About CA
Contact CA

Search Journal

Editorial Board

Permission to Reprint


Site Map


In Praise of Ambiguity: Musical Subtlety and Merleau-Ponty

  Tiger C. Roholt

When a jazz, rock, or hip-hop drummer strikes certain notes in each measure slightly late, instead of hearing the degree to which those notes are late, we typically hear the effects of those variations; namely, a groove, the "feel" of a rhythm. Slight variations of pitch function similarly.  In this essay, I argue that certain analytic theorists go astray due to their preoccupation with the variations themselves.  By invoking Maurice Merleau-Ponty's insights into subtle visual perceptions, and his notion of perceptual indeterminacy, I avoid an account of musical subtlety suggested by Daniel Dennett that is too coarse-grained, as well as the bleak conclusion that certain musical subtleties are ineffable, Diana Raffman's view.  I conclude that elements of music that are perceived ambiguously can perform a positive function in such aesthetic experiences:  they can mediate or foster emergent qualities; moreover, they must be perceived in this way to do so.

Key words
aesthetic experience, Paul Cézanne, Daniel Dennett, expressive variation, groove, indeterminacy, ineffability, Maurice Merleau-Ponty, musical nuance, musical subtlety, phenomenology, Diana Raffman, rhythm

1. Introduction

The active nature of aesthetic experience is emphasized by R. G. Collingwood, Roman Ingarden, John Dewey, to name a few.  One way in which aesthetic experiences are active is that they are exploratory.  For instance, in order to apprehend one feature of a painting or musical performance, you may need to direct your attention to other features.  A rock music critic may write, "The rhythm of this track has an intriguing, frantic quality; in order to hear this, listen to the way the bass guitar repeatedly races ahead of the drums." This may be followed by advice about what not to attend to: "This frantic quality is difficult to hear; if you focus too much on the keyboard and voices, you may fail to hear it."  Aesthetic experiences are often active at least in the sense that we can, and often must, look and listen to different features of an artwork. 

Here are three of the questions that drive this essay:  What are the effects of perceptual attention upon what we perceive in music?  Does a given feature affect other features differently when we attend—or do not attend—to that feature?  In cases in which certain features are responsible for emergent qualities, does perceptual attention upon those responsible features affect this emergence?  I explore questions such as these through examples of musical nuances (a.k.a. "expressive variations"), such as a slightly raised F-sharp or a slightly early eighth note, and emergent qualities, such as a harmonic brightness or a groove.  Although I focus upon music, I believe that some of what I say can be applied to similar perceptual circumstances in visual art; indeed, I will draw upon examinations of visual perception and visual art.           

It is a common methodological starting point in continental phenomenology to focus upon the phenomena under consideration as they show up in ordinary experience.  The alternative, according to phenomenologists, is to allow one method of investigation or another to illegitimately put its imprint upon the subject matter in the early stages of an investigation.  In ordinary experiences of musical subtleties, we hear the effects of musicians' minute manipulations of pitch and timing rather than the manipulations themselves (musicians perform such variations, in fact, for the purpose of our hearing such effects).  For example, a singer may sing a certain note slightly high; we hear the effect of that variation, a "brightened" harmony (say), rather than hearing the degree to which the note is raised.  A jazz, rock, or hip-hop drummer may strike certain notes in each measure slightly late; we hear the effect of these manipulations, a groove (the "feel" of a rhythm), rather than hearing the precise degree to which the notes are late.  Of course, with practice, one can hear (detect) these slight variations themselves, but this is not a typical way to listen. 

In investigating musical subtleties, many psychologists, music theorists, and philosophers focus on the slight variations themselves, often upon our abilities to detect or to discriminate between them.  In contrast to this, in order to offer an account of the musical subtleties we typically hear, the effects of slight variations, I draw upon Maurice Merleau-Ponty's insights into subtle visual perceptions, such as his use of the figure-ground structure of perception and his notion of perceptual indeterminacy.  I aim to avoid an account of musical subtleties that is too coarse-grained (which would emerge from an approach suggested by Daniel Dennett), as well as to avoid the bleak conclusion that certain musical subtleties are ineffable (Diana Raffman's conclusion); Dennett's and Raffman's views ultimately result from focusing upon the variations themselves rather than their effects.  Although these terms and relations must be clarified below, I will claim that elements of music indeterminately perceived perform a positive function insofar as they mediate emergent qualities; they only do this as indeterminate, i.e., when they are preserved as ambiguous in perception.

2. Daniel Dennett's E-string

Daniel Dennett has suggested that some perceptions which, at first, seem ineffable turn out not to be once they are analyzed effectively.  In his "Quining Qualia,"[1] he considers the sound of a guitar's low, open E-string, through a three-step thought experiment.  He begins by asking us to imagine simply plucking the string.  The sound seems rich, ineffable and unanalyzable.  His method, reasonably enough, is to attempt to break-up the sound into parts. Thus, next, we are asked to play the string's harmonic (by placing a finger lightly on the twelfth fret while plucking). Upon hearing this harmonic, "Suddenly a new sound is heard:  'purer' somehow and of course an octave higher."[2] According to Dennett, we have now isolated one of the original sound's seemingly ineffable characteristics, the harmonic overtones.  Finally, Dennett believes that when we listen to the open E again, after hearing the harmonic, we will be able to clearly perceive the overtones of the sound, which will render the composite sound that much less ineffable:  "On a third open plucking one can hear, with surprising distinctness, the harmonic overtone that was isolated in the second plucking.  The homogeneity and ineffability of the first experience is gone, replaced by a duality as 'directly apprehensible' and clearly describable as that of any chord."[3]            

The analysis Dennett leads us through is misleading:  rather than simply clarifying the original perception, his instructions lead us to a different perception.  Dennett himself describes the third perception as being different (in the third perception the overtones are more distinct).  He does not seem to realize that in the different steps of his experiment we are listening to the E-string in different ways—and when we do, a change in the structure of the perception occurs. Invoking the early Gestalt psychologists, in his Phenomenology of Perception,[4] Merleau-Ponty maintains that in order to correctly describe perceptions we must describe them in terms of the figure-ground structure (the figure consists of the area to which one attends; the ground consists of the other portions of the visual field):  "a figure against a background is the most basic sensible given we can have. . . .  The perceptual 'something' is always in the middle of some other thing" (PoP, p. 4).  Merleau-Ponty explores the relationship between perceptual attention and perceptual structure by considering a Necker Cube (figure 1). Regarding the labeled cube, he writes, "When I focus upon the face ABCD of the cube, this does not mean simply that I make it enter into a state of being clearly seen, but also that I make it count as a figure, and as closer to me than the other face; in short, I organize the cube" (PoP, p. 275).

Figure 1: Necker Cube

Focusing upon different parts of the cube changes the structure of the perception.  Failing to describe that structure accurately leads to a misdescription of the perception.  These structural changes also affect other aspects of a perception: in this case, what we take to be the figure determines whether we see the cube as from below or above.  We wouldn't say that these different perceptions of the cube are the same.           

Returning to Dennett, he believes that the third perception of the E-string is similar enough to the first perception that the third is simply a clarification the first; this is his mistake. He doesn't seem to realize that focusing upon different features of the stimulus changes the structure of what we perceive.  In the first hearing of the E-string, the overtones were not the focus of attention (this is the typical perception, in the sense mentioned in my introduction); however, hot on the heels of the isolated perception of the harmonic (the second hearing), in the third hearing, the overtones are the focus, they are the figure.  Thus, Dennett's analysis does not merely clarify the first perception; rather, the analysis results in a different perception; it generates a different perceptual structure in which the overtones become the figure.  He does not render the first perception effable.  The third perception can be described as clearer than the first but that is not all that distinguishes it from the first. 

The important point for our purposes is that the perceptions are different, and this difference can be fruitfully characterized in terms of different ways of perceiving the E-string sound, a different perceptual structure.  What Dennett takes himself to be doing is clarifying an aspect of the experience (the overtones), which have been there all along.  And he assumes that there is no difference between an experience in which the overtones remain unnoticed and the experience in which they are noticed.  We will see that when more specifiable effects of such subtleties are involved, in actual music, this seemingly minor difference between experiences becomes crucial.

3. Subtleties of pitch and duration

An instrumentalist or vocalist may perform certain notes that are slightly high or slightly low; two such notes may be accurately categorized as A-sharps while one is slightly higher than the other (but not high enough to be categorized in terms of the next highest pitch concept). Regarding duration, a drummer, vocalist, or pianist, may perform slightly early or late notes.  Like the pitch examples, this earliness or lateness is not characterizable by means of music-theoretic concepts, such as eighth note, sixteenth note, or dotted sixteenth note.  Two notes exhibiting a timing subtlety may be accurately categorized as eighth notes while one is performed slightly later.  Diana Raffman calls these subtleties "musical nuances";[5] psychologists and music theorists prefer "expressive variations."           

Importantly, philosophers, psychologists, and music theorists who examine these musical subtleties typically do so by characterizing them in the figure role (this is not to take the typical musical experience to be the target of investigation). We can see that this is the case by recalling that focusing upon a pitch or duration places it in the figure role; it is common in the relevant articles on music perception to find subjects being asked to attend to pitches or durations in order to report on which variations they are able to detect, or to discriminate between.  For example, in Eric F. Clarke's "The Perception of Expressive Timing in Music," he writes, "The experiments reported in this paper are an attempt to investigate the ability of listeners to detect small-scale timing changes, similar to those in expressive performance, in various kinds of musical sequence."[6]  Raffman adopts this orientation by following such research:

In hearing these nuances, we are hearing differences within—that is, more fine-grained than—the C-pitch [chromatic pitch] and C-interval (chromatic interval) categories.  Each C-pitch category subsumes many discriminably different pitches, just as each "determinable" color category subsumes many different "determinate" shades; there are many A-naturals and many B-flats, just as there are many reds and many blues. Under laboratory conditions of minimal uncertainty, the human ear can discriminate anywhere from 20 to 300 pitches to the semitone, depending upon the frequency range and testing procedure employed.[7]

Raffman uses pitch terms with subscripts to denote specific pitches; this also indicates that she is characterizing variations as occupying the figure role:  a slightly high F-sharp is an "F-sharp(2)," "F-sharp(4)," and so on.  She claims that such terms accurately characterize these subtleties; the terms "serve perfectly well for enunciating the [representational] content in question."[8]  Construing nuances in the figure role sets the stage for her resulting observation that nuances are ineffable.  Once these subtleties are conceived as in the figure role, this leads to the observation that our capacities of discrimination outstrip our capacities of conceptualization.  Just as we can discriminate or detect many more color shades than we can conceptualize, so too, we can discriminate many more pitches than we can conceptualize.  According to Raffman, fine-grained pitches are ineffable insofar as we cannot conceptualize them.[9]

4. Pitch and duration subtleties in the figure-ground structure

So far, we have seen that Dennett believes that our subtle perceptions can be clarified conceptually, so they are perfectly effable; we've considered the mistakes in his approach.[10]  Raffman believes that discrimination outstrips conceptualization, and this leads to ineffability.  And as I have indicated, like Dennett, Raffman does not take the import of the figure-ground structure into consideration.  The question to consider now is this:  what is the figure and what is the ground in a perception of a slightly high pitch? In order to take the next step, I want to very briefly consider my criticism of Raffman, from my "Musical Musical Nuance."[11] 

In that essay, I criticize Raffman by highlighting her descriptions of perceptions of musical nuances.  Although I don't put it this way in that essay, perceptions of nuances in the figure role are her explananda.  Without appealing to the starting point common in phenomenology (describing the phenomena under consideration as they show up in ordinary experience), I argue in that essay that the target of an investigation into these musical subtleties should not be the slightly high pitches themselves but rather certain effects of those pitches.  And I am not referring to the effects standardly mentioned by Raffman, music theorists, and psychologists writing on such subtleties, namely, that a performer employs a pitch nuance in order to lead a listener to hear the musical structure as he, the performer, hears it. Rather, I am referring to nonstructural effects.  In that essay, I consider a quotation uncharacteristic of Raffman's book in which she mentions a nonstructural effect—the "brightening" of an interval (she subsequently downplays this and other nonstructural effects):

Many fine-grained differences in interval width—indeed the most interesting and important ones, for our purposes—are fully intended expressive features, as when a flutist ever so slightly raises (“sharpens”) an F-sharp sustained over a D-natural across a modulation from b minor to D major. . . . The flutist’s objective is to widen (“brighten”) the major third between D-natural and F-sharp, thereby emphasizing and strengthening the new key of D major.[12]

I suggest in that essay that one reason to prioritize nonstructural effects is that they constitute one set of reasons that musicians perform slight variations to begin with.  A musician strives to perform a slightly high A-natural (say) because it has an effect in the music, and one kind of effect is nonstructural, such as a "brightening." I go on to develop one piece of the puzzle for rendering such effects effable:  we can employ indirect description, metaphor, and comparisons to characterize such effects.[13]           

Now, the figure-ground structure is useful in clarifying subtleties of pitch and timing because it enables us to further clarify these subtleties, which, importantly, further preempts the ineffability conclusion. In order to begin to see why, it will be instructive to consider Raffman's example in terms of Dennett's three-step analysis. In an initial, typical perception, the figure is the "brightening" (analogous to the initial, vague E-string sound); the slightly high pitch is in the background (analogous to the E-string's overtones).  In the second perception, we need a way to imagine focusing our attention toward the slightly high pitch (recall that Dennett accomplished this by plucking the harmonic).  Imagine that we are at a rehearsal, and could isolate the flute by simply walking toward the flutist.  Upon clearly hearing the flute's slightly high F-sharp, that pitch becomes the figure.  We are hearing the pitch as slightly high, so let's follow Raffman and name it, call it an "F-sharp(3)."  In the third perception, we return to our original position in space, and hear the music altogether.  However, having been highlighted, the high F-sharp is still the figure.  Clearly, this is a different structure from the first perception; it is a different perception; it sounds different.           

Here is the question I have been working up to:  how should these initial perceptions be described?  Is there a way to describe the slightly raised F-sharp while it is in the background, in the first, typical hearing?  If it is correct that musicians often perform such subtleties for the very purpose of generating effects such as a "brightening," then it stands to reason that a salient kind of musical perception would be structured so as to hear the "brightening" as the figure.  If the F-sharp is in the background in that sort of perception, then in order to get clearer about such musical subtleties, we had better find a way of describing it.

5. Merleau-Ponty on perceptual indeterminacy

I want to work up to offering a characterization of the background features as perceptually indeterminate, in Merleau-Ponty's sense.  We should begin by noting that, along with Edmund Husserl, Merleau-Ponty maintains that when I see a house (say) I see it as a three-dimensional object, even though my perception is perspectival.  However, whereas Husserl holds that I hypothesize the parts of the house that I do not see, such as the back of the house, Merleau-Ponty maintains that I actually experience the sides of the house that are not determinately presented to me in perception.  Here, I am following Sean D. Kelly's account of the distinction between these philosophers on this notion of "object transcendence."[14]  Kelly writes, "Merleau-Ponty ... thinks that my current visual experience contains something that is itself an indeterminate presentation of the back [-side of the house]."[15]  In Merleau-Ponty's words, "The region surrounding the visual field is not easy to describe, but what is certain is that it is neither black nor grey.  There occurs here an indeterminate vision, a vision of I do not know what, and, to take the extreme case, what is behind my back is not without some element of visual presence" (PoP, p. 6).[16]         

Merleau-Ponty's notion of indeterminacy does not have to do only with that which is hidden.  In fact, as Kelly writes, "The canonical kind of indeterminate visual presence, for Merleau-Ponty, is the visual presence of the background against which a figure appears.  The background, insofar as it is experienced as a background, is visually present to a subject even though it makes no determinate contribution to his experience."[17]  Kelly argues that the indeterminacy of a background feature (say, the relative brightness of a light) consists in its normative effect; "the experience of the lighting context is essentially normative; I see how the lighting should change in order for me to see the color better."[18]  I will not follow Kelly to this normative claim; rather, I focus on the positive effects of indeterminate features vis-à-vis related emergent perceptual qualities, and I will draw support directly from Merleau-Ponty's texts, beginning with this important claim:  "We must recognize the indeterminate as a positive phenomenon.  Quality appears within this atmosphere" (PoP, p. 7).  Notice that while Dennett does not acknowledge the difference between the way features show up in the background versus how they show up as objects of attention, Merleau-Ponty is focused on just this distinction.            

Consider what Merleau-Ponty says about the Müller-Lyer lines (figure 2).  The horizontal lines are, of course, the same length, but they appear to be different lengths in the Müller-Lyer context. Merleau-Ponty claims that the horizontal lines are indeterminate in a normal perception of the illusion. Now, if we focus on each of the horizontal lines, so as to extract them from their context (ushering them into the figure role), then we can see that the two lines are actually equal in length; Merleau-Ponty takes this to be an unnatural way to view the illusion.  If we perceive the illusion normally, by not scrutinizing the horizontal lines, the horizontal lines do not look equal.  Interestingly, Merleau-Ponty says that they also do not look unequal.  In addition to the ordinary distinction between looking equal and unequal, he is suggesting a third option.  He says the lines look different. This third option will turn out to be perceptual indeterminacy.


         Figure 2: Müller-Lyer lines          

Describing the way these horizontal lines look in a typical perception of the illusion is to describe them neither as equal nor unequal.  A perceived horizontal line in isolation (when it is the object of attention, the figure) possesses characteristics that a perceived line in this context does not (that is, with the addition of auxiliary lines, when it is not in the figure role).  Therefore, in describing such an indeterminate perceptual feature, we will be mistaken if we characterize it as having the kind of determinate specificity that can ground qualities such as sameness or difference in length.  If we take determinate length to be a characteristic of these lines in this context (even determinate unequal length), we mischaracterize them.  This is what Merleau-Ponty is getting at when he writes, "The lines in Müller-Lyer's illusion cease to be equal without thereby becoming 'unequal'—they become 'different'" (PoP, p. 11).           

With regard to length, then, the Müller-Lyer lines are perceived as ambiguous.  What I want to emphasize is that this perceptual ambiguity is a perceptual-interpretive resting place; the ambiguity is preserved.  By allowing the lines to remain ambiguous in perception, the illusion is generated. Contributing to, or fostering, the illusion is the positive influence of the indeterminate features that Merleau-Ponty is referring to.  We ought to allow such indeterminate features to have this positive influence without seeking to render them determinate.  Such indeterminate features are not ambiguities to be clarified; indeed, if our goal is to correctly describe such an experience, we must not give a description of these features as clarified.  We must not clarify indeterminate features because this would mischaracterize their role in the experience.           

What I have been driving at is that the examples of musical subtleties I have mentioned are analogous to the Müller-Lyer lines:  the overtones in Dennett's first perception of the E-string are fruitfully described as indeterminate in Merleau-Ponty's sense.  They made a positive contribution to the E-string sound, and they made this contribution as ambiguous.  In an ordinary perception of the E-string their ambiguity is preserved.  Dennett treats this ambiguity as a kind of summoning (in Plato's sense [19]); Dennett sought to clarify the overtones, and as we have seen, the overtones became the figure; they became determinate in subsequent perceptions.           

Dennett's approach covers over a distinction between perceptions that becomes crucial in aesthetic experience.  To see why, consider the example of the "brightening" that is brought on by the slightly raised F-sharp.  I want to suggest that the F-sharp is perceptually indeterminate in an ordinary perception of this subtlety.  Mull over this question:  when one focuses upon the flute's raised F-sharp (noting that it is slightly high)—in that precise moment—does the quality of "brightening" emerge in experience?  If this example is sufficiently analogous to the Müller-Lyer lines, then the "brightening" will not arise unless the F-sharp is perceived as indeterminate, preserved as ambiguous; one cannot hear the F-sharp as an "F-sharp(5)" (say) and hear the "brightening" at the same exact moment.[20]        

My claim does not rely upon an analogy to the Müller-Lyer illusion alone; Merleau-Ponty makes this point more generally: some elements of a perception perform a function as indeterminate background features that they would not perform were they to be perceived as the figure, determinately.  Consider another example raised by Merleau-Ponty that concerns the perceptual effect of the reflection on human eyes.  (Merleau-Ponty is ultimately making a point about perceiving actual human eyes, but he makes the point by referring to techniques of painting.)

It took centuries of painting before the reflections upon the eye were seen, without which the painting remains lifeless and blind, as in the paintings by primitive peoples.  The reflection is not seen for itself, since it was able to go unnoticed for so long, and yet it has its function in perception, since its mere absence is enough to remove the life and the expression from objects and from faces. (PoP, p. 322)

The perceptual effect of the reflection on the eye is to give the face life and expression.  What Merleau-Ponty says next is based upon the idea that there are different ways of perceiving this reflection.  "The reflection is only seen out of the corner of the eye.  It is not presented as an aim of our perception; it is the auxiliary or the mediator of our perception.  It is not itself seen, but makes the rest be seen" (PoP, pp. 322-323).  The idea is that we can perceive an eye-reflection either as a figure (an object of attention:  "an aim of our perception") or as in the background ("out of the corner of the eye").  The reflection makes an important contribution to our perception of life and expression in a face; namely, it mediates that perception; seeing an eye-reflection indeterminately fosters our perception of the life and expression in a face.  Importantly, the reflection—as well as other background features such as lighting—would not have the effects they do were they not perceived indeterminately:  "Lighting and reflection only play their role if they fade into the background as discreet intermediaries, and if they direct our gaze rather than arresting it" (PoP, 323, emphasis in original).

6. Subtleties of duration:  groove

Consider another musical example, a musical subtlety of duration.  Ringo Starr's manner of playing various rhythmic patterns in early Beatles' recordings, such as "All My Lovin'," makes the rhythms "feel" as though they "lean" forward or "push."  This forward-leaning quality of his performances is one kind of groove (which, generally speaking, is the "feel" of a rhythm).  In contrast, the grooves of Led Zeppelin's John Bonham tend to "lean backward"; for example, the groove on "Blackdog" is backward-leaning.  The way these drum performances sound—the "feel" around which the other musical elements coalesce—is not merely a matter of the rhythmic patterns.  Of the elements that contribute to Ringo's grooves, timing nuances are the most crucial; Ringo strikes certain notes slightly early; similarly, by striking certain notes slightly late, John Bonham makes the rhythm "feel" as though it is leaning backward.  Leaving these particular examples behind, consider a particular kind of rhythmic pattern, a swing.  The ride cymbal element of a swing rhythm is shown in figure 3.  In performing a swing so that it feels as though it leans backward, a drummer strikes certain eighth notes late; the notes that are struck late are represented in figure 4 with arrows.  This backward-leaning groove is extremely common in jazz; for example, in the Count Basie Band's drummer Sonny Payne's performance on "Fly Me to the Moon (in Other Words)."[21]


Figure 3:  The ride cymbal element of a swing rhythm

Figure 4:  Ride element with slightly late notes indicated

Now, consider the perception in which the backward-leaning quality of the groove is the figure; this is an ordinary experience.  Drummers play these notes slightly late primarily so that other musicians and listeners experience this rhythmic "feel," not primarily so that other musicians and listeners notice that this strike is slightly later than the ordinary eighth note, and so on.  On the grounds outlined above in relation to pitch, I want to suggest that the slightly late eighth notes that generate the backward-leaning groove are not perceived determinately in that experience. In an ordinary perception, they do not show up in a way such that they could accurately be described as "eighth note(-3)" or "eighth note(-5)."  Rather, their specific duration is not perceived.  In the moment during which one focuses upon the late eighth notes, when they become the figure, in that precise moment, the backward-leaning quality drops out of the experience.  This claim rests not only upon an analogy to Merleau-Ponty's account of the Müller-Lyer lines but also his description and general claims about the perception of background features discussed above, such as reflections, lighting, and so on.         

It may help to consider a visual analog.  In his Cézanne:  A Study of His Development,[22] Roger Fry discusses one of Paul Cézanne's portraits of his wife, "Madame Cézanne in a Red Dress" (1888-1890).[23]  Fry mentions three aspects of the painting.  First, the portrait is boldly symmetrical; the model is facing forward, nearly centered, her arms are at her sides, the chair is rigidly rectangular, and so on.  But second, Fry emphasizes that, in spite of the straightforward, symmetrical design, the painting possesses a certain "vitality . . . . the palpitation of life."[24]  I believe that such a vital quality, an emergent Gestalt quality, is analogous to a musical groove.  Third, Fry seems to understand this vitality as being due to the slight variations of various elements of the painting.

Cézanne has instinctively corrected this [plain symmetry of design] in his detailed treatment. Everywhere this symmetry is modified by deformations:  the body leans slightly to one side, and the perpendicular of the dado behind becomes drawn into the movement . . . . The strongly marked edge of the dado seen in perspective affords by its emphasis the counterpoise to these uniform movements. Finally, as though not to break anywhere this general play of slight variations, this dado does not even keep a continuous line, but appears as though refracted where it passes behind the chair.[25]

Fry notes that the out of kilter dado was no mistake. Indeed, two additional portraits of Cézanne's wife, which Fry does not mention, also include an out of kilter dado / baseboard: "Madame Cézanne in a Yellow Chair" (1888-1890)[26] and "Madame Cézanne with Green Hat" (1891-1892).[27]  In all three of these paintings, an edge of the room's dado / baseboard passes behind the chair but is out of kilter; the edge that we see emerging from one side of the chair does not line up with the edge emerging from the other side.  I want to suggest that the out of kilter edges are analogous to the early or late eighth notes of a groove.  "Madame Cézanne with Green Hat" possesses not only a vital quality but also a quality of movement.  I am suggesting that, in all three of these paintings, the out of kilter edges mediate or foster emergent qualities of vitality or movement.

The relevant question for us to ask is whether, in the precise moments during which one scrutinizes the out of kilter elements, one perceives the quality of movement or vitality.  By analogy to the examples above, such as the reflections in eyes, my claim is that we do not perceive the movement or vitality; we only experience these emergent qualities in the moments in which we allow the out of kilter elements to remain in the background as indeterminate elements, preserving their ambiguity in perception so that they can function to mediate the emergence of the quality of movement or vitality.  Characterizing such elements as indeterminate in Merleau-Ponty's sense is an effective way of describing them while they are in the perceptual background.  This conclusion can also serve as a practical suggestion regarding active aesthetic experience:  one effective way to attempt to perceive emergent qualities such as a groove, a harmonic brightening, or a visual quality of vitality or movement, is to allow the background, contributing elements to remain ambiguous in perception.

7. Conclusion

I have claimed that describing perceptual experiences of musical subtleties by characterizing all of the relevant elements as they show up when they are objects of attention (as occupying the figure role) leads to inaccurate descriptions.  The background features of some musical subtleties—for example, a slightly high F-sharp, late eighth notes, or an E-String's overtones—are typically not determinately perceived.  Support for these claims is drawn from my criticism of Dennett's analysis, claims about what is typical in musical experiences, as well as analogies to claims Merleau-Ponty makes about certain visual experiences.  I have claimed that a promising option is to characterize background features as indeterminate in Merleau-Ponty's sense; such features perform a positive function, according to Merleau-Ponty, insofar as they mediate emergent qualities; I suggest that grooves, brightened harmonies, and so on, are examples of such emergent qualities.  Performing this positive function requires that these features be preserved as ambiguous in perception.  Support for this last claim rests on Merleau-Ponty's claims about the Müller-Lyer lines as well as background features such as lighting and reflection.          

A benefit of adopting my suggestions is that we can avoid descriptions of musical subtleties that are too coarse-grained, which fail to capture differences among perceptions (like Dennett's), while also avoiding the conclusion that such subtleties are ineffable (Raffman's claim).  Regarding the latter, first, as I suggested in "Musical Musical Nuance," we can render subtleties effable by (1) describing nonstructural effects of slight variations of pitch or duration (such as a "brightened" chord) via metaphor and comparison.  And as I have argued above, we can make further progress by (2) describing the perceptions in terms of the figure-ground structure, where the effects occupy the figure role, and slightly high pitches (etc.) are a part of the background; and (3) by clarifying how the background features show up and function in experience i­­n terms of Merleau-Ponty's understanding of perceptual indeterminacy and their mediating function.[28]


Tiger C. Roholt

Tiger Roholt is Assistant Professor of Philosophy at Montclair State University. His research focuses on philosophy of art, philosophy of music, and phenomenology. He is currently writing, Groove: A Phenomenology of Rhythmic Nuance (Bloomsbury).

Published on June 3, 2013.


[1] Daniel Dennett, "Quining Qualia," Philosophy of Mind: Classical and Contemporary Readings, ed. David J. Chalmers (Oxford:  Oxford University Press, 2002).

[2] Ibid., pp. 243-244.

[3] Ibid., p. 244.

[4] Maurice Merleau-Ponty, Phenomenology of Perception, trans. Donald A. Landes (Routledge, 2012 [1945]).  References to this book will be given parenthetically in the body of the essay, as "PoP," followed by the relevant page number.

[5] Diana Raffman, Language, Music, and Mind (MIT Press, 1993).

[6]  Eric F. Clarke, "The Perception of Expressive Timing in Music," Psychological Research, 51: 2–9 (1989), 3.

[7] Raffman, Language, Music, and Mind, p. 65.  Raffman is here drawing from Burns, E.M., and Ward, W.D. (1982), "Intervals, Scales, and Tuning," in D. Deutsch ed. The Psychology of Music (New York:  Academic Press), pp. 241-269.

[8] Ibid., p. 140.

[9] "Do not misunderstand:  certainly we could coin (type-identify) names for the N-pitches-names like, say, 'A-natural(1)', 'A-natural(12)', 'B-flat(10)', and so forth—just as we could coin names for all the determinate shades we can see.  But for want of the requisite schemas in long-term memory, we would not be able to apply those names 'by ear'.  In this respect the nuances are known ineffably are, as I shall say, nuance ineffable" (Ibid., p. 88).  For more on Raffman's ineffability claim, see my, "Musical Musical Nuance," The Journal of Aesthetics and Art Criticism, 68 (2010), 1-10.

[10] In her book, Raffman criticizes Dennett on other grounds:  for not acknowledging that we can discriminate more finely than we can conceptualize.

[11]  Tiger C. Roholt, "Musical Musical Nuance," The Journal of Aesthetics and Art Criticism, 68 (2010), 1-10.

[12] Raffman, Language, Music, and Mind, p. 66.

[13] By comparisons, I mean that we can point out, for example, that this "brightening" is brighter than the "brightening" on that recording.

[14] See Sean D. Kelly, "Edmund Husserl and Phenomenology," in The Blackwell Guide to Continental Philosophy, eds. Robert C. Solomon and David Sherman (Blackwell, 2003).

[15] Sean D. Kelly, "Seeing Things in Merleau-Ponty," in The Cambridge Companion to Merleau-Ponty, eds. Taylor C. Carman and Mark B. N. Hansen (Cambridge University Press, 2005), p. 81.

[16] Kelly draws upon a portion of this quotation; I am adopting Kelly's adjustment to the translation; see pp. 80-81 of his "Seeing Things in Merleau-Ponty."  The visual field includes objects of attention as well as the background; Merleau-Ponty's conception of the perceptual background is obviously quite broad.

[17] Ibid., p. 82, emphasis in original.

[18] Loc. cit., emphasis in original.

[19] For Plato, ambiguous perceptions are "summoners;" that is, unclear perceptions, he writes, "summon the understanding to look into them" (Plato, Republic, trans. G. M. A. Grube and C. D. C. Reeve [Hackett, 1992], 523a); they are unclear perceptions of what Plato would call sensible particulars, which draw us toward knowledge of the Forms.  From this perspective, these properties are seen negatively, merely as calling out to be clarified.

[20] Eric F. Clarke seems to recognize something like this while not noticing the implications for his project:  “Auditory events, or more specifically musical events, are inherently multi-dimensional, and. . .although they may be theoretically, and even empirically, decomposable into unitary components, this may destroy or conceal emergent properties of the whole event” (Clarke, “The Perception of Expressive Timing in Music,” p. 3, emphasis added).

[21] From the Frank Sinatra LP, Sinatra at the Sands (Reprise Records, 1966).

[22] Roger Fry, Cézanne:  A Study of His Development
(New York:  Macmillan, 1927).

[23] Paul Cézanne, "Madame Cézanne in a Red Dress," oil on canvas, 1888-1890  (Metropolitan Museum of Art,
New York).
(accessed 27 April 2013)

[24] Fry, Cézanne:  A Study of His Development, p. 68.

[25] Ibid., p. 68.

[26] Paul Cézanne, "Madame Cézanne in a Yellow Chair," oil on canvas, 1888-1890 (Chicago Institute of Art, Chicago).
(accessed 27 April 2013)

[27] Paul Cézanne, "Madame Cézanne with Green Hat," oil on canvas, 1891-1892 (The Barnes Foundation, Philadelphia).
(accessed 27 April 2013)

[28] I would like to thank those who made helpful suggestions following presentations of earlier versions of this paper—especially at the 2012 meeting of the Royal Musical Association Music and Philosophy Study Group (London, Kings College) and at the Bucknell University Philosophy Colloquium (2012).  For thought-provoking discussions about these issues, I want to thank Hanne Appelqvist, Chris Bartel, David Clowney, Lydia Goehr, Ted Gracyk, Sean Kelly, Jerrold Levinson, Margaret Moore, Jonathan Neufeld, and Jill Rosenberg.