Aesthetic Cognitivism

Is beauty a feature of objects in the world; or a feature of our individual experience of objects; or the result of aesthetic education and training, modulated by social culture? I argue that aesthetic properties are like perceptual properties, understood as Constitutionalism explains them: as real features of objects in the world that fully manifest themselves only in causal interaction with suitable perceivers. I here develop Constitutionalism to provide a metaphysical account of aesthetic properties in terms of causal powers. Like perceptual properties, aesthetic properties are (multi-track and multi-stage) powers of objects that mutually manifest with relevant powers of the perceiver. On this basis, I argue that aesthetic judgements, like perceptual judgements, are apt for truth and falsity, and their truth value derives from the reliable experiences that


Introduction
Would the world be lushly coloured, tasty and smelly as we experience it, if there were no one to perceive it? Are colours, tastes, smells, etc. features of the world, or features of our experience? Some philosophers hold that perceptible properties like colours are 'out there in the world', but are exhaustively accounted for in terms of the physical properties of objects -let us call this view Physicalism; whilst others claim that colours are 'in the mind of the beholder' and wouldn't exist without perceivers -let us call this view Projectivism. Various attempts have been made in the philosophical literature to do justice to the contrasting and yet compelling intuitions motivating these two views about colours and suchlike properties; intuitively, colours seem to be 'out there' and also 'in the mind'. If the qualities of the objects in the world and the qualities of our experience of them were somehow connected, we could 'save the phenomena'; but what sort of connection would this be? Primitivism, a third view in the literature, holds that colours (and perceptual properties in general) are primitive intrinsic properties instantiated by physical objects; and that they are somehow constitutive of our phenomenal experience of them. 1 Primitivism is an appealing position: it does justice to the idea that objects are truly coloured, and colours are out there in the world, and it connects somehow -constitutively -what there is in the world with what there is in our mind. Yet, Primitivism posits, but doesn't account for, this all-important constitutive connection (which isn't identity) between our phenomenal experience of the world and the qualities of the world that we perceive. That there is no identity between the two is clear from a variety of cases, which we can subsume for convenience under the label of 'phenomenal variance': there is incontrovertible evidence that phenomenal experiences may vary without the perceived properties (instantiated by the physical objects) varying -to the extreme that it is possible to have phenomenal experiences of objects that don't even exist, e.g. when an object is hallucinated. Phenomenal variance is a datum of our experience of the world that Primitivism cannot ac-commodate in view of the constitutional relation between phenomenal experiences and perceptual properties. Our desideratum is a theory that accounts for perceptible properties, such as colour, as real properties of objects in the world, taking into account the existing variety of types of perceivers and perceiving conditions, and experiential cases such as non-veridical perception, hallucinations and inverted-spectrum scenarios (which Primitivism does not explain).
The key idea I argued for in previous work (Marmodoro 2006; Marmodoro, Grasso 2020) is that objects have qualitative features (e.g. colours) but, crucially, such properties are sensuous. Namely, they are such that they need to interact causally with perceivers (and not with trees or stones, or just light), to 'come to their full' (i.e. come to fully be what they are). They 'come to their full' only when and while the objects to which they belong interact causally with perceivers, under certain conditions. 2 What is the role of the perceiver on this account? It is to enable the objects to activate their colours in full, in a certain environment; thus serving as a necessary condition for that activation/activity, but without projecting colours onto the world. This requires causal interaction between the object and the perceiver; the realisation of powers of objects in the sense-organs of the perceiver. This causal interaction is constitutive of both of the object's manifested qualities and of the perceiver's experience of them -hence, the theory is called Constitutionalism. Objects are in full colour as they interact with perceivers.
I here lean on Constitutionalism as developed in previous work, and its central idea that some properties are sensuous, to provide a metaphysical account of aesthetic properties in terms of causal powers. In sections 2. and 3. I will introduce Constitutionalism and the key arguments that support it, and in section 4. I will argue that like perceptual properties, aesthetic properties are (multi-track and multi-stage) powers of objects that mutually manifest with relevant powers of the perceiver. On this basis, I argue that aesthetic judgements, like perceptual judgements, are apt for truth and falsity, and their truth value derives from the reliable experiences that produce them.

2
The Ontology of Sensuous Properties I will here assume the stance defended elsewhere (Marmodoro 2020) and shared by other metaphysicians, that the properties we admit in our ontology as bona fide ones should satisfy the so-called Eleat-2 Furthermore, changes in the conditions wherein the interaction takes place bring about different such manifestations of the qualitative features of objects, as we will see in more depth in sections 3 and 4.
ic Principle according to which only what is causally powerful is real (see Plato's Sophist 247e1-4; and e.g. Armstrong 1978, 5, in current philosophy). Properties, in short, are causal powers. Powers (e.g. heat) are essentially directed towards their manifestation or exercise (heating), which defines the type of power they are. While this account of causal powers is mainstream and goes under the name of dispositional essentialism, fewer metaphysicians (e.g. Molnar 2003 among others) hold the additional thesis, which I endorse, that a power's manifestation or exercise always happens as mutual manifestation of partner powers. Partner powers serve reciprocally as necessary conditions for each other's manifestation. 3 (Heat heats only when something is heated). Uniquely, in the current debate, I further hold that it is numerically the same power that is inactive in potentiality and then manifests. (The power that heats is the power that can heat). It is both intuitively compelling and philosophically sounder (for reasons given in Marmodoro 2020 and elsewhere) to think that a power's exercise is its activity, i.e. what the power does; rather than thinking that the exercise of a power is a numerically different, new power, causally related with the original one -which is the mainstream view in current debates (see e.g. Mumford 2011 among others). Powers, as I conceive of them, can endure being exercised; they may also endure various types of alteration by being exercised, as for example their strength may increase or diminish (e.g. the strength of the electric charge of a discharging capacitor diminishes). Some powers may endure repeated manifestations (as for instance the repelling power of an electron). 4 To understand how a power endures transitioning from potentiality to exercise it is helpful to recall that instantiated powers are tropes of physical powerfulness; they are real within nature, even if inactive. Power tropes in potentiality are physically present in the world; this is their reality, and is what grounds how they endure exercise, alteration, and repeated exercise. When powers exercise/manifest, they produce change in partner powers, which we can detect. 5 Drawing on my general metaphysics of powers, in the case of (genuine, not hallucinated) perceptions, I hold that there is a causal interaction between the powers of an object in the world and the perceiver's perceptual system. This causal interaction (under appropriate 3 Speaking more accurately, I distinguish (in Marmodoro 2020) between transitive and intransitive powers; the former (which include the powers of objects to cause certain experiences in the perceivers, and the powers of the perceivers to perceive them) are those which require manifestation partners.
4 There is ontological economy in individuating powers in a way that allows for repeated manifestation, rather than positing a different instance of the power each time.
5 The causal relation however, being contingent, gives rise to epistemological issues for the cognition of a power through its manifestation.

Anna Marmodoro
Aesthetic Cognitivism 45 circumstances, e.g. light conditions), grounds the co-occurrence of a specific phenomenal property in the perceiver (e.g. the experience of seeing a specific colour) and the qualitative character of the coloured surface. Thus the power of the object to (e.g.) appearing red, and the power of the perceiver to have an experience as of red are co-activated as mutual manifestation partners. 6 The perceiver's experience of red and the appearing red of the coloured surface are manifestations of two different powers, but constitutively connected on account of their mutual interdependence for their occurrence. 7 The fact that colours and suchlike properties are powers of the kind here outlined is key: colours are real properties of objects; however, the full manifestation of the qualitative properties of objects is dependent on the environment and on the operation of the observer's perceptual system. Thus, Constitutionalism explains in which sense colours are sensuous properties, and provides the explanation of what Primitivism leaves unexplained on account of positing properties as primitive. 3 The Role of the Perceiver If a tomato weren't red in the absence of any perceiver, it would seem the tomato's colour is somehow generated by its being perceived, in a projectivist, non-realist manner. On the other hand, if the tomato's being red in the absence of any perceiver were the manifestation of a numerically different power from the one that gets manifested in the presence of a perceiver, it would seem that what the perceiver sees is different from what's there when the perceiver is not there -which is counter-intuitive and deems the phenomenology of our perceptual experience systematically erroneous. I submit that perceptible properties are 'sensuous', namely, they are such that they require the presence of, and a causal interaction with the perceivers to be fully activated/manifested, in appropriate conditions. What the appropriate conditions are will vary for each type of power. Generalising, as I argued elsewhere (Marmodoro 2014;Marmodoro, Grasso 2020), sensuous properties are 'multi-stage' powers; and the role of perceivers is to enable their full manifestation, which co-occur with the perceiver's perceptual experience.
Let us now turn to phenomenal variance: suppose that how things are in the world is held fixed with regard to which sensuous proper-6 For clarity, sensuous properties are dependent on their co-manifestation partners, whether they are in potentiality or are manifesting. This does not entail that a tomato is red only if someone is seeing it, but only if someone can see it.
7 The mutual dependencies between phenomenal properties of our experience and the full qualitative character of objects are of co-determination, co-dependence, and co-variation, as discussed in Marmodoro 2006. ties an object has, and with regard to the obtaining of the appropriate conditions for their manifestations. Can there be variation in how different perceivers experience such a world? It seems more than plausible to answer in the positive, not only on the basis of everyday experience (think of possible illnesses or malformations affecting the sense organs), but also because of thought experiments such as the inverted spectrum one. So, suppose that perceiver A and B, while looking at the same tomato, experience phenomenal properties blue and red, respectively. Are they seeing the same colour? My answer is "Yes"; my theory can accommodate this type of case, on the understanding that colour is a multi-track power, whose manifestation is both blue and red, in A and B respectively, at the same time. Many even if not all power ontologists posit the existence of multi-track powers, such that a numerically single power may have manifestations of different types (called tracks). 8 A commonly referred to example of a multi-track power is the power of an electron to be affected by other electrons at various distances from it. In this case, the electron does not have as many numerically different powers to be affected as there are repelling powers of other electrons, but only one numerically same power that admits different manifestation types, and which is manifested in combination with different partner powers (i.e. the powers of other electrons). Thus, the (numerically) same power of the tomato in our example can give rise to different manifestation types: red and blue, for different perceivers (or even, in different manifestation conditions, e.g. in green light, for one perceiver).
To recapitulate, multi-track and multi-stage powers are such that they may have different manifestation types, each with multiple manifestation stages. Thus the numerically same power can be possessed by an object but not manifested; it can be enabled to manifest in the absence of a perceiver; and it can be fully manifested in the presence of a perceiver. Furthermore, the numerically same power (but a different track of it) can be equally fully manifested in the presence of a perceiver A with a different type of sense organ than perceiver B. The feature of being multi-stage is crucial, because it guarantees that what we perceive is really the power of the object; hence it provides a realist account of perceptible properties as properties of objects, and yet, the causal interaction with the perceiver 'makes a difference' to what there is in the world. (So e.g. dogs enjoy different 8 For further definitions of multi-track powers see Martin, Heil (1998, 1999, according to whom the same power can manifest itself differently in conjunction with different manifestation partners, and Choi, Fara (2016), for whom they are "[…] conventional dispositions that correspond to more than one pair of stimulus condition and manifestation (Ryle 1949, 43-5;Bird 2005, 367;Bird 2007, 21-4;Ellis, Lierse 1994, 29). The thought is that exactly the same conventional dispositions may be picked out by multiple characterisations in terms of stimulus condition and manifestation". colours in the world than we do). From this account follows that perceptible qualities are real properties of objects; however, their full activation/manifestation depends on the environment and on the observer's perceptual system. This is Constitutionalism. Constitutionalism provides a richer account than other theories of perceptible properties, via the apparatus of multi-track and multi-stage powers and the appeal to the existence of mutual dependencies between the phenomenal properties of experience and the qualitative character of objects. 9 The non-realist might at this point raise the following consideration: does the interdependence, in the mutual activation between an object's perceptible power and the power of a perceiver to perceive, undermine the objectivity of perception? Is the tomato, itself, really red or blue? Are there any properties out there in the world? If there is variation in how a certain perceptible property of an object might be activated/manifested in different perceivers, along different tracks, which manifestation of it is veridical? Are perceptual observations at all apt to have truth value? I argue that perceptual observations (as well as aesthetic observations, as we will see in the following section), are indeed true or false, because there are proper observers and observation conditions, which my account explains. In this sense, my position may be characterized as Perceptual Cognitivism. 10 My Perceptual Cognitivism is underpinned by a form of reliabilism. 11 The truth value of perceptual observations is determined by the causal process whose outcome is the perceptual experience (as well as, on my account, the full manifestation of a certain quality in the world). When this process takes place, the perceiver and world conditions within which it occurs determine the reliability of the observations. I submit that our common practice is to classify things as thus and so (e.g. as red) on the basis of our perceptual observations having taken place in 'appropriate' or 'standard' conditions. We discriminate between veridical and non-veridical perceptions and appearances of things in the world on the basis of the obtaining (or not) of such conditions, which make the perceptual process (the causal interaction between a perceptible and the relevant sense organ) reliable.
9 Which allow for their occurrence, and are further explained in Marmodoro 2006. 10 This view is Aristotelian in spirit, in the sense that it is built on Aristotle's theory of perception, which I defined elsewhere a subtle realist one (Marmodoro 2014). For the reader who might be interested in Aristotle's cognitivism in general, a helpful resource is Ian Mccready-Flora (2014). Aristotle has been an inspiration for pragmatist metaphysics of properties, from Dewey to Putnam and more, and to this degree, my account shares similarities with theirs.
11 See e.g. Laurence Bonjour: "the central idea of reliabilism is that what makes a belief epistemically justified is the cognitive reliability of the causal process via which it was produced" (2002,244).
What counts as 'appropriate', 'standard' or 'normal' conditions? We do have the notion of well-functioning sense organs in appropriate observation conditions, which facilitate the well functioning of the organs. I propose to understand the notion of a well-functioning sense organ, and correspondingly of a malfunctioning sense organ, as statistical notions, within the perceiver's species. 12 Thus, I hold that there are real properties of objects in the world, which we perceive; there is no privileged access to reality that confers truth on any particular perceiver's perceptual content. Yet, in human society, we have developed the notion of 'healthy perceivers and appropriate perpetual conditions'; accordingly, perceptual experiences can be true of false; the veridical ones are only those by well-placed and well-functioning perceivers. In that sense, 'not all perceptions are born equal', and so, "Man is the measure of things" in a different sense than as conceived by relativism. It is not the case, in my theory, that any perception is as veridical as any other; well-functioning perceivers operating in appropriate conditions are 'measures' of truth, because the process underlying the formation of their perceptual content is reliable; it counts as reliable because it is of the kind that the majority of individuals within the species have (rather than because it is somehow 'closer' to reality).

Aesthetic Properties
Perceivers observe redness in the world, but also beauty, elegance, harmony, etc. Is there beauty 'out there' in nature? My answer is 'Yes!' I argue that aesthetic properties are as real as perceptual properties; they qualify objects; but their nature, too, is to be sensuous properties, i.e. powers dependent for their full activation/manifestation on the observers. I will here concentrate on addressing a relativist challenge to the existence of aesthetic properties in the world (rather than on giving a detailed account of how my account of sensuous properties provides a metaphysics for aesthetic properties). The question I want to consider here is this: can both types of observations -of e.g. red, and of beauty -be veridical, latching on to something real in the world? I want to defend here the view that I call Aesthetic Cognitivism.
The relativist challenge to the reality of aesthetic properties, and the veracity of our observations of them, has been expressed in a variety of ways in the literature; one way is to deny the idea that there can be a well-functioning observer of aesthetic properties operating in appropriate conditions, that can serve as a 'measure' of e.g. beauty. Robert Hopkins, among others, contrasts the existence of a reference class that can serve as 'measure' of perceptible properties to that of an (impossible, to his mind) reference class for aesthetic properties: if on looking at something I judge it red, but everyone else I ask to look at it judges it brown, this can be reason enough for me to think my view wrong […]. In an aesthetic case, in contrast, I am never justified in going that far […]. (2001, From this stance follows that aesthetic judgements cannot be right or wrong, because there is no standard against which to measure their veracity. The idea is that, while for perceptual properties we can distinguish between what e.g. Philip Pettit (1983) has called a "primitive" and a "rectified" report, 13 we cannot for aesthetic properties, because, there is no way to determine which group of observers might be considered the 'normal' one, with reference to which, aesthetic reports can be rectified. Thus, there cannot be a reliable process that leads to a correct report on aesthetic properties.
Interestingly, Roger Scruton takes this position in the direction of non-realism about aesthetic properties, as follows: "in aesthetics you have to see for yourself precisely because what you have to 'see' is not a property" (1974,54). What is it that one 'sees', according to Scruton, when looking at an artwork? I understand Scruton's position as it is glossed in the passage below: As an example, consider the face in a picture. This is not visible to a dog, but only to a being with imagination (which is a rational capacity). There are physical features of the picture which explain the fact that I see a face in it, and which could be described in primary-qualities terms. But the face is not part of them. Nor is there any law which says, to a being with certain sensory capacities (for example sensitivity to light rays), the face will automatically appear. (Scruton 2010, 100) The step from conceiving aesthetic properties along these lines, to being non-realist about them, is short, and Scruton takes it; his lines quoted above are followed by this claim: "Because of this dependence on rational capacities, it is possible that the face can be 'argued away'" (Scruton 2010, 100).
The contrast with Constitutionalism and with the Aristotelian position from which I derive Constitutionalism, is stark.
In a 'linguistic turn' mode, Scruton does 'argue away' aesthetic properties, thus: Consider the sadness of a piece of music or the gravity of a verse. Few of us feel tempted to follow Berkeley in thinking that secondary qualities are not really "in" the objects which seem to possess them. But we all feel tempted to say something like that of tertiary qualities. There comes a point, we feel, when it is only a matter of speaking to refer to a property of an object. The real fact of the matter is the response of the observer. If we speak of a property of an object, this is just a matter of saying that the response was justified (as when we describe a landscape as 'fearful'). (2010,100) So for Scruton, the fact that our response to an aesthetic quality is justified, is not sufficient ground for realism about these properties.
In response to Scruton's eliminativist move, I want to argue that we have ways to support the same combination of realism, cognitivism and reliabilism concerning aesthetic properties that I have put forward for perceptible properties. The interesting question to address is: What counts as normal conditions, for aesthetic observations to be reliable? My answer is that every society has observers who are recognized for their well-functioning aesthetic judgement; they set aesthetic 'norms'; and they can educate society in such norms. This is a plausible view to hold, and commonly shared. Therefore, there is a public domain wherein there is 'right' and 'wrong' in aesthetic judgements. Reliabilism, modulo society, is not relativism, because there are reliable 'observers of truth' in each and every society; hence other observers can be corrected or confirmed, using public criteria of truth, within a society. This position is not relativist: while relativism would claim that truth is relative to each individual, where there cannot be 'rectified' (perceptual or aesthetic) reports, Aesthetic Cognitivism is the view that the truth of aesthetic judgements is dependent on society's practice with respect to aesthetic norms, as much as the truth of perceptual judgements is relative to the human species.

Conclusion
Are colours, sounds and beauty in the world, or in our heads? Is the world as we experience it? How do the properties of the world relate to the observers' experience of them? According to Constitutionalism, colours and suchlike perceptual and aesthetic properties are powers of objects whose (full) manifestation depends on the mutual manifestation of relevant observation powers of perceivers, and is co-realized JOLMA e-ISSN 2723-9640 1(1), 2020, 41-52

Anna Marmodoro
Aesthetic Cognitivism 51 with them in their interaction. In addition, Constitutionalism takes such properties to be multi-track, and also multi-stage powers. The track-dimension explains why the numerically same property of an object can have different types of manifestation (for instance, appear as different colours to different perceivers). The stage-dimension allows us to preserve realism about the properties, while accounting for the crucial role of the perceiver/observer in the causal interaction. The causal interactions between observers and objects in the world, in the case of perceptual and of aesthetic properties, admit of 'standard conditions', and thus support a form of Cognitivism according to which, perceptual and aesthetic reports about properties have truth value, based on the reliability of the process they result from.