Where scholarship meets technological innovation . . .
Commentary
María José Alcaraz León
Universidad de Murcia, ES
León, María José Alcaraz. 2025. “Comments on Expressiveness: Perception and Emotions in the Experience of Expressive Objects.”
Journal of Philosophy of Emotion 7, no. 1: 13-19. https://doi.org/10.33497/2025.summer.3.
Abstract: In this commentary, I address two aspects of Benenti’s approach to the problem of expressive experience. First, I examine her characterization of response-dependent account of expressive properties, and I contend that her concern with the possible relativistic consequences of this approach could be appeased if she would have taken into account some versions of this account à la McDowell. Secondly, I question the need to ground expressive experience in the presence of inherently expressive features of the music eliciting expressive experiences. After a particular understanding of the possible representational contents afforded by the duck-rabbit figure, I propose that the multiple representational capacity ascribable to the figure does not require that the features grounding those representational contents possess any inherent representational character. Similarly, I contend, we can make sense of the normative character of expressive experiences without endorsing a particular commitment with the existence of inherently expressive features grounding those experiences.
Keywords: response-dependent properties, relativism, McDowell, expressive properties
Marta Benenti’s book on the phenomenon of expressive experience offers not only a thorough revision of some of the most important positions and discussions about this phenomenon, but also a serious attempt at finding a middle point between what she rightly recognizes as the two poles of the phenomenon.
As she puts it (p. 86) “on the one hand, accounts of the content of expressive experience must do justice to those constraints that reality puts on it. Expressive experience is not a sort of daydreaming or phantasy where one is free to ascribe whatever expression to whatever object; rather, it appears to be related to objective features of
[page 13]
objects. This suggests that some adequacy conditions are required. On the other hand, such accounts must do justice to the relational and therefore “subjective” nature of the experience, that is to the fact that important differences may subsist between two expressive experiences of the same object on behalf of different subjects.”
Benenti’s approach is meant to make justice to both poles and to find the right anchor for the phenomenon. In what follows, I would like to address some aspects of her proposal that I think can be shown to be disputable. The aspects that I would like to discuss are:
Benenti’s characterization of response-dependent accounts of expressive properties in music.
The need to anchor expressive experience in certain low-level inherently expressive properties.
In what follows I will try to comment on these two points.
BENENTI’S CHARACTERIZATION OF RESPONSE-DEPENDENCE
ACCOUNTS OF EXPRESSIVE PROPERTIES IN MUSIC
Benenti’s proposal partly derives from her analysis of some of the problems that other alternative accounts have. In fact, she thinks that a proper account of expressive experience is to be found midway between causal approaches and response-dependent ones.
On the one hand, causal approaches succeed in establishing a strong dependency connection between non-expressive perceptual features and expressive properties, but they fail to make that relation normative or justificatory. Since the relation is supposed to be merely causal, it falls sort of providing a justification for the particular expressive character of the experience. Under this view, expressive experience is, while explained in causal terms, not properly grounded upon the perceptual character of the object. What, in a particular instance, causes the expressive experience in the listener might not be subject to normative assessment.
Furthermore, the lack of a normative connection between the (perceptual) properties that are allegedly causally responsible for a work’s expressive character and the expressive character itself leads to the problem known as the “heresy of separable experience” (Budd 1985). This problem, initially formulated by Malcolm Budd (1985), states that any explanation of the expressive character of a particular work that involves its expressiveness “can be fully characterized [and valued] without reference to the nature of the work itself” (Budd 1985, p. 123) is flawed. Thus, any account should respect the intuition that a work’s expressive character -and the corresponding expressive experience— must be anchored in the perceivable properties of the work. Otherwise, the very same expressive experience could be caused by other, alternative, purely causal mechanisms—such as a drug. Therefore, any proper account of expressive experience should accommodate the normative character of the relationship between the work’s perceptual, non-expressive properties and its expressive character. Causal accounts cannot do so.
On the other hand, response-dependent accounts typically explain expressive experience in terms of the response that informed subjects will have to certain non-expressive perceivable features. However, Benenti’s analysis of the accounts -such as Noordhof (2002, 2008)—which are supposed to give credit to this position concludes with the remark that a purely response-dependent account is subject to lead to relativistic results—for different responses will constitute different expressive experiences. As she puts it: “no adequacy conditions can be found for expressive experience. (. . .) the same landscape could look desolate to somebody and cheerful to somebody else, and these two ascriptions could never be verified by actual properties of the landscape, because the experience would entirely consist in the responses the two subjects would give to the same causal stimulus. At best, causal conditions may explain the rise of certain expressive experiences whose content would not find in the
[page 14]
world any criterion to establish its correctness.” And she concludes “this view does justice to the subjective character of expressive experience. However, it is not in a position to capture its more objective side, or at least, our tendency to appeal to the way objects perceptually look, in order to justify our ascriptions" (Benenti 2020, 91).
Thus, according to Benenti, response-dependent accounts are ill-equipped to fully explain the veridical character of expressive experience, its being somehow responsive to the (right) properties of the work. They fail because there seems to be no independent manner to assess whether a particular response is well, or ill, grounded in the work’s perceptual properties.
Before addressing Benenti’s own proposal, I would like to focus on her understanding of the response-dependent approach for, I think, the kind of account that she presents as the standard response-dependent approach does not correspond to some well-established versions of this view, such as McDowell’s (1985, 1998), that are better equipped against the relativistic charge. This fact may, in turn, lead her to endorse some strategies that may be unnecessary or misguided.
Some reasons speak in favor of considering that a view about secondary properties and values such as McDowell’s could have been relevant for Benenti’s overall approach to the phenomenon of expressive experience. First, the way she characterizes expressive experience and her explicit commitment to the idea that expressive properties are in a non-reducible manner grounded in the perceptual non-expressive properties of the objects that instantiates them seems to parallel the understanding of aesthetic value fostered by views such as McDowell’s. Within the latter account, values cannot be reduced to non-evaluative properties in the sense that their identification is not independent from the experience that they typically trigger. To judge that something is, for example, funny, is not independent from finding it amusing or responding to the object with amusement. The amused response is constitutive of the judgment. Secondly, Benenti’s acceptance of the similar status of expressive and aesthetic properties -she actually endorses the view that expressive properties are a subset of aesthetic properties-, gives further reason to reconsider the response-dependent character of expressive properties in light of those views. An important consequence of having explored the response-dependent character of expressive properties after these views about value would be that the threat of relativism would be less warranted; and, hence, that the need for identifying a response independent source of the expressive character of a work would be less motivated. That is, only under a certain reading of the response-dependent condition will the problem of relativism be a serious problem.
THE NEED TO ANCHOR EXPRESSIVE EXPERIENCE
IN CERTAIN LOW-LEVEL INHERENTLY EXPRESSIVE PROPERTIES
As I see it, Benenti’s partial characterization of the response-dependent approach might lead her to think that this approach fails to answer the requirements of a proper account of expressive experience. In her view, response-dependent accounts can only account for the subjective pole of expressive experience. This, in turn, leads her to think that, unless we can find certain properties that are inherently affective, we would not be able to account for the proper normative relation between the properties of the work and the expressive experience it provokes. In her view, without acknowledging some external constraint to the response-dependent criterion, there is no guarantee that expressive experience bears the right sort of relation to the non-expressive perceptual properties that grounds it. So, we need to find something that, to some extent, prevents us from experiencing the very same object as being expressive of many, and maybe inconsistent, emotions. Something that guarantees that certain responses are right, while others fail to capture the true expressive character of a given object.
[page 15]
In order to fill in this requirement, Benenti’s approach exploits two different sources: one is a version of the resemblance view (which she claims to account for the weeping willow and the expressive gesture cases); and the other is a view according to which certain perceptual features are inherent affective (which she thinks can explain our expressive experience of single features such as colors or the minor chord case). In each case, the presence of the relevant resemblance or the inherent affective feature would guarantee one’s expressive experience to be rightly anchored in the object’s non-expressive features.
In what follows I would like to explore a possible way to undermine the need for this move. While I think Benenti is right that not any expressive experience can be correct of a given object and that the object’s properties partly determine whether a particular expressive experience is proper, I think her appeal to some response-independent features--be it in terms of resemblances or in terms of inherent affective features- is unnecessary. Moreover, this appeal is, I think, at odds with the often--invoked idea in aesthetics that there are no aesthetic principles governing particular ascriptions of aesthetic value or properties (Sibley 2001). If I have not understood Benenti’s proposal wrongly, her view can be seen as implying the existence of certain principles governing expressive attributions and expressive experience. In particular, her proposal that underlying resemblances and inherently affective features are what justify our expressive experiences can be understood in terms of these features to function as pro tanto reasons for a particular expressive character rather than another. This understanding of her view is, I believe, motivated by some of the remarks at the end of her book.
After establishing (Benenti 2020, 166) that “certain low-level features are originally experienced as expressive, as affectively loaded, in virtue of their perceptual structure (described as its rythmos” she claims that “the same perceptual feature is hardly experienced both as positive and negative, static and dynamic -at least in ideal conditions where it appears isolated from context” (Benenti 2020, 167). Two important things can be said about this paragraph. The first one is that Benenti is aware of the possible contextual factors that, over and above the alleged inherent affective character of the perceived feature, could contribute to the overall expressive experience. So, in a sense, she is not assuming that the normative force of these features is such that they sufficiently establish the particular expressive character of an object. But this acknowledgment leaves some room for the inherent affective feature to play some normative role in providing us with a way to tell whether a particular experience is correct or incorrect. So, in a sense, Benenti will be endorsing the existence of pro tanto reasons or weak expressive (aesthetic) principles. In this sense, her approach would depart from the often-assumed view that aesthetic normativity is compatible with the absence of aesthetic principles. As I will try to show later, there is no need to give up on this view about aesthetic normativity.
The other thing I think is worth commenting on is her claim that the same feature could not “be experienced both as positive and negative, static and dynamic” (167). I think that this sentence can be understood in two alternative ways each implying different commitments regarding the normative structure of expressive experience. One way in which this statement can be understood is as saying that the same perceptual feature cannot be experienced as grounding an expressive experience with a positive character and at the same time as grounding an expressive experience with a negative character. I think this reading is uncontroversial. I take it that even those who think that the same feature can ground different -even contrary- expressive experiences would claim that these alternative experiences cannot take place at the same time. But I think Benenti is stating something different. If I have not understood her wrongly, an inherently affective feature could not be correctly experienced as grounding an expressive experience with positive valence at one time and with a negative valence at another time (all things being equal). This claim is, I think, more controversial.
[page 16]
To show why it is unnecessary that there is something inherently affective in the low-level properties in order to explain the normative role they play in the validity of the expressive experience, I would like to draw an analogy with the Jastrow-like figures. Let’s take the duck-rabbit figure as an example of this phenomenon. Part of what makes it appropriate to see either the duck or the rabbit in the figure is that we can perceive the different parts of the drawing as parts of a duck-figure and as parts of a rabbit-figure. Also, as it is often pointed out, when one is seeing the duck in the figure, one cannot at the same time see the rabbit -albeit it is possible to see both alternatively. Those features that are invoked as parts of the duck and as parts of the rabbit are the same lines -the figure has not changed-, but it would be very odd to say in this case that the features are inherently representational of a duck -or of a rabbit, for that matter. What makes certain lines and dots depictive is not something inherently representational in the lines or the dots, but something that has to do with the kind of things one can say about the drawing and its parts when we try to explain our experience. This does not imply that their contribution to the experience is not explanatory— if we can see a duck’s head it is partly because there are certain curved lines and a dot in the middle—but their being explanatory of the representational experience of a duck does not require the existence of an inherent representational character in the features involved in the representation or a given resemblance relation between the lines of the drawing and ducks and rabbits. What makes them representational is that they make possible this or that particular visual experience. If we apply this way of understanding how certain features can contribute differently to the depictive content of a single figure to the case of expressive experience, we could say that the very same features can ground different alternative expressive experiences given that we can properly relate these experiences to the features we take to be responsible for them. For example, we could show that a particular expressive experience of a passage as sad would lose its sadness by changing the rhythm or that her sadness would somehow be lifted up by playing the passage with a different instrument. So that rhythm and that instrument’s timbre are, in these cases, partly responsible for the passage's expressiveness. This would be a way to show that the expressive experience is grounded upon certain perceptual features without assuming they possessed any inherent expressive or affective character. In short, we can give due credit to the way in which expressive experience is grounded in the perceivable aspects of the music without assuming these aspects must possess an inherent affective character or bear a resemblance relationship with the expressive profile of the emotion musically perceived.
I think this way of approaching the phenomenon also better justifies the way in which we argue about aesthetic experience and in particular about expressive attributions. It would be very odd that someone insisted on the inherent or intrinsic affective character of a singular feature—or on the alleged resemblance between a passage and an emotion—in order to justify that a particular musical passage has a particular expressive character of which we, as listeners, are not convinced of. Should we give up on our judgment if someone pointed out to us that a certain resemblance existed or that that chord has a particular inherent affective character? The standard way to motivate a particular expressive experience is by pointing to certain perceivable features that we think that, in that particular occasion, make for the music’s expressive character. But we do not need to endorse that those features will systematically have the same contribution to believe that, in this particular case, these are the features responsible for the expressive character of the work. In this sense, endorsing a particularist approach can better account for the normative relation we think exists between low-level properties and expressive features, without giving up on the idea that aesthetic normativity is compatible with the absence of aesthetic principles governing the application of aesthetic predicates. Benenti often reminds us that designers and certain composers—film music is often quite formulaic—know which features will trigger certain affective responses in the viewers and the listeners respectively. This seems to support her view that there are certain resemblance relations and certain inherently affective features that designers and composers are knowledgeable of. But it is also true that more explorative designers and composers often discover unexpected effects of features that, when presented in a particular way or in certain combinations, result in expressive experiences difficult to explain in terms of an alleged inherent affective character. As in the more conventional cases, the expressive experience afforded in these
[page 17]
cases is no less grounded in the perceptual properties of the objects, but they can hardly be expressive by instantiating the same principles that explained their more conventional counterparts.
A particularist approach to expressive experience grants its anchor in the properties of the object, avoiding thus the risk of relativism, without requiring a commitment to the existence of either inherently affective properties or resemblance relations. My contention is not that such properties do not exist—maybe they do. My contention is that even If they did exist, they would not perform the kind of justificatory role that they are supposed to perform according to Benenti’s account.
[page 18]
References
Benenti, Marta. 2020. Expressiveness. Perception and Emotions in the Experience of Expressive Objects. Berlin, GER/Boston, MA: Walter de Gruyter GmbH.
McDowell, John. 1985. “Values and Secondary Qualities.” In Morality and Objectivity, edited by Ted Honderich, 110-129. London: Routledge. Reprinted in McDowell, John. 1998. Mind, Value, and Reality, 131-150. Cambridge, MA: Harvard University Press.
McDowell, John. 1998. "Aesthetic Value, Objectivity, and the Fabric of the World." In McDowell, John. Mind, Value, and Reality, 112-130. Cambridge, MA: Harvard University Press.
Noordhof, Paul. 2002. “Imaginings Objects and Imaginings Experiences.” Mind and Language 17 (4): 426-455.
Noordhof, Paul. 2008. “Expressive Perception and Projective Imagining.” Mind and Language 23 (3): 329-358.
Sibley, Frank. 2001. “Particularity, Art, and Evaluation.” In Approach to Aesthetics. Collected Papers on Philosophical Aesthetics, edited by John Benson, Betty Redfern and Jeremy Roxbee Cox, 88-103. Oxford, UK: Clarendon Press.
Sibley, Frank. 2001. “General Criteria and Reasons in Aesthetics.” In Approach to Aesthetics. Collected Papers on Philosophical Aesthetics, edited by John Benson, Betty Redfern and Jeremy Roxbee Cox, 104-118. Oxford, UK: Clarendon Press.
[page 19]
María José Alcaraz León © 2025
Author email: alcaraz.mariajose[at]gmail[dot]com
