Marta Benenti
Spring 2023
10.7413/18258630134
It. Espressività; Fr. Expressivité; Germ. Ausdruckskraft; Span. Expresividad.
The concept of expressiveness refers to the capacity someone or something has to express, i.e., to manifest a psychological state, emotions in particular. Expressiveness characterizes human emotional manifestations, but also objects’ and – especially – artworks’ appearances: a piece of music can be described as melancholy, a landscape as desolate, a shade of color as cheerful. The phenomenon of expressiveness has been widely studied by both philosophy and psychology (Parovel 2012). Most philosophical theories of expressiveness have been put forward in the domains of philosophy of the arts, particularly philosophy of music, philosophy of aesthetic experience, and philosophy of mind - including philosophy of emotions, of perception, and of imagination (Benenti 2020).
Philosophers have so far addressed the topic of expressiveness on three intertwined levels:
(i) A phenomenological level, asking what it is like to experience something as expressive of an affective state. Philosophers disagree on whether the phenomenology of “expressive experiences” is akin to that of standard perceptual experiences, or it is rather sui generis.
(ii) An epistemological level, assessing our mental (and bodily) access to expressive properties. To be debated is the role played by mental states like imagination, memory, and emotions in making us experience objects as expressive, and thereby providing us with information about them.
(iii) A metaphysical level, assessing the nature of expressive properties, and especially their link to emotions and to aesthetic properties.
The origins of the debate about expressiveness trace back – at least – to Theodore Lipps’ Ästhetik (1903), conceiving of expressiveness as a matter of empathy (Einfühlung); Wassily Kandinsky’s Concerning the Spiritual in Art (1911), accounting for expressiveness as the outcome of natural emotional triggering; and Rudolf Arnheim’s Toward a Psychology of Art (1966), focusing on the perceptual nature of expressive Gestalten or “tertiary qualities”. Notably, the philosophical discussion has been more recently revived in analytic philosophy.
Importantly, although they share some points of intersection, the topic of expressiveness and that of expression do not overlap. For this reason, the present entry deliberately sets aside not only philosophical and psychological discussions about human facial and bodily expression, but also prominent approaches to expression as a metaphysical notion, such as Gilles Deleuze’s reflections on Spinoza’s Ethics (Deleuze 1968).
1. An Apparent Paradox. Expressiveness can be introduced as a paradox (or inconsistent triad) with the following structure:
1) In order to be expressed, an emotion must be felt by someone
2) Inanimate objects do not feel emotions
3) Inanimate objects express emotions
Notably, one strategy to avoid the paradox has been introduced by Alan Tormey in his 1971 The Concept of Expression. Countering “expressivist” theories of art (defended by, e.g., R.G. Collingwood and John Dewey) according to which the expressive qualities of an artwork are the direct outcome of the artist’s emotional expression guiding the creative process, Tormey distinguished expressiveness from expression. While the former should be understood as the instantiation, in a look, of detectable expressive qualities (the look will be expressive of a certain state), the latter amounts to the actual manifestation of felt emotional states (the person or the object will express those states). This explains why, for example, we can talk of more or less expressive expressions (Robinson 2007). In light of this distinction, proposition (3) of the paradox can thus be amended by saying that: inanimate objects do not express emotions, but rather, are expressive of emotions.
Although Tormey’s distinction has been admittedly taken on by most subsequent theories of expressiveness, residuals of expressivism can still be found in many of them. That is, although formally acknowledging that it is one thing to express an affective state and another thing to be expressive of that affective state, philosophers keep struggling to account for the allegedly strong phenomenological, epistemological, or metaphysical relation between expression and expressiveness. In the following, prominent accounts of the phenomenon of expressiveness are introduced, along with the way in which each of them – explicitly or implicitly – deals with the paradox. Importantly, all listed approaches to expressiveness have problems and face objections that cannot be explored here. Yet, each of them seizes relevant aspects of a complex and multifaceted phenomenon that deserves further philosophical and psychological inquiry.
2. Pathetic Fallacy. Coined by art critic John Ruskin, the term “pathetic fallacy” designates the illusory experience a sentient subject has when, misled by uncontrolled emotions, they ascribe affective states to inanimate natural objects or artworks. Expressiveness thus consists in the result of an irrational and therefore erroneous attitude towards the world, one that has left its traces in infants, primitive societies, magicians, and poets (Santayana 1905). Accordingly, (3) can be denied and replaced by: (3a) We happen to misattribute emotions to objects when we are in the grip of our affective states.
3. Metaphor. That of a misattribution is a strong intuition underlying those theories of expressiveness according to which the ascription of emotions to objects - artworks in particular - must be explained in terms of metaphorical descriptions. In Languages of Art (1968), Nelson Goodman has famously defended a non-literalist account of expressiveness. Emotion labels such as “sad” or “cheerful” apply to objects only metaphorically, so that – it is claimed – expressive properties are metaphorically possessed by artworks. While Goodman’s view remains neutral as to the metaphysical nature of expressive properties, other metaphorical accounts take a more radical stance, arguing that the use of metaphorical descriptions is necessitated by the absence of any obvious metaphysical relation between inanimate objects and emotions (Zangwill 2001). Thus, according to metaphorical views, (3) can be denied and replaced by: (3b) We metaphorically apply emotion labels to objects because we lack literal descriptions.
4. Arousal. Another strategy to deal with the paradox appeals to causal psychological mechanisms of emotional arousal. Since emotions cannot be literally expressed by objects that do not feel them, arousal theorists claim that we ascribe to objects those emotions such objects elicit in us. In a nutshell, similarly to when we recognize others’ facial or bodily expressions as – say – sad, by feeling sadness or pity for them, listeners would describe a piece of music as “sad” in virtue of its capacity to trigger sadness (Matravers 1998; Robinson 2005). In this view, (3) can be denied and replaced by: (3c) We ascribe to objects those emotions (or emotions somehow consistent with those) they actually elicit in us.
5. Projection. Upholders of projectivist approaches to expressiveness elaborate on those associative mechanisms that make us recognise affective features in objects that cannot literally possess them. In a particularly sophisticated and openly Freudian proposal, Richard Wollheim (1993) argued that expressive properties are automatically ascribed to objects based on the emotions such objects elicited in us in the past. This projective process is held responsible for expressive ascriptions even in the absence of emotions felt by the subject. Other projectivist accounts attribute a key role to imagination as what makes us project psychological features onto inanimate objects (e.g. Noordhof 2008). According to projectivism, (3) can be denied and replaced by: (3d) We ascribe emotions to objects based on consolidated associative mechanisms that make us see those objects as apt for such ascriptions.
6. Persona Theory. Imagination is at the core of so-called Persona theory, originally put forward by Jerrold Levinson (1996), suggesting that ascriptions of emotions to artworks, and music in particular, result from an imaginative engagement that makes us experience those objects as if they were someone’s emotional expression, a phenomenon that partly resembles that of pareidolia. Such a character might take the shape of a composer, a musician, or a fictional entity. Accounts of expressiveness that mobilize imagination, typically rely on theories of simulation as the psychological process that allows one to empathize with others, thanks to the subpersonal mimicry of their expressive behaviors (e.g. Walton 1999; Currie 2011; Cochrane 2010). Accordingly, (3) can be denied and replaced by: (3e) We ascribe expressive features to objects that we imagine as the emotional expression of a fictional character.
7. Contour Theory. Focusing on the perceivable contour of music, Peter Kivy and Stephen Davies have put forward so-called Contour theory, holding that expressive properties of music are perceptual features of musical gestures, in the same way as notes, chords, pitches, rhythm, and so on (Kivy 1980; 2002; Davies 1994; 2011). According to Contour theory (that has been extended to objects other than music. See e.g. Lopes 2005), we directly hear musical expressiveness in virtue of the resemblances between musical contours and people’s expressive bodily and vocal manifestations. Thus, (3) can be denied and replaced by: (3e) We ascribe expressive features to objects as long as we perceptually detect resemblances between their perceivable patterns and typical human emotional expressions.
8. Atmospheres. One further approach to expressiveness stems from the research domain of “atmospherology” (see Böhme 1995 and Griffero 2014 for seminal accounts of atmospheres). By denying the purely inner nature of emotions, philosophers of atmospheres claim that there is no need to postulate any arousal, projective, nor imaginative process to account for expressive properties of objects. Rather, it is part of our concepts of emotions that they can be instantiated by inanimate objects as well as by sentient beings. Expressive features accordingly behave like atmospheres: they are both felt and perceived, and subjects can adjust their own feelings to them, or rather resist such elicitation. On this perspective, proposition (1) of the paradox can be denied, for, in order to be expressed, emotions do not require any sentient expresser.