, The first, of course, is that Chopins score is not
together with the filtering out of the smallest counterfactually dependent on Argerichs perfor-
events of the recording, represents a compromise mance, since the score predates the performance
between LARAs spectrographic richness and our by over a century. Just as any object within the
cognitive and motor limitations. frame of a photograph will be captured by the
By analogy, suppose it were possible to take an camera whether the photographer wills it or no,
incredibly detailed photograph of the Mona Lisa, every event in Argerichs performance is captured
such that its molecular structure were somehow by LARAs processes, whether or not Beaudoin
captured visually. In order for us to see these de- (or anyone else) wills it. Unlike LARAs output,
tails, perhaps the photograph would have to be the Chopins score is impervious to Argerichs perfor-
size of a planet. Such a photograph would be use- mance. She, and any other number of performers,
less to us as a photograph of the painting, because could have interpolated as many extra notes as
we could not see the painting in it, for reasons they wished; this would not affect the number of
analogous to why we cannot see the molecular notes in Chopins Prelude.
structure of the painting when standing in the Lou- This leads us to the (related) second main dif-
vre. A representation in musical notation of Arg- ference between Chopins and Beaudoins scores.
erichs performance, of the sort described above, Chopins score is not a transcription of a perfor-
even if possible, would be useless to us as a musical mance; it prescribes how certain performances
photograph of that performance in a similar way. those of this preludeshould be, rather than
There are two basic ways to produce a more use- describing any particular performance. In this re-
ful (that is, comprehensible) musical photograph spect, it is more like the circuit diagram an en-
of a performance using microtiming data, both of gineer produces for a new electrical component
which have been employed by Beaudoin. One is to than a photograph of such a component. It tells
stretch the data out across time, to produce a kind others what they ought to do in a generic fashion,
of slow-motion transcription of the performance. rather than showing them what some particular
Beaudoin did this in many of the pieces in the thing is like. Of course, Beaudoins score is pre-

Etude dun prelude series, including the first piece scriptive, too, but before it was prescriptive it was
in the series, Chopin desseche, which expands the descriptive.20 Beaudoin took a transcription (or,
time axis from 1 51 to 7 25 . The other is to put more accurately, helped produce a transcription
the data through a coarse filter, retaining only du- with certain features) of a certain performance
rations that are discriminable by the human eye (Argerichs) of a given work (Chopins) and
and ear. This is one of the transformational tech- transformed it into the prescription for produc-
niques Beaudoin uses in Latticed Window, as de- ing performances (anyones) of a different work
scribed in Section II. We have already mentioned (Beaudoins). Such a process is as unusual in the
how this technique reflects the limitations of Tal- visual arts as in music, but we can easily imag-
bots photographic technology, but, of course, ine analogues (and works of this sort probably
photographers from the very birth of the technol- exist). Suppose an artist takes a photograph of a
ogy have used such limitations, whether dictated lithographic print that was designed by a second
by the contemporary state of the technology or artist and executed by yet a third. Then suppose
freely chosen, for artistic ends. One might con- the first artist publishes the photograph, clearly
Beaudoin and Kania A Musical Photograph? 125

indicating, LeWitt-like, that the photograph itself others have done, a machine that mechanically
is not her artwork, but that other artists should produces a linguistic description of the visual ap-
execute instances of her work by painting images pearance of a scene.23 No one argues that such de-
of the print captured in the photograph. The fact scriptions enable us to see the scene represented,
that the artist uses the photograph to prescribe precisely because the scene is not depicted. Musi-
what others should do in producing instances of cal notation, as a quasi-linguistic form of represen-
her artwork does nothing to take away from the tation, is similarly not a form of depiction, which is
fact that it is a photograph. Beaudoins score pre- one reason we cannot see a musical performance
scribes what a performer should play in order to through Beaudoins score. Another, perhaps more
perform Latticed Window, just as our imagined obvious, reason is that a musical performance is
artist prescribes what others should do in order not a visual entity. It is something appropriately
to produce instances of her work. Thus, we argue, heard, rather than seen.24 Thus, no visual repre-
the fact that Beaudoins score is prescriptive does sentation of a musical performance can be a de-
nothing to detract from the reasons for thinking piction of a musical performance, nor a fortiori a
that it is a musical photograph. musical photograph.
We have said three things relevant to this objec-
tion already. First, we do not claim that the score of
iv. transparency
Etude dun prelude VIILatticed Window is sim-
ply a photograph, in exactly the same sense as
Throughout this article, we have made reference a photograph by Cartier-Bresson or this months
to two features of representational photographs cover of Vanity Fair. Our argument is not that
widely discussed in the literature: mechanical Beaudoins Latticed Window has relinquished its
counterfactual dependence and visual similarity. status as a piece of music, nor that its score should
These features have been most widely discussed be hung alongside Mapplethorpe prints. Rather,
in the context of the transparency of photographs, having outlined their respective ontologies, we
that is, the question of whether, in looking at argue that the similarities between Beaudoins
a photograph of some object you thereby see score and Talbots Latticed Windows make it more
(indirectly) the object itself, through the photo- than metaphorical to call this score a musical
graph.21 Now consider the following argument. photograph.
Typical photographs are transparent. Thus, if Second, given that this is an extended use of
Beaudoins score for Latticed Window is a (mu- the term photograph, it does not follow from the
sical) photograph, by looking at it one thereby fact that Beaudoins score lacks some feature or-
sees Argerichs performance (or the recording of dinary photographs possess that the score is not a
that performance) through the score. But we do photograph in an extended sense. Musical record-
not see Argerichs performance (or the record- ings are sometimes referred to as musical pho-
ing of it) through the score. Therefore, it is not a tographs, though there are obvious, clear differ-
photograph. ences between the media of sound recording and
One might reject this argument by rejecting the photography. What we hope to have shown here
supposition that photographs are transparent, but is that there is another, at least equally good can-
this does not seem very promising. The point of didate for the title musical photograph that has
the argument is to show an important dissimilarity thus far been overlooked.
between the score and photographs; that dissimi- Third, though it is true that Beaudoins score
larity remains even if photographs are not trans- is not, and could not be, visually similar to Arg-
parent. What is the dissimilarity? It is not a lack of erichs performance in the way an image is visually
counterfactual dependence, the most widely dis- similar to a visual object, it is still true that one
cussed feature of transparency. Beaudoins score can see that performance in the score, just as we
is counterfactually dependent on the recording of might say we hear an emotion in a musical passage,
Argerichs performance and thereby on the per- even though emotions have no sonic features.
formance itself, since the recording is counterfac- The visual similarity of photographs (and other
tually dependent on the performance. What the depictions) to the things they represent is surely
score lacks that a photograph possesses is visual important in large part because it grounds our see-
similarity with what it represents.22 Consider, as ing those things in the images. Thus, the fact that
126 The Media of Photography

we can see Argerichs performance in Beaudoins 2. This claim has, in fact, already been made, but
score lends plausibility to the claim that it is rele- not extensively argued for, in Richard Beaudoin, Stephen
Davies, and Jonathan McKeown-Green, Micro-measured
vantly like a photograph.
Interpretations as Material for Composition (unpublished
The score of Richard Beaudoins Latticed Win- manuscript).
dow is mechanically counterfactually dependent 3. Reprinted in : William Henry Fox Talbot, Some
on a particular musical performance of another Account of the Art of Photogenic Drawing, or, The
musical work, a performance one can see in the Process by Which Natural Objects May Be Made
to Delineate Themselves Without Aid of the Artists
score.25 These features lead us to describe Beau- Pencil, Photography: Essays and Images: Illustrated
doins score as a musical photograph of that per- Readings in the History of Photography, ed. Beaumont
formancean extended use of the term photo- Newhall (New York: The Museum of Modern Art, 1980),
graph, to be sure, but not a capricious one. On the pp. 2331.
4. Talbot, Some Account, p. 28.
other hand, the score is not visually similar to the
5. Geoffrey Batchen points out that there are at least
performance in the way ordinary photographs are twice as many squares of glass in the window as suggested by
visually similar to the objects one can see in them. Talbot, in A Latticed Window, in Singular Images: Essays
This makes the analogy imperfect. But the best al- on Remarkable Photographs, ed. Sophie Howarth (London:
ternative candidate for a musical photograph of a Tate Modern, 2005), pp. 1521.
6. The methodology for these measurements can be
particular performance, a sound recording, is also found in Olivier Senn, Lorenz Kilchenmann, and Marc-
imperfectly analogous to ordinary photographs. Antoine Camp, Expressive Timing: Martha Argerich plays
Though recordings are mechanically counterfac- Chopins Prelude op. 28/4 in E minor, in Proceedings of the
tually dependent on their sources, they are not International Symposium on Performance Science 2009, ed.
A. Williamon, S. Pretty, and R. Buck (Utrecht: European
visual entities; one does not see anything in them,
Association of Conservatories, 2009), pp. 107112.
nor are they visually similar to their sources. Of
7. For instance, Etude dun Prelude VIThe Real Thing
course, they are sonically similar to their sources, (string quartet, ca. 5 30 , August 25, 2009) was composed
and we can thus hear those sources in recordings.26 in response to a 2000 painting of the same name by the
To claim that either a recording or a score like UK artist Glenn Brown, who uses photography and Photo-
shop to produce, in oil on board, versions of iconic paint-
Beaudoins is a musical photograph is to use the
ings; while Etude dun Prelude VIIIKertesz Distortion
term photograph in an extended sense. What we (string quartet, ca. 7 30 , August 25, 2009) was inspired by
hope to have shown here is that Beaudoins com- Andre Kerteszs
1933 photograph, Distortion No. 172.
positional process generates a new kind of musical For a summary of the compositional process and descrip-
tions of several of the works, see Davies, Beaudoin, and
object, one that warrants this extension of the term
McKeown-Green, Micro-measured Interpretations, and
into the domain of notated music.27 Richard Beaudoin, The Principles of Microtiming and Mu-
sical Photorealism (unpublished manuscript), available at
RICHARD BEAUDOIN http://nrs.harvard.edu/urn-3:HUL.InstRepos:3415685.
Department of Music 8. Olivier Messiaen, Music and Color: Conversations
with Claude Samuel, trans. E. Thomas Glasow (Portland,
Harvard University
OR: Amadeus, 1994), p. 64.
Cambridge, Massachusetts 02138 9. For a philosophical introduction to musico-visual
synesthesia, see Kathleen Higgins, Visual Music and Synes-
internet: beaudoin@fas.harvard.edu
thesia, in The Routledge Companion to Philosophy and
Music, ed. Theodore Gracyk and Andrew Kania (New York:
ANDREW KANIA Routledge, 2011), pp. 480491.
Department of Philosophy 10. Beaudoins score is unique in this respect as far as we
know. But the uniqueness claim is a contingent one; there
Trinity University
is no reason why Beaudoin or future composers could not
San Antonio, Texas 78212 produce further such musical photographs.
11. It is possible to make a musical score in standard no-
internet: akania@trinity.edu
tation that preserves every rhythmic and dynamic aspect of a
performers interpretation (a sort of Latticed Window with-
out any alterations to key, handedness, or detail); such ex-
1. Other such works include Modest Mussorgskys Pic- periments were in fact part of the process of making Latticed
tures at an Exhibition (1874), Franz Liszts Von der Wiege Window. While such a work might be a simpler example of a
bis zum Grabe (1881), Sergei Rachmaninovs The Isle of musical photograph of the sort we are discussing, such true
the Dead (1907), Paul Dessaus Guernica (1937), Gun- imitations of performances were not Beaudoins aim; more-
ther Schullers Seven Studies on Themes of Paul Klee over, the alterations made in Latticed Window strengthen
(1959), and Henri Dutilleuxs Timbres, espaces, mouvements the analogy with Talbots image. We return to these points
(1978). below.
Beaudoin and Kania A Musical Photograph? 127

12. For more on the temporality of artworks, see Jerrold defending, because (i) it does not transcribe a single perfor-
Levinson and Philip Alperson, What Is a Temporal Art? mance (let alone a performance of a preexisting work) and
Midwest Studies in Philosophy 16 (1991): 439450. (ii) his description of these musical items is intentionally
13. Andrew Kania, Definition, in The Routledge Com- mediated in a way that disqualifies it from being mechani-
panion to Philosophy and Music, pp. 313. cally counterfactually dependent. Thanks to the editors for
14. Thomas Stainsby and Ian Cross, The Perception raising this issue.
of Pitch, in The Oxford Handbook of Music Psychology, 21. For an introduction to this literature, see Nigel
ed. Susan Hallam, Ian Cross, and Michael Thaut (Oxford Warburton, Photography, in The Oxford Handbook of
University Press, 2009), pp. 4758. Aesthetics, ed. Jerrold Levinson (Oxford University Press,
15. Christopher Mole, The Motor Theory of Speech 2003), pp. 614626.
Perception, in Sounds and Perception: New Philosophi- 22. It lacks aural similarity, too, since the score makes no
cal Essays, ed. Matthew Nudds and Casey OCallaghan sound.
(Oxford University Press, 2009), pp. 211233. Of course (in 23. Kendall L. Walton, Transparent Pictures, reprinted
the musical case, at least), we hear the different sounds as in Marvelous Images: On Values and the Arts (Oxford
different (for example, we hear the C#s as progressively University Press, 2008), pp. 105109.
sharper), but we still hear them as falling into the pitch-class 24. At least, the aspects able to be captured in musi-
C#. cal notation are appropriately heard rather than seen. Jer-
16. For an introduction to such matters, see Roger rold Levinson, for instance, argues that we must understand
Scruton, Rhythm, Melody, and Harmony, in The the gestures that occur in musical performances in order
Routledge Companion to Philosophy and Music, properly to appreciate those performances (for example,
pp. 2437. the caressing nature of the cellists bowing) (Authentic
17. Thanks to Joseph Moore for bringing this point to Performance and Performance Means, in Music, Art, and
our attention. Metaphysics [Cornell University Press, 1990], pp. 393408).
18. Stephen Davies, Notations, in The Routledge Com- And seeing musicians perform is the most obvious way of
panion to Philosophy and Music, pp. 7079. gaining such understanding. However, even if this is true, it
19. For some examples of transcriptions that capture does not affect the argument, since musical scores do not
more detail than most, yet still clearly illustrate the gap look like, say, performing cellists.
between musical event and transcription, see Paul F. 25. Or, strictly speaking, the score of the close counter-
Berliner, Thinking in Jazz: The Infinite Art of Improvisation part we have been considering has these features.
(University of Chicago Press, 1994), pp. 505757. 26. Perhaps we do not just hear these sources in record-
20. One might think that Chopins score is still descrip- ings, but literally hear the sources; that is, perhaps musical
tive, not of a performance of his prelude, but of the various recordings are transparent. This would be a further argu-
playings he engaged in while composing the work (whether ment for considering recordings to be musical photographs.
actual, or in his head), or of some preexisting, eternal On the transparency of musical recordings, see Andrew
sound structure. (Similarly, one might think a diagram of a Kania, Musical Recordings, Philosophy Compass 4
novel circuit is descriptive of a preexisting, eternal circuit (2009): 2238.
structure.) We think this is a rather odd way of conceiving 27. For helpful discussion, we thank Stephen Davies,
of composition, but even if it is correct, Chopins score is Joseph Moore, Olivier Senn, Michael Schreyach, and the
still not a musical photograph in the sense we have been editors of this issue.

You might also like