The Musicality of Speech

1. Introduction

Much of the sound of our speech — its prosodic profile — has little or nothing to do with the words we utter, or what we mean to communicate by uttering them. For instance, you and I could each speak the same sentence, and in uttering it, we could each mean the same thing. And yet our two utterances will sound quite different as a result of the unique timbres of our two voices, our accents, and our idiosyncratic styles of expressing words through melody and rhythm.

All speech can be heard as music, as song. Almost any evaluative criteria for assessing the aesthetic merit of singing can, it seems, also be applied to speaking. Speech can be musically beautiful, and it can also be kitsch (perhaps by being heavily affected, for instance). The combination of pitch, rhythm, tempo, timbre, and dynamics in any given vocal performance can strike one as lovely, as intriguing, or in one way or another as tone-deaf, hollow, or perhaps even clichéd. Indeed, the possibilities for musical evaluation of speech are multiplied further by the aesthetic significance of the interaction between the semantic and pragmatic content of speech, on the one hand, and its strictly prosodic features, on the other. There could be significant aesthetic quality, for example, in the way the melody of a person’s accent can be heard more clearly when they speak of matters closer to their heart. Attention to this interaction too can yield rich aesthetic experiences.

Such experiences typically affect relations between people. Many of us are in the habit of allowing aesthetic impressions made by a person’s speech to affect how we are inclined to feel about them. At the extremes of this habit, we may allow the impressions of a person’s speech to lead us to distrust them, dislike them, or, indeed, to be strongly endeared to them. This raises a question. Can such feelings ever be acceptable? That is, can it be legitimate to like or dislike a person on the basis of the apparent aesthetic qualities of their speech?

My goal in this paper is to defend an affirmative answer to this question. But first it is necessary to give expression to a compelling negative answer. While little has been written in philosophy about evaluating the voice, more has been said about other modes of aesthetic appraisal of persons, especially about judgements of the visual appearances of people’s bodies. Drawing from those discussions, in section 2, I articulate two challenges to the idea that it could ever be legitimate to like a person in virtue of their aesthetic qualities. In short, the challenges are that, first, our aesthetic judgement of others is too clouded by prejudice to be reliable and, second, even if we could make unprejudiced aesthetic appraisals of people, their aesthetic qualities are irrelevant to the matter of how likeable they are. In section 3, I focus on vocal aesthetics and show the dynamics of these two problems for interpersonal aesthetic valuing in general as they play out in that domain in particular.

My response to the challenges is to carve distinctions, in section 4, between different types of judgement, often conflated: of attractiveness, conformity with beauty norms, and beauty itself. I argue that it is possible to judge some speech to be beautiful in a way that could bear on the worthiness of the speaker for love or trust that could go beyond merely reproducing prejudices. Then in sections 5 and 6, I put forward the positive case for thinking that beyond being conceptually possible, voices do in fact sometimes bear such aesthetic qualities as to legitimately ground attitudes such as love, trust, and possibly dislike and distrust.

The aesthetic dimension of speech is a facet of ordinary experience that calls, in its own right, for philosophical reflection. At the same time, the puzzle that it raises about how we ought to respond to aesthetic impressions in the course of interpersonal life is a more general issue and is not specific to the aesthetics of speech. As a microcosm of that issue, the musicality of speech can provide useful concrete examples for thinking through the more general problematic while also serving to suggest that the kinds of aesthetic experience that is relevant to our social lives are more richly multi-sensory than is sometimes assumed.

2. The General Problem of Judging People Aesthetically

Irvin (2017) has recently drawn attention to the social problem of aesthetic interpersonal judgement. Summarising a wealth of empirical evidence, she diagnoses the situation thus:

The issue, then, is not that attractive people are treated a bit more nicely than unattractive people. Instead, we have a picture whereby, from the moment of birth, attractive people (with a few exceptions) accrue positive social capital in families, schools, and workplaces, while unattractive people pay a very substantial penalty that may involve less positive parental attention, less support from teachers, less recognition for their qualifications, less help when they need it, more punishment, and so forth. Some are routinely teased, bullied, dehumanized, and ostracized. These judgments often interact in disturbing ways with race, gender, disability, age, and gender identity, among other aspects of social identity. (p. 5)

It is in response to this first challenge, combined with the injustices inflicted upon those deemed less attractive, that Irvin (2017, pp. 9–10) proposes that we reimagine what it is to have aesthetic experiences of our bodies — and the bodies of others. This involves, consequently, reimagining the kinds of aesthetic qualities people and their bodies might be found to bear. Her proposed practice of ‘aesthetic exploration’ promises to subvert the conventions of aesthetic judgement. As such, it is a strategy for resistance against the harmful social effects that such conventions tend to bring about. An example that Irvin offers is of competing aesthetic approaches to the scarred skin of someone who has been severely burned. A conventional and all-too-common reaction may be to cringe, to recoil, perhaps also to judge the burned skin to be ugly. But if we encounter this skin in a spirit of aesthetic exploration, then we may displace the conventional reaction and instead take aesthetic pleasure in the visual and tactile qualities of the burns themselves (Irvin, 2017, pp. 12–13).

In spirit, Irvin’s proposals are in line with other subversive responses to the injustices of dominant beauty standards. Think, for instance, of the movement around the hashtag #bodypositive.¹ At the time of writing, that hashtag appeared in over 14 million Instagram posts asserting the beauty of bodies whose beauty is conventionally denied. Irvin’s strategy of aesthetic exploration, like the body positivity movement, challenges society’s beauty standards directly, claiming that conventional aesthetic judgements are problematic just because they are wrong about what is and what is not in fact beautiful.²

However, despite their commonality, there is a striking tension between Irvin’s approach and that of the body positivity movement. The tension arises over an issue that Irvin does not expressly address but which her account throws into sharp relief. The central current of the body positivity movement leaves intact a fundamental assumption of the dominant mode of aesthetic interpersonal judgement; indeed, it takes this assumption for granted. That is, it is assumed that if someone is beautiful, then that is a valuable feature about them in virtue of which others may be right to think of them highly. However, though she does not say so directly, Irvin’s notion of aesthetic exploration points towards a rejection of this assumption.

On the dominant, conventional approach towards beauty, being beautiful is a trait that some people have, but others do not. It is a special trait to be cherished, savoured by others, and admired. Whether it is thought of as a gift or an achievement, it is a valuable trait of the person. A dominant ideology about beauty may even go so far as to include putative justifications, explaining why beautiful people deserve to be liked, while ugly people deserve to be distrusted and shunned. Perhaps, for instance, the beautiful are thought to have earned their beauty through ‘looking after themselves,’ whereas the ugly have ‘let themselves go.’³ The dominant ideology putatively offers some justification for liking a person in virtue of certain of their aesthetic qualities.

But if, as Irvin suggests, the aesthetic pleasure we take in the bodies of others is a creative response to their diverse affordances, as the aesthetic exploration strategy would have it, then it becomes less clear that any quality of a person’s body could license a valuing-attitude towards that person. In aesthetic exploration, the connection between the body that affords the experience, and the person whose body it is, seems merely incidental. That is, whatever qualities a person might have in which one might take pleasure through a creative process of aesthetic appreciation — qualities such as the ethereal smoothness of some scarred skin — those qualities are not necessarily connected in any way to the character of their bearer; they need not be qualities for which the bearer is in any way responsible.

Such qualities are irrelevant to the question of whether one should like the person whose qualities they are. In discussions of love and friendship, it is common to cite Yeats’ poem ‘For Anne Gregory’ which features a girl who wishes to be loved for herself alone and not for her yellow hair. Proponents of otherwise divergent views agree with the girl in the poem that such qualities as having yellow hair ought to be irrelevant to the matter of one’s love-worthiness.⁴ The same can presumably be said for qualities like those found in scars via the process of aesthetic exploration.⁵

The argument to this point has been too quick on one score because it remains unspecified exactly what kinds of valuing-attitude are in question. An objector could point out that if some aesthetic exploration of a person and their body yields the discovery of an aesthetic quality the perception of which is enjoyable, then for that very reason the perceiver does have reason to value the person. That is, they can value them simply qua the source of a pleasant aesthetic experience. This objection is right as far as it goes, but it invokes a different sort of valuing-attitude than the one I mean to call to mind in this paper, which should thus be clarified. A good name for the type of valuing that should be in focus is personal liking.

Personal liking (or, for short, simply liking) is distinguished from other valuing-attitudes by the type of object to which it can be an appropriate response and by the type of response that it is. Liking a person is a way of treating them as having final, rather than instrumental, value. Thus, on the one hand, personal value is distinct from various instrumental valuing-attitudes that one might have towards a person. This takes care of the objector mentioned in the previous paragraph who values a person because and insofar as that person is aesthetically pleasurable to perceive. Even when such instrumental valuing-attitudes have people as their object, their proper object is not a person as such, but a person as (in this case) a source of pleasurable experience. By contrast, the proper object of liking is always a person as such.

On the other hand, liking is distinct from other valuing-attitudes whose object is also a person as such but which respond to the people who are their objects with reserve. One widely used sense of the term ‘admiration’ captures just such an aloof valuing-attitude. In admiring a person in this way, one recognises something about them that is good and is a good thing about them, as a person, but one need not thereby feel drawn any closer to that person, or have any desire to interact with them or for their recognition. By contrast, liking a person just is responding to something good about a person by feeling drawn to them, which is to say, by wanting some form of interaction or communion with them.

For my purposes in the remainder of the discussion, I will make some assumptions about personal liking so conceived: that it is a familiar attitude from ordinary interpersonal life; that it bears some close association with ‘affective trust’ (Jones, 1996); that it is an attitude that can be justified, or unjustified, according to whether the judgement that it entails about its object being worthy of liking is true, or not;⁶ that it can be justified by the presence of certain qualities that mark people apart from one another, such as charms and virtues; and, more precisely, that the set of qualities that can justify liking includes moral and aesthetic qualities.⁷ I also assume — though less hangs on this — that there is a corresponding negative valuing-attitude that might be called personal disliking.

With these assumptions in hand, it is possible to restate my driving question and summarise the two challenges to any positive answer. The question is whether the aesthetic qualities of a person’s speech could ever justify personally liking them. The first challenge is that any judgement of a person’s aesthetic quality seems always to rely on unjustifiable social prejudice. This is the matter to which Irvin responds, suggesting that the problem can be surmounted through a radical reimagining of the aesthetic experience of other people and their bodies. But this response to the first problem illuminates the second: that a person’s aesthetic qualities seem irrelevant to whether they should be personally liked. As I will now argue, these two challenges are manifested in quite complex ways in the particular context of vocal aesthetics.

3. Judging Voices

It would be naïve to suppose that the vocal domain is impervious to the social prejudices in which other modes of interpersonal aesthetic judgement are mired. And yet, at the same time, it cannot be taken for granted that the same types of prejudice that Irvin discusses also threaten to undermine liking someone in virtue of their speech. Rather, it seems worthwhile to consider the ways in which prejudice and stigma pertaining to race, gender, class, and disability are conventionally associated with various facets of prosody, including pitch, timbre, fluency, and accent. My attempt in the final two sections below to vindicate the practice of liking people in virtue of the aesthetic qualities of their speech can only be plausible if it takes stock of a clear picture of the current widespread injustices caused by that very practice.

The aesthetic presentation of the voice is subject to its own beauty norms. It can be difficult to say with confidence exactly what the socially dominant criteria are, in any given context, that govern which voices people find to be attractive or beautiful.⁸ One thing that is quite clear, though, is that having a ‘normal’ sounding voice is widely perceived as a good thing (Bruckert et al., 2010). Having an attractive voice is regarded as important. It informs judgements of trustworthiness and other positive personality traits (Zuckerman & Driver, 1989). For this reason, it is important to recognise both the injustice and the arbitrariness of the dominant customs of aesthetic sensitivity to the voice.

There are several axes of injustice in the sphere of vocal appraisal. Perhaps the most glaring is the way that the importance attributed to vocal beauty makes life difficult for those who are unable to speak using the larynx (or voice box). The surgical removal of all or part of the larynx is not uncommon in the treatment of laryngeal cancer, and consequently this process accounts for a significant number of people — and there are yet others — who must use other means than the larynx in order to speak. Alaryngeal speech sounds unfamiliar to many people’s ears and does not easily conform to typical expectations of what a beautiful voice ought to sound like. As such, alaryngeal speakers are regarded with suspicion, treated as ‘other,’ or, in the words of one alaryngeal speaking YouTuber, QuietBob, ‘I sound like a robot or a really big bug’ (Marshall, 2014, p. 3). The harsh judgement of those who are disabled vis-à-vis the capacity to speak using the larynx is a judgement that begins with the aesthetic impression of their speech and proceeds to make substantial inferences about character traits and the way such people ought to be treated: as suspicious or difficult to befriend. Since the fact of their laryngectomy (or, indeed, any other likely cause of their speech being the way it is) has nothing whatsoever to do with their likeability, their trustworthiness, or any aspect of their personality, these judgements are, obviously, arbitrary, and they constitute an injustice.

Although those without a fully functioning larynx are disadvantaged in the most absolute way by the importance that many people attach to a person’s voice, they are far from the only group that suffers the injustice of being systematically misjudged by their speech. Listeners’ expectations of what a voice ought to sound like might often be keyed to the social identity that they attribute to the speaker. Such coded expectations are refracted through the prisms of race, gender, and class. The result is that certain apparently aesthetic judgements are associated with certain identities inferred from the pitch, timbre, accent, and fluency of speakers’ speech, among other qualities. A telling example of this phenomenon is provided by the concept of shrieking.

‘Shriek’ appears to be a doubly thick concept: it has a descriptive component and two distinct evaluative components. The descriptive element identifies shrieking with high-pitched sounds and thus associates the sounds with adult female voices rather than male. One evaluative component, which is at least a common connotation of the concept, is that shrieking is aesthetically bad, other things equal. That is, to describe a voice as shrieky is to imply that to listen to it is an unpleasant experience. The second, related evaluative component of the concept is ethical and proscribes shrieking as overly dramatic, demanding, or even hysterical. Admittedly, perhaps, the concept does not imply anything as strong as that shrieking is impermissible but that it is better not to shriek if it can be avoided, and being unnecessarily shrieky is not merely an aesthetic but also an ethical flaw in a person’s manner of speaking. Given that these evaluative elements are bundled together by the concept with the descriptive association with the adult female voice, they are gendered evaluations. In conventional usage, then, the availability of the concept ‘shriek’ is a resource primarily for the negative aesthetic and ethical evaluation of women’s voices but not men’s.⁹

The development of certain evaluative concepts that regulate some voices but not others is one mechanism by which the practice of judging people on their voices can produce injustices. As noted, gender is not the only axis along which such mechanisms serve to infuse social prejudices into the immediate aesthetic reactions to vocal expression. Another axis is race (Delfino & Kosse, 2020; Eidsheim, 2014; Mendieta, 2014). There exist a variety of racial expectations that pertain specifically to the voice: not just to dialects and choices of words, but to particular timbres and patterns of intonation. Via these expectations, racial stereotypes permeate common appreciation of vocal beauty.

The beauty or ugliness of a person’s accent is an especially familiar form of aesthetic judgement of speech. For instance, one study seems to confirm the stereotype that the Brummie accent — local to the city from which I am writing — is perceived to be the ugliest regional accent in the UK (Malarski, 2013). Reinforcing the connection mentioned above between aesthetic judgements of a voice and more substantive judgements about the person, that same study also found Brummie speakers to be perceived as the least intelligent. Aesthetic attitudes towards accents are, unfortunately, harmful. As an immediate indicator of class, they are a medium for class prejudices which operate, in economically stratified societies, to preserve the class privilege of those already better off (see, e.g., Lippi-Green, 2012).

Gender, race, class, and disability are salient dimensions of social identity that are apparent in a person’s speech. In each of these dimensions, there are familiar evaluative criteria through which stigmas and prejudices towards social identities are manifested in aesthetic judgements of people’s voices. The existence of these unjust aspects of the sensitivity to the music of speech reintroduces in the auditory sphere the first challenge that I raised above for aesthetic interpersonal judgement in general: namely, that it may be that aesthetic judgement of speech always relies on such unjustifiable prejudices. Even if this challenge could be answered, the second problem lies in wait. That is, again, that there may be no connection between the aesthetic quality of a person’s speech and their value as a person. If judgements of a person’s likeability or trustworthiness were inferred from judgements of a person’s voice as ugly in virtue of being shrieky, or in virtue of a Brummie accent, or any similar putatively aesthetic quality born from social prejudice, then such inferences would patently be fallacious.

4. A Tripartite Distinction in ‘Aesthetic Judgement’

In the discussion so far, the possibility of legitimate interpersonal valuing-attitudes grounded in aesthetic impressions has been called into question by the prevalence of a variety of factors at play in typical interpersonal aesthetic judgement, especially in the auditory sphere. Those factors were often unjust. But even if they were not unjust, they were irrelevant to the evaluative questions at issue: about whether a given instance of speech is beautiful and whether such beauty licenses liking the speaker. However, in this section I want to argue that the prevalence of these erroneous factors does not in fact show that there cannot in principle be well-founded interpersonal aesthetic judgements.

The strategy for showing that such well-founded judgements are at least conceptually possible is to carve some distinctions among three categories of judgement that might otherwise be conflated into the general notion of ‘interpersonal aesthetic judgement.’ Those are judgements of attractiveness, judgements of conformity with social beauty norms (which I will call judgements of “beauty”), and judgements of beauty per se. They are conflated in the sense that often in ordinary interpersonal experience, people make impressions on one another via their appearances (visual, auditory, and otherwise), and those impressions prompt corresponding valuing-attitudes without the subject of the impression distinguishing between impressions of attractiveness, “beauty,” or beauty. As a result, one’s judgement of another person’s beauty might be erroneous if it were made just on the basis of their being attractive or “beautiful.” The idea, then, is that by teasing apart these categories, it could be possible for each of us — as aesthetic judges of others — to critically identify the factors discussed above as figuring in judgements of attractiveness and judgements of “beauty” but to exclude them as far as possible from judgements of beauty. Once such distinctions are drawn, the question of whether a person’s beauty could legitimately ground liking them can be posed again without risk of being clouded by the many prejudices involved in liking others in virtue of their attractiveness or their “beauty.” If this strategy is successful, then, it will answer the first of the two challenges articulated above.

To begin, consider judgements of attractiveness.¹⁰ One way to draw the beauty-attractiveness distinction is loosely inspired by Kant’s approach to aesthetics. Kant distinguished between two kinds of pleasure that one can take in perceptual experience: interested and disinterested.¹¹ Interested pleasure is grounded in some desire that one has for the thing that one perceives. Kant used the term ‘agreeable’ to refer to things the mere perception of which is pleasurable in virtue of some contingent desire of the perceiver’s. So, judgements of attractiveness are of agreeableness in Kant’s sense, and the pleasure one might take in perceiving an attractive person, as such, is an interested pleasure.¹²

Kant held that judgements of beauty cannot be judgements of agreeableness. Experiences of beauty carry the implication that the perceiver judges the object to be aesthetically good. If it is aesthetically good, then it has a value that in principle others also ought to be able to appreciate and enjoy. However, one can experience something as agreeable without taking it to be aesthetically good or to be such that others would like it too. Judgements of agreeableness, then, relevantly including judgements of attractiveness, are not the same thing as judgements of beauty.

To illustrate, consider an example of the impression of attraction that can be created by a voice. Let’s suppose that A hears a recording of a stranger, B, speaking in a language that A cannot understand. A is nonetheless struck by seductive qualities in B’s tones: a husky timbre; audible breathing; and slow, halting, rhythmic speech. Such prosodic characteristics might imaginably be attractive to A. What makes them attractive is an interesting question, but it does not matter for present purposes. All that matters is that since A is listening to a recording, and one the meaning of which A does not grasp, what A is attracted by has nothing to do with any existing relationship between A and B, nor with the content of B’s speech, but rather with B’s voice and manner of speaking. This might be the auditory analogue of being attracted to a person on the basis of seeing a photograph of their silhouetted figure.

In finding B attractive, A judges B to be agreeable but not to be beautiful. A’s appreciation of B’s voice attributes a value to B but does not imply that any other well-placed listener should attribute the same value to B. Attraction does not imply an attribution of value that is universal in this sense. This is for the simple reason that A’s attraction to B finds a value in B that is partly grounded in A’s idiosyncratic desires: to be attracted to certain kinds of people, to desire intimacy of certain kinds. Since these desires are not universal (in the sense of being shared by everyone), B’s value for A is also not universal, and such value thus contrasts with the sort of value one attributes in making a judgement of beauty per se.

As an example of the latter, take Sonny Rollins’ recording of the Duke Ellington tune ‘In a Sentimental Mood’ (especially the way Rollins plays the head the second time round). I think that Rollins’ playing there is beautiful, which is to say, I think it is aesthetically good. Listening to that music, merely perceiving it, is a pleasurable experience for me, and I contend that that is a result of its aesthetic goodness. In so judging Rollins’ playing, I hold — just as anyone must hold when they sincerely make a judgement of beauty — that anyone else who recognises the quality ought to, thereby, have a pleasurable experience, other things equal. If someone in the right frame of mind and without distractions finds listening to the track unpleasurable, then I must think that either they are missing something and have not grasped its key qualities or that I was wrong after all about the beauty of the playing. So, judgements of beauty per se are distinct from judgements of attractiveness because the former are universal, whereas the latter are not.

The second category to be separated from judgements of beauty is the category of judgements of conformity to beauty norms, or “beauty.” As in the case of the previous category, judgements of “beauty” are often conflated with judgements of beauty.¹³ But “beauty” must be distinct from beauty. Something is “beautiful” if it conforms well to the social standards that dictate what “beauty” consists in, in each social setting. Ideals of how people ought to appear are constructed by powerful voices and industries in a society.¹⁴ But within any cultural setting which has some established convention for what counts as “beautiful,” it is always an open question whether what is “beautiful” is indeed aesthetically good. That is, thoughts of the following form are always intelligible: it is “beautiful,” but it is not beautiful.

To illustrate, consider a speaker whose diction, intonation, and cadences of speech are a perfect match with the social setting they find themselves in. A listener could be struck by this fact itself: by the fact that the way the speaker talks is exactly how it is deemed cool, or proper, or otherwise good and admirable to talk in company like this. This might be the vocal analogue of being struck by how a person’s clothes are perfectly in line with the season’s fashion. My present point is simply that being struck by the fact that another’s speech is exactly such as to be deemed good or beautiful by their society need not imply that one finds the speech good or beautiful. Even if, as the perceiver, one happens to endorse the social standards that find some speech to be beautiful, it is one thing to judge that those standards apply perfectly to this particular piece of speech and another thing to judge that those standards themselves are correct about what kind of speech is good. Once again, such speech is “beautiful” but not necessarily beautiful.

This last point, which echoes Moore’s (1993, pp. 62–69) open question argument, gestures towards a useful conception of beauty per se that is clearly conceptually intelligible and, just as clearly, distinct from attractiveness and from “beauty.” Namely, beauty is just the property of an object’s being aesthetically good. One popular analysis of aesthetic goodness is that for an object to be aesthetically good is for it to be such that, in appropriate conditions, appreciation of it brings about a certain kind of pleasure.¹⁵ Moreover, this pleasure must be grounded in the object itself rather than in some extraneous desire of the perceiver. Since my goal in this paper is not primarily to defend that conception of beauty against rival accounts, I will take the analysis as an assumption. However, it will be helpful to address two objections to clarify this conception of beauty and to further support my claim that it is distinct from those other categories of judgement, attractiveness, and “beauty.”

The first objection is voiced from the perspective of those who remain suspicious that beauty is different from attractiveness in the Kantian way that I have suggested. The objection says that it is psychologically implausible that we could find something pleasurable to perceive unless we had some sort of desire for it.¹⁶ That is, one might think that the Kantian conception of beauty as the appropriate object of disinterestedly pleasurable perception is implausible if it dissociates pleasure too far from the satisfaction of desire.

But as far as this paper is concerned, we need not suppose that aesthetic pleasure is quite so radically removed from the perceiver’s desires — the Kantian distinction between agreeableness and beauty can be made in a slightly more modest way. Let us isolate a certain set of attitudes that one can have towards a desirable object and group those attitudes together as desires for union with the object. These include sexual desires and gustatory ones and hoarding-like desires. There is a certain kind of pleasure to be had from the mere sight of one’s lunch, not to mention the smell. Such pleasure is plausibly connected with one’s desire for the relevant type of union with the object (in the case of one’s lunch: eating). Call this set ‘desires of union.’ Now, Kant’s point can be remade with this notion in hand.¹⁷ It is possible for us to take pleasure in the mere perception of an object even when we do not, plausibly, have any kind of desire for union with it. Salient examples of aesthetic judgements (like musical appreciation) are not plausibly grounded in desires of union. Guided by such examples, we might follow Kant in thinking of aesthetic pleasure as independent of desires of union while keeping open the possibility that there could be some other type of desire that accounts for any such aesthetic pleasure.

The second objection strikes on the other front, challenging the distinction between beauty and “beauty.” This objection begins with the thought that all aesthetic judgements are made with reference to intersubjective standards that are in some sense socially constructed. From here, it seems that if judgements of “beauty” are judgements of conformity with social standards of beauty, then there is no contrasting category of judgement that is any less based on social standards.

Again, it is possible to concede quite a lot to this objection and still to recognise the conceptual possibility of judgements of beauty per se as a distinct category. The open question–style argument mentioned above has the consequence that whatever one wants to say about the criteria on which each of us finds objects of our experience to be aesthetically enjoyable, those criteria must be conceptually distinct from the dominant norms in a society that govern attitudes of how people are made to feel that they ought to present themselves and their bodies.

Suppose that in a particular social setting the extension of judgements of “beauty” exactly overlaps with the extension of judgements of beauty. Thus, good judges in that social setting take disinterested aesthetic pleasure in exactly the same qualities in people that are prescribed by societal beauty norms, that is, imposed as ideals of personal beauty by powerful forces in that society. Suppose further that the origins of the criteria of beauty and of “beauty” are also identical. Thus, the same historical factors that led people to find inherent value in certain qualities also led them to hold those same qualities up as an ideal. Even in such a context, it must be intelligible to ask whether what is “beautiful” is really beautiful. Or, to put the point in other words, aesthetic judgements are inherently normative whereas judgements of conformity with beauty norms are not.¹⁸

In this section I have argued that judgements of beauty are distinct from judgements of attractiveness and of “beauty,” despite these three categories often being conflated both in ordinary evaluative thought and in academic discussions. The point of making these distinctions is to make possible the realisation that many of the arbitrary and unjust features of interpersonal aesthetic judgement discussed in the previous sections play a role not in judgements of beauty per se, but in judgements of attractiveness and of “beauty.” Hierarchies of power and status go into the very construction of what (and who) counts as attractive and what counts as “beautiful.”¹⁹ This is not to say that judgements of attractiveness or of “beauty” are necessarily prejudiced or unjust, just that they often are, and that many of the prejudices associated with beauty stem from them. It may be that judgements of beauty per se can themselves be prejudiced — this is a point I will return to below. Nonetheless, if we can undo the conflation of beauty with these distorting loci of prejudice and stigma, then our sensitivity to beauty itself will have a better chance of avoiding such unjust patterns of thought.

Moreover, the tripartite schema makes clear that we must undo that conflation for a further reason. The distinctions help to illuminate the connection between aesthetic qualities and personal value. Once we recognise judgements of attractiveness and of “beauty” for what they are — that is, depriving them of their illicit association with judgements of beauty — we can see that they have little to do with the personal value of the people whom they judge. In the earlier discussion, I made clear that the notion of personal liking that should be in focus is a way of responding to a person as being finally valuable. Being attractive is a way of being valuable to someone, but it is a way of being valuable to them insofar as one is the fitting object of their idiosyncratic desires, and thus of being valuable insofar as one could serve the satisfaction of those desires. Therefore, attractiveness is a form of instrumental, not final, value. As such, one cannot justifiably like a person in virtue of their being attractive. Yet to be “beautiful” is not a way of being valuable at all, since the fact that one conforms with a set of social standards of appearance (vocal or otherwise) is compatible with those standards being wrong in what they find to be beautiful. So, again, judging someone to be “beautiful” is no justifiable basis for liking them.

There are two tasks remaining for my attempt at a vindicatory account of the custom of being sensitive to the music of others’ voices. The first is to consider the nature of the beauty of speech. Having argued for the conceptual possibility of judging a voice to be beautiful — as opposed to attractive or “beautiful” — I now want to consider whether people’s speech really does bear scrutiny from this distinctly aesthetic perspective. I will try to show that some speech does, and this is something to which we are — sometimes at least — already attuned. Moreover (and secondly), I will offer some basis for thinking that the aesthetic qualities of speech can sometimes provide epistemic access to (dis-)valuable qualities of people’s characters, and so in this way that aesthetic judgements of speech can in principle provide sound bases for interpersonal valuing-attitudes. In the final section below, I will pull together the elements of my discussion and present a summary of the vindicatory account.

5. Lyricism

It is fairly easy to imagine what kinds of vocal qualities people might find attractive and, relatedly, which qualities might be promoted by beauty norms. Namely, these will likely be vocal traits that suggest bodies of the sort that people are otherwise attracted to, or that are socially prescribed, respectively. For instance, testosterone levels, height, and weight can all be heard, to varying extents, in the voice’s depth, the resonance and clarity of the timbre, its volume.²⁰ But once these qualities are set aside, what kind of further qualities might speech bear that are simply aesthetic?

In answer to this question, a core concept in the appreciation of the musicality of speech, I suggest, is lyricism. Lyricism is a form of musical beauty: for the production of a sound to be lyrical is for it to be aesthetically good and to be so in virtue of some complex function of its sonic elements. In other words, lyricism is not a quality of any specific sonic component of speech such as timbre, intonation, rhythm, dynamic variation, or tempo. This point might not be entirely obvious: one might think that, for example, the timbre of a person’s voice could itself be more or less lyrical. But an idea familiar from instrumental music helps to clarify matters. Any musical instrument has capacities and limitations for the timbres it can produce, as well as its range of notes, and of dynamics, among other things. The capacities of an instrument thus form the boundaries of expression within which a given player’s playing of the instrument could be judged to be more or less lyrical. In this manner, it is possible to subject the playing, rather than the instrument, to musical evaluation. The same idea serves to isolate the musical quality — in this case the lyricism — of a speaker’s act of speaking rather than merely their voice.

It is instructive to consider in more detail what it is about an act of speaking that makes it more or less lyrical. Lyricism is a type of expressiveness — namely, a beautiful or aesthetically good type of expressiveness.²¹ As such, an act of speaking is lyrical when it uses its purely prosodic elements well to express its sentiment. Such expression through purely prosodic elements might involve variation between tempi using slow and fast sequences of words or syllables; variation from the expected melodic intonation as predicted from the speaker’s accent and idiosyncratic speaking style (thus, neutralising or leaning into one’s accent at choice moments could count as expressive along this dimension); speaking in such a way as to bring out the distinctive qualities of the timbre of one’s voice, i.e., allowing the mellowness or warmth or breathiness or resonance of one’s voice to be heard; and so on.

It bears repeating, though, that more than any one of these factors alone, the lyricism of speech is determined by the interplay between them. Consider someone whose words are spoken softly just when one expected them to be firm, and whose cadences in this moment involve intervals ever so slightly more pronounced than they normally might be. If such speech is beautiful to behold, then, in being impressed by it, a listener is impressed not just with the bare sound and its subtleties but also with the speaker as the producer of those subtleties. To be sensitive in one’s speech to such aesthetic factors is impressive. In this respect, lyricism in speech is analogous to the aesthetic quality in instrumental music that gives a listener the impression that the player has beautiful sensitivity over the minutiae of their tone, that the intonation, the timing, and the timbre are just so. Just as in the case of instrumental playing, the speaker is responsible for their lyricism — and therefore deserving of approbation for it too.

One might think that a skilled instrumental musician is deserving of approbation for the beauty of their playing only because they have consciously honed their craft through practice. Everyday cases of lyrical speech are presumably different: they are not, in the same way, the product of consciously crafted musical skill. However, while the capacity to speak with musical sensitivity may not be cultivated consciously, as perhaps the capacity to play the bass guitar with similar musicality might be, it is nonetheless a capacity that the musical speaker deserves credit for. It is, in this respect, unlike the inadvertent, pleasing qualities of their snoring, or, more pointedly, the equally inadvertent qualities of their youthfulness or slimness as those qualities can be heard in their vocal tones.

My claim here is that lyricism is a quality that can be attributed to a person in the same way that a virtue or vice can be attributed to a person, as opposed to the way that qualities of someone’s snoring cannot be attributed to them, as a person. Whether, and on what grounds, virtues and vices can be properly attributed to their bearers, as people, is a fraught topic in the philosophy of action — one which I could not hope to settle here. It will thus have to suffice for me to take it as an assumption that this distinction holds up between the attributability of virtues and, for instance, inadvertent qualities.²²

This description of lyricism will have served its purpose if it has at least summoned to mind the idea that one might be sensitive to a distinctly aesthetic dimension of another’s speech for which they as the speaker are singularly responsible. If these remarks about lyricism are approximately true to experience, then they suggest that more than being conceptually possible, genuinely aesthetic interpersonal experience is a familiar feature of verbal exchange. From here, it is possible to extend the sketch to show that some aesthetic evaluations of others’ speech might ground interpersonal valuing-attitudes. The question is whether lyricism could either constitute or represent a kind of personal value of the speaker — i.e., a ground for personal liking.

I think that lyrical speaking can indeed constitute evidence of the personal value of the speaker. To see this, note that lyricism is a quality that is similar to moral virtues such as patience and charms such as wittiness, which may not be moral virtues but which do seem to provide reason to like their bearer. That is, lyrical speech — an aesthetically pleasurable thing to perceive — has the effect of making the speaker an easier, more enjoyable person to be around. It is a trait that facilitates smooth and pleasant social relations. It is legitimate to value someone for their wit because that is a trait that can ‘introduce levity into difficult times’ (Abramson & Leite, 2011, p. 694) and can intensify the enjoyableness of good times. Wit is a likeable trait in a person because it is an enjoyable trait in a person and, specifically, one attributable to them as a person. Likewise, lyricism, which brings musical life to conversation, is an enjoyable trait in a person and so just as likeable as wit, and for the same reason. To complete the analysis of lyrical speech, I would like to allay a pair of objections by briefly sketching an example. The objections are that both of the challenges discussed in the earlier sections of this paper can re-emerge in judgements of lyrical speech. The first challenge discussed above — that aesthetic judgement of other people is clouded by prejudice — could re-emerge in judgements of lyricism, for instance, if that concept were to be associated with ‘speaking well,’ where that category itself relies on prejudices that favour fluent, educated, able-bodied native speakers. The second challenge — that a person’s aesthetic qualities are irrelevant to their worthiness for personal liking — appears to arise again if lyricism is a quality that can be manufactured by skilled rhetoricians to manipulatively induce the personal liking of their audience. If lyricism is a quality like ‘speaking well’ that can be learnt and deployed tactically in this way, then again it seems as though it is an aesthetic quality that is irrelevant to the question of whether someone is worthy of personal liking.

However, lyricism is not a matter of ‘speaking well’ in the conventional sense. Imagine an example of an elderly person who speaks with her friends in their mother tongue, which to her is a second language learnt in adulthood. Her speech is faltering, strewn with grammatical errors and inefficient phraseology. On any conventional account of ‘speaking well,’ she is failing in this regard. And yet her speech is also full of dry humour and playful jokes at the expense of her friends, jokes which are themselves jokingly disguised by her lack of fluency as she maintains an ambiguity over the extent to which any hesitation is deliberate and mocking. She also speaks a lot, filling much of their meetings with the sound of her voice and her familiar-though-foreign accent. Perhaps the constancy of her speech is itself an expression of a vulnerability that she allows her friends to hear — namely, the fact that she needs and cherishes their company. Let’s suppose also that the timbre of her voice, far from bearing dulcet tones, is gravelly and slips uneasily between registers of pitch. All of the speaker’s endearing qualities — her mirth, her affection, her sheer vulnerability — are embodied in the music of her speech which is itself an object of aesthetic evaluation.

I hope that it is intuitive that to her friends, merely listening to the speech of the elderly woman could bring disinterested pleasure. Her speech is paradigmatically lyrical despite being in several ways contrary to the archetype of ‘speaking well’ that might be practiced by a rhetorically skilled politician. That already shows that whatever lyricism is, it is not just a category reliant on the prejudices that inform the notion of ‘speaking well.’ So the first challenge has been met: it is possible to positively appraise the lyricism of speech in a way that is not reliant on unjustifiable prejudice. This does not mean that judgements of lyricism cannot be prejudiced. They surely can, and examples are at hand already: to deem the polished politician’s speech to be lyrical just on the basis that they were ‘speaking well’ would be prejudiced, as would be the judgement that the elderly woman’s speech lacked lyricism just on the basis that she was not ‘speaking well.’ Nonetheless, by invoking the hopefully familiar, everyday acquaintance that one might have with the loveworthy qualities of close friends, and the way such acquaintance can be mediated by the beautiful expression of such qualities through such aesthetic media as lyrical speech, the example serves to illustrate that unprejudiced judgements of beauty per se are not only conceptually possible but also familiar in the course of ordinary life.

The example also goes some way to addressing the second worry, because it illustrates the possibility that the aesthetically good qualities of speech could be connected in the right way to likeable qualities in a person. This does not rule out the possibility that listeners could be deceived by a skilful speaker into liking them. But that is a real possibility — indeed a real phenomenon — that is compatible with judgements of personal liking sometimes being legitimately grounded in the aesthetic qualities of a person’s speech.

6. Conclusion: The Vindicatory Account

I began this discussion by noting that our aesthetic experiences of other people, and the interpersonal feelings that are generated by those experiences, are permeated by prejudices and also shaped by a variety of factors — such as hormone levels and the arrangement of facial features — that seem to have little to do with how worthy people are for those feelings that arise in interpersonal interaction. This raised the general question of whether it can ever be legitimate to like a person on the basis of their aesthetic qualities. It is now possible to summarise my argument for thinking that, at least in the auditory domain, it sometimes can.

1. Interpersonal aesthetic experience commonly involves a bundle of three types of judgement: of attractiveness, of “beauty,” and of beauty per se.

2. Many of the prejudices that distort interpersonal aesthetic evaluation are associated not with judgements of beauty, but with judgements of attractiveness and of “beauty.”

3. Therefore, judgements of beauty per se are less distorted by prejudice than interpersonal aesthetic experience in general (from 1 and 2).

4. Personal liking involves an attribution of final value to the other person, as a person, for which they are responsible.

5. Of the types of judgement commonly involved in interpersonal aesthetic experience, only judgements of beauty attribute final value to the other person.

6. Therefore, of the types of judgement commonly involved in interpersonal aesthetic experience, only judgements of beauty could be relevant grounds for personal liking (from 1, 4, and 5).

7. Some beautiful qualities, such as the lyricism of a person’s speech, are finally valuable qualities of the bearer, as a person, for which they are responsible.

8. Therefore, in response to qualities such as lyrical speech, judgements of beauty per se can provide a sound basis for personal liking — thus, not all aesthetic bases for personal liking succumb to the prejudice or irrelevance objections (from 1, 3, 6, and 7).

This argument establishes that well-founded attitudes of personal liking based on appreciation of others’ aesthetic qualities are conceptually possible. Beyond that, the way that the likeable characters of friends and acquaintances can be expressed in the pleasing sounds of their voices (indeed, in the very pleasingness of the sounds of their voices) is something that will be familiar to many people from the course of ordinary life. Thus, the picture I am presenting is not of a mere conceptual possibility, but of a form of aesthetic attunement to other people’s likeable characters that evades the charges of prejudice and irrelevance and which is within reach of our everyday interactions.

However, my account has not said much about what all of us, as judges of others’ voices, should do with the erroneous factors that cloud our judgement. Without saying much, I have nonetheless identified the rudiments of what might be called a politics of listening.²³ Namely, in the absence of any further, more extensive vindicatory stories, it seems that the only aspects of another’s voice that can genuinely be beautiful, and in a way that can license valuing them, are the musical qualities that are attributable to them as a person. Given this, we must unlearn and disavow all other positive or negative judgements of another’s voice as having anything to do with their beauty — as far as this is possible. Furthermore, a politics of listening will separate judgements of attractiveness, and judgements of conformity with beauty norms, from judgements of beauty. This facilitates the political enterprise of subjecting auditory beauty norms to critical scrutiny, and developing a clearer understanding of how they might operate to create and perpetuate systems of prejudice, control, and injustice. Drawing these distinctions also facilitates clear-headed scrutiny of one’s own intuitions about attractiveness. Admittedly, this leaves unanswered the difficult question of how best to cultivate those intuitions.^{24, 25}