Commentary

Comments on Expressivenesss: Perception and Emotions

 in the Experience of Expressive Objects

Copenhaver, Rebecca and Jay Odenbaugh. 2025. “Comments on Expressiveness: Perception and Emotions in the Experience of Expressive Objects.”

 Journal of Philosophy of Emotion 7, no. 1: 6-12. https://doi.org/10.33497/2025.summer.2. 

Abstract: In this essay, we first summarize Marta Benenti’s views articulated in her book Expressiveness. Second, we situate Benenti’s views amongst other views regarding expressiveness including projectivism, arousalism, persona theory, and contour theory. Third, we raise questions about her views regarding representationalism, cognitive permeation, over-intellectualizing expressiveness, and to what extent expressiveness is a unified phenomena.

Keywords: affect, emotion, expression, representationalism, cognitive permeation, perception 

Marta Benenti’s Expressiveness is one of the most sophisticated accounts of expressiveness to date. It brings together the philosophy of art and the philosophy of mind, providing many novel insights. In this commentary, we summarize our understanding of Benenti’s major claims and present some questions and criticisms. Much is shared between our views–for example, we are all representationalists (or intentionalists) about experience–but we differ in interesting ways. We begin by summarizing Benenti’s views as we understand them.

SUMMARY

The phenomena we are both trying to understand is expressiveness. One puzzle in the philosophy of art and philosophy of mind concerns expression. It seems that both animate and inanimate objects can express emotions. But only the former is capable of having emotions. So, what is it for something to express an emotion (or any other mental state for that matter)? Marta Benenti considers the thesis that inanimate objects express emotion. 

[page 6]

Call this the of inanimate object’s expressiveness (IOE) thesis. Two arguments lead to the denial of IOE (Benenti 2020, 8). 

P1 Emotions are psychological states. 

P2 Only animate beings possess psychological states.

C1 Only animate beings can feel emotion.

C1 Only animate beings can feel emotions.

P3 To express an emotion implies that one feels that emotion. 

C2 Only animate beings can express emotions. 

The view that inanimate objects can express emotions is often referred to as the “pathetic fallacy.” On this view, when we say a piece of music is sad, for example, we are guilty of a false animism. Following Alan Tormey (1971) and Peter Kivy (1989), Benenti distinguishes between expression and expressiveness. One expresses an emotion only if one feels that emotion. However, one’s face can be expressive of an emotion that one does not feel; for example, in the way that a St. Bernard’s face is expressive of sadness. The face of the St. Bernard has the “characteristic look” of that emotion. With the distinction between expression and expressiveness in hand, one may concede that inanimate objects cannot express emotions while admitting that inanimate things and features may yet be expressive of various emotions. 

Benenti’s aim is to explain what expressiveness is. Benenti provides several desiderata that any reasonable theory of expressive experience must satisfy (Benenti 2020, Ch. 1).[1] Such a theory must correctly account for (a) the phenomenal character of expressive experience, (b) the intentional content of expressive experience, (c) the role of the emotions in expressive experience, and (d) empirical results relevant to expressive experience. Using these desiderata, Benenti evaluates a variety of theories of expressive experience. 

Benenti considers four initial approaches–projectivism, arousalism, persona theory, and contour theory. In each case, Benenti argues that the violates one or more of the theoretical desiderata. Consider, for example, projectivism. Often, we feel emotions and find correspondences between emotions and things in the environment (Wollheim 1993). Emotions can permeate perceptual experience. In fact, projecting emotions onto our environment can be a therapeutic intervention. However, projectivism has difficulty accounting for cases in which we have expressive experience without currently experiencing any emotions. Projectivism also doesn’t explain the supposed correspondences between emotions one the one hand and things and features on the other–what exactly are these correspondence relations ? 

Another account of expressive experience is arousalism. On this view, we have an expressive experience when a relevant emotion is caused in us under normal conditions (Matravers 2001). However, as before, there are obvious cases in which we have an expressive experience in the absence of emotion. Arousalism is also subject to the “heresy of separable experience”: the contents and causes of experience are distinct. By committing this fallacy, arousalism misdescribes the phenomenal character and misidentifies the content of expressive experience.[2] 

Another approach to expressive experience is persona theory. According to persona theory, an expressive experience consists in imagining of an object that it feels some emotion (Levinson 2006). We might do so through affective empathy or emotional contagion, but in doing so we personify things by attributing emotions to them. The main worry for persona theory is again that the view gets the phenomenal character of expressive experience wrong. In expressive experience, we do not always attribute emotions, we do not always imaginatively see or hear objects as having emotions. Indeed, emotional attribution can be cognitively taxing. In order to 

[page 7]

imaginatively hear a piece of music as mournful, I must possess and deploy the concept MOURNFULNESS in my experience. As with other views, persona theory seems to get the experience itself wrong. 

Finally, there is contour theory, according to which expressive experience consists in perceiving similarities between emotional expressions and features of objects (Lopes 2005). We do sometimes sub-personally detect similarities of this sort. However, again, such similarities are not always part of the character or content of expressive experience. Thus, as stated, none of these theories satisfy Benenti’s desiderata for a theory of expressive experience. 

Like us, Benenti is committed to representationalism (or intentionalism). According to representationalism, experiences have adequacy (or better yet, accuracy) conditions: ways the world would have to be for the experience to be accurate. The phenomenal character of an experience is determined by its intentional content: what it is like for the subject to have an experience is a matter of how her experience represents the world as being.[3] For example, expressive experiences can misrepresent properties of our environment. If I hear Pharrell William’s song “Happy” as sad, my experience is misrepresenting a feature of the song. 

Benenti argues that expressive experience is perceptual or perception-like. It often occurs quickly, without inference, and little voluntary control.[4] Expressive experience is also vivid (we might characterize its vivacity by saying that it has analog rather than digital content). Three phenomenal qualities are often associated with expressive experience: temporality, movement, and complexity. Temporality and movement are important for expressive musical experience but also for the visual arts. What seems to matter across the arts is “intrinsic dynamism” (Benenti 2020, 76). However, intrinsic dynamism faces problems when it comes to simple expressive features such as minor chords that do not appear to be dynamic.[5]

Benenti argues that there are three types of expressive experience that can be captured by paradigm cases: the weeping willow, the musical gesture, and the minor chord (Benenti 2020, Ch. 3). In the case of the weeping willow, we see it as sad because we detect an isomorphism between the willow and the posture of a sad person, which results in an expressive experience as of sadness. The isomorphism need not be phenomenally manifest. 

In the case of a musical gesture, we recognize dynamic properties that are shared, for example, between parts of musical works and the human voice. We often recognize such groupings of properties sub-personally. According to Benenti, recognitional expressive experiences are weakly cognitively permeated by metaphorical concepts.[6] In such cases, recognitional concepts do not merely causally influence the perceptual processing of isomorphisms, but render experiences themselves at least partially conceptualized. Such experiences also involve non-conceptual content representing the low-level features involved in the isomorphisms. 

According to Benenti, emotions are complex entities constructed from core affect along the dimensions of arousal and valence. Interestingly, emotions qua experiences and the properties they represent have affective features. That is, both may be described correctly through arousal and valence. As a result, Benenti provide a way to have accuracy conditions for emotions. Emotion concepts are cluster concepts. We deploy concepts like SADNESS by attributing sadness to people, faces, and even inanimate objects. We do so by recognizing correctly, or when correctly are unavailable, the terms such as ‘sadness’ acquire a “secondary meaning.” Secondary meanings allow us to attribute ‘sadness’ and other emotions to simple features such as chords, colors, lines, and the like, thus accounting for the case of the minor chord.

Benenti provides us with a rich and complex theory of expressive experience. We now turn to questions and criticisms.

[page 8]

QUESTIONS AND CRITICISMS

We have a variety of questions for Benenti, but it is worth covering where we are in agreement. First, we also find the traditional views, including projectivism, arousalism, persona, and contour theories importantly incomplete or wrong. Second, we are also representationalists (or intentionalists) (Copenhaver and Odenbaugh 2020).[7] Third, we also agree that the philosophy of mind provides essential tools for understanding expression and expressiveness. 

Our first concern is that Benenti’s views conflict with her representationalism. Recall that according to representationalism, experiences have accuracy conditions, and that phenomenal character is determined by intentional content. On Benenti’s view, expressive experiences are weakly cognitively permeated by metaphorically-applied emotion concepts that often (though not always) represents some isomorphism. There is no problem with tracking non-affective properties using affective concepts. However, if a concept is applied metaphorically, then it represents the object as having a feature it strictly speaking does not have. As such, Benenti’s view looks more like a form of projectivism or error theory. She might reply that such cases are instances of secondary meaning. But this raises the question of what secondary meanings are.

A second concern also arises from Benenti’s appeal to cognitive permeation. Emotional concepts are applied via cognitive permeation, which is a top-down process not internal to perceptual processes themselves. However, in the philosophy of perception, there are liberal views that provide alternatives to cognitive permeation (Crane 1992, Siegel 2011). In particular, one may learn to make sophisticated perceptual discriminations of kind, causal, moral, and aesthetic properties in experiences that have non-conceptual content. In short, we worry that Benenti has “conceptualized,” and thus intellectualized, expressive experience without considering other alternatives consistent with her representationalism. 

A third set of issues highlights our worries about intellectualizing expressive experience. If expressive experience requires conceptualization, then non-human animals and young children cannot have expressive experience given they lack concepts. But it is reasonable to think that animals and young children can perceive a dark forest as scary or a Batman doll as angry. Additionally, sometimes the contents of expressive experience are more fine-grained than the concepts we possess. When we listen to Miles Davis’ “Kind of Blue”, we do not hear it as merely sad but as having very specific melancholic qualities. As with our experience of colors, our expressive experience is often much richer than our concepts. Benenti might respond that expressive experiences have conceptual and non-conceptual content (she in fact does). However, so long as expressive experience remains cognitively permeated we are left with the worries we have regarding metaphors, projectivism, and error theory mentioned above. 

A fourth and final question concerns Benenti’s focus on experience. Many philosophers of art have been interested in understanding what expressive properties are and whether inanimate objects can possess expressive properties. Benenti shifts the debate to expressive experience. She justifies this shift as follows: 

For reasons that will become clearer below, most existing accounts are interested in explaining what I will hereafter refer to as “expressive experience”, that is the experience of objects – natural objects as well as artefacts – as instantiating expressive qualities. The nature of these qualities is accordingly defined on the basis of the sort of experience that mobilises them. Instead of asking “what are expressive qualities (or properties)?”, the question “how do we experience expressive qualities (or properties)?” will be the main point of contention in the debate that is about to follow. The desiderata to be satisfied will

[page 9]

therefore concern a theory of expressive experience rather than a metaphysical assessment of expressive properties. (Benenti 2020, 20-1)

However, Benenti accepts a theory of emotion according to which affective properties such as arousal and valence are attributable to experiences and to objects independent of experience both. It looks like we are left once again with the puzzle of how a single property kind–affective properties–can be attributed to things (experiences and objects) that are so ontologically heterogenous. It is not clear that we can address this puzzle if we attend only to expressive experience and not also to expressive qualities. 

CONCLUSION 

Marta Benenti’s Expressiveness is an extremely rich book. It provides new tools and theories to address the difficult topics of expression and expressiveness. We hope philosophers of art and philosophers of mind will read and learn much from it. 

[page 10]

Notes

[1] One might try to provide an account of expressive properties. However, Benenti suggests that many are trying to account for expressive experience instead. In fact, one might argue that if expressive properties are response-dependent then we can explain them only by appeal to expressive experience. 

[2] Of course, sometimes the what it is like to have an expressive experience includes felt emotion. But the objection here is that it can occur without it. 

[3] As Benenti states the view, this is actually too strong. Though some representationalists are reductionists, some are not. They insist that the phenomenal character of an experience supervenes on its intentional content even if it is not reducible to it. 

[4] It is worth noting that inference can occur quickly as well. So, we might want to distinguish perception and inference by other means. 

[5] Peter Kivy (1989) argues that minor chords are expressive of sadness because of cultural convention. Whether or not this is right, the expressive qualities of minor chords do not seem to be relational in the same way that musical gestures are.

[6]  By ‘cognitive permeation’ we intend the same meaning as the more commonly used phrase ‘cognitive penetration,’ which we find to be…unsubtle.

[7] We include the term intentionalism here because not all content is representational. It can also be directive (i.e., imperatival) rather than descriptive. One interesting potential example of an experience with directive content is pain (Klein 2015). When an experience has directive content, it has satisfaction conditions rather than accuracy conditions.

[page 11]

References

1. Benenti, Marta. 2020. Expressiveness. Perception and Emotions in the Experience of Expressive Objects. Berlin, GER/Boston, MA: Walter de Gruyter GmbH. 

2. Copenhaver, Rebecca and Odenbaugh, Jay. 2020. “Experiencing Emotions: Aesthetics, Representationalism, and Expression.” In The Epistemology of Non-Visual Perception, edited by Berit Brossard and Dimitria Gatzia, pp. 213-35, Oxford University Press. 

3. Crane, Tim, ed. 1992. The Contents of Experience: Essays on Perception. Cambridge, UK: Cambridge University Press.

4. Kivy, Peter. 1989. Sound sentiment: An Essay on the Musical Emotions, Including the Complete Text of the Corded Shell. Philadelphia, PA: Temple University Press.

5. Levinson, Jerrold. 2006. “Musical Expressiveness as Hearability-as-expression.” In Contemporary Debates in Aesthetics and the Philosophy of Art, edited by Matthew Kieran, pp. 192-204. Oxford, UK: Blackwell.

6. Klein, Colin. 2015. What the Body Commands: The Imperative Theory of Pain. Cambridge, MA: MIT Press.

7. Lopes, Dominic. 2005. Sight and Sensibility: Evaluating Pictures. Oxford, UK: Clarendon Press.

8. Matravers, Derek. 2001. Art and Emotion. Oxford, UK: Oxford University Press.

9. Siegel, Susanna. The Contents of Visual Experience. Oxford, UK: Oxford University Press.

10. Tormey, Alan. 1971. The Concept of Expression. Princeton, NJ: Princeton University Press.

11. Wollheim, Richard. 1993. The Mind and Its Depths. Cambridge, MA: Harvard University Press.

[page 12]

Rebecca Copenhaver & Jay Odenbaugh © 2025 

Author email: becko.copenhaver[at]gmail[dot]com; jay[at]lclark[dot]edu