Rob van Gerwen
Aesthetic Argument and Tertiary Qualities

1. Grand aesthetic terms uninformative.

Disagreement about aesthetic values is quite common. However, we hardly ever argue about aesthetic judgements like "this is beautiful". A philosophical tale that is being told for two hundred years or more assumes that such aesthetic judgements are typically expressed with grand terms like 'beautiful', 'sublime', or 'excellent' at the predicate place. These terms do nothing to specify the 'this' at the grammatical subject place of the proposition, but rather express a specific, aesthetic, type of experience of judgement.[1] They inform us about the speaker's claim of having based her judgement upon such an experience and of, therefore, being justified to judge. However, these categories do not specify the state or states of mind that supposedly legitimize our judgement, nor even do they attract our attention to them. Instead we think they attract attention to the object and its details. In reality, however, they do nothing of the kind. Saying of a Rembrandt self-portrait that it is excellent merely invites people to watch it more intently. This suggests the relative unimportance of these grand aesthetic terms, since they do not inform us: not about the object nor about the judging subject. Starting, not from the perspective of the terms used in our judgements, but from the object, an analogous problem arises: we cannot infer the aesthetic value of an object from its 'objective' properties, due to a lack of rules or theories that link specific (combinations of) objective properties with aesthetic evaluations. The named grand aesthetic terms may often occasion disagreement but shall be of little help if any in subsequent aesthetic arguments. We use a different terminology to argue matters aesthetic: devoid of highbrow pretensions these terms primarily are descriptive. However, of these descriptive terms it is not evident how they should be relevant for the aesthetic judgement they supposedly help us explain. And this confronts us with a dilemma: either we give up the grand categories and their experiential claim to evaluational legitimacy, and consequently restrict the analysis of aesthetic discourse to allegedly descriptive, critical, remarks; or we honour the grand categories and their experiential implications, but shall be helpless in specifying these categories' relevance for our aesthetic discussions. The latter strategy has been the one included in the old, Kantian, tale; the former, defending the relevance of critical language stems from more recent date, from the analytical approach to linguistics. I propose to dissolve this dilemma by expanding our ontology with a kind of wonder properties, tertiary qualities, which although they are attributable to aesthetic objects, depend for their discernment on specific mental activities that are central for our aesthetic experience as well. What we need to achieve this mixture of seemingly incompatible strategies are, first, disputable categories to describe the aesthetic object with. These categories must, secondly, be so deeply involved in the aesthetic experience that allegedly justifies our judgements, that they at once clarify how they can form the reasons for our seemingly incorrigible grand claims. I shall start with the first demand of descriptive relevance, and shall return in the end to the second one of experiential relevance.

2. Grand and descriptive terms.

The reasons we provide for our aesthetic judgement must be put in descriptive terms lest they be as irrelevant to third party onlookers as grand aesthetic terms appear to be. But on what grounds do we take a judgement to be aesthetic? Traditionally, at least with Kant, a judgement is stipulated to be aesthetic if and only if it is expressed in a proposition involving predication with an aesthetic term, but equally traditionally, Kant reduced the candidates for such predication to the terms 'beautiful', and 'sublime'. Kant argued that the correct application of these terms presupposes certain negative and positive constraints.[2] Put in a nutshell, such terms as these are aesthetic if and only if their application is ruled not by understanding, but by taste, and just what this taste should amount to Kant has analyzed in terms of the free play of the cognitive faculties. None of the arguments he used to establish his point of view has changed the stipulative linkage of 'aesthetic' with taste. So Kant thought that there are only two terms that are applicable by taste only. Frank Sibley proposed a rather more elaborate set of terms, but did not change the stipulation either. According to Sibley we describe an aesthetic object in terms of 'tenderness', 'tenseness', or 'harmony', or we call them 'frightening', or 'appalling' without meaning these to be literally applicable; or 'dainty', 'graceful', and 'elegant', used in more straightforward ways.[3] Even the propositions tagged 'aesthetic' by Sibley derive their aestheticness from the fact that their 'correct' assessment is alleged to be an exclusive matter of taste. I think Sibley was right in sustaining the taste-'aesthetic' link,[4] but he was certainly wrong in thinking that other terms than the Kantian 'grand' ones deserve to be tagged 'aesthetic'. This much shall follow from my argument. Thus, if one wants other kinds of terms to be relevant to aesthetic discourse additional arguments are needed, arguments that relate these terms to the grand ones, and to taste.[5] Evidently, the terms we are looking for must inform us of the whereabouts of the object, but it is an open question whether some object or event described correctly with whatever terms apart from the grand ones is thereby judged aesthetically, or, as we have it, with taste. One can deny that this is an open question by arguing that calling something 'elegant' means to judge it aesthetically, because this descriptive term's aesthetic implications supposedly forms an analytical part of its meaning. In such a case 'beauty', or its fellow term 'sublime', supposedly forms part of the meaning of 'elegance'. However plausible this claim to analyticity may seem, it is a problem to establish exactly which of the grand categories ought to correspond with each relevantly descriptive term. The answer to this problem may depend not on the meaning of the terms involved but rather on the objects we want them to describe. So that, once we accede to the idea of there being more than one distinguishable grand category it is indeed an open question whether by applying any of these descriptive terms to an object we also judge this object aesthetically. Since we are about to explain the relation between the grand and the descriptive terms we are going to need different arguments. I want to argue that the terms that are or can be relevant for our aesthetic arguments describe properties in the world that we cannot perceive merely by looking at or listening to them, but only by using our empathic imagination in doing so. I am not falling back on a simplistic sense data theory of perception but distinguish between two kinds of imagination to make my point. "Merely looking" I argue presupposes a different brand of imagination than is required for the appreciation of aesthetic properties. We use a grand aesthetic term to judge the force with which our empathic imagination is being engrossed. I.e. by using a grand aesthetic term we evaluate not so much the aesthetic properties in the object to which we possibly refer as forming the reasons for our aesthetic judgements, but rather the specific process of perceiving them.[6] Thus, in what follows I shall argue, first for the need to ontologically expand the make-up of the world as I intend to do by introducing tertiary qualities, and, secondly, for a logical distinction between kinds of imaginative activities.

3. Primary and secondary qualities.

Sheer enumeration of objective properties shall do little if anything to explain why we find an object beautiful. We do not point at the 'lively' Kandinsky painting saying "Look, it is square, three inches high, it has a red patch over there, and a yellow stripe beneath it, and, there, from left to right this blue diagonal daub of paint".[7] Recognition of the painting's liveliness presupposes that we project certain psychological and behavioral considerations onto the plane of paint; considerations, for example, like the ones attributed to a lively boy's wild movements and frivolous yellings. Aesthetic properties involve this subjective projection instead of a more or less passive taking in. As Richard Wollheim has argued such projections imply that an account of aesthetic evaluation should be subjectivist.[8] To specify the nature of such projection let me first look at the nature of qualities that admit of being taken in more or less passively. 18th century empiricism suggested that there exist two kinds of perceptive properties, primary and secondary qualities, neither of which, I shall argue, sufficiently explains the subjectivist projection characteristic of aesthetic properties. Locke used three arguments.[9] An epistemological argument in terms of the role of our mental faculties; an ontological one in terms of what does and does not belong to the object in itself, and a third argument in terms of whether or not the ideas we have of these qualities resemble them. In what follows I shall not go into the third argument, as I don't think it is intelligible. (My idea of a red patch is not itself 'red', but neither shall my idea of a table possess the same form or motion that the relevant table has). Locke's epistemological argument runs as follows:[10] Primary qualities such as mobility, solidity, number, and figure are perceived by more than one of the senses; in particular they can be sensed by touch and sight alike.[11] Secondary qualities, such as colours, tone, taste, and smell, on the contrary, are revealed to one of the senses only, and it is impossible for a person missing the appropriate sense to ever form the right idea of a relevant property. One cannot possibly explain a tone or even, for that matter, the concept of 'tone', to a person born deaf. In his second, ontological, argument Locke argues that primary qualities, such as figure, are inseparable from the bodies they adhere to: splitting a grain of wheat still leaves us with extended, solid, bodies that are mobile or at rest, and which have a certain number. Secondary qualities, on the contrary, are said to be nothing in the object but dispositions to produce, by way of the primary qualities of the object, some specific sensation in a subject suitably equipped.[12] If it weren't sustained by the epistemological argument, however, this ontological argument would not hold, for several reasons. First, if secondary qualities are identified by the impressions they produce in us, then there is hardly a reason to posit in the object such ontological oddities as 'dispositions to produce them'. How would we know that secondary qualities are to be identified by such odd powers in the object, if not by our perceiving these powers. But either we perceive them by perceiving the secondary qualities they cause in us, or we have an independent access to them which would reduce them to purely primary qualities in the end. Secondly, and more importantly, if the dispositional powers are there in the objects there appears to be no reason for them to disappear whenever the object is changed in any way. I agree with the Lockean assumption that a changed object will still have a figure, even though it be different from its original figure, and also some or other extension, even though it be different from its original extension, but in the same vein a changed object shall also have some or other colour, tasty or tactile quality. The related argument that only primary qualities are causally effective is oversimplifying: secondary qualities are causally effective as well, as the hot sunshine on the black roof easily attests.[13] Finally, we may all agree that specific faculties are needed to develop general concepts of secondary quality kinds, such as 'colour', or 'tone'. But this general faculty is insufficient for the perception of a red rather than blue, shade of colour. For singular perceptions of secondary qualities real objects are presupposed, not mere dispositions: actualities. So out goes the idea that secondary qualities are illusory and that only the primary ones really adhere to the object. In short, the ontological argument cannot be sustained independently from the epistemological argument.

4. Against the distinction.

The epistemological argument, that secondary qualities are perceived by only one of the senses whereas primary qualities are perceived by several, is incapable of proving the point it is designed to bring home, of there being an ontological distinction between primary and secondary qualities. It serves two theses, and this leads to confusion if not conflict. The first thesis relates poly-modal perceivability to ontological provability, the second one relates it to knowledge of the nature of the perceived. As an existential proof polymodality - if elaborated sufficiently - seems to form a convincing criterion,[14] but as a source of establishing the exact nature of properties - what they are like - it is rather weak. Anyway I am not convinced that these theses put secondary qualities on a different footing from primary ones. First of all: secondary qualities are as real as primary ones. We have produced artificial means of poly-modal efficiency that are sufficient to establish the existence of secondary qualities. Even though we must have seen a red patch to understand what 'red' means, once we are in the possession of the relevant faculty (colour vision) we can prove just what colour is there in the object by referring to sample sheets, or by intellectually interpreting some scientific diagram or number. Through samples, scientific measurement, and possibly yet other operationalizable procedures we can accurately establish the existence of secondary qualities even though such means as these do absolutely nothing to further our insights in the exact nature of the properties involved. The conditions under which secondary qualities shall be correctly perceived can be operationalized as convincingly as we claim they can with primary qualities. We may conclude that our perception is equally efficient regarding secondary qualities as it is regarding primary ones. Therefore, the existence of both kinds of qualities can be proven by poly-modal means. Secondly, if primary qualities are such that they be perceivable by more than one of our senses, then it is presupposed that they be perceivable, and that notwithstanding their 'primacy' they resemble secondary qualities in being dispositions of objects that cause perceptions in us. Thus, the exact nature of species of both kinds of properties is response-dependent, as our contemporaries would call it, and, consequently, this nature cannot be assessed objectively.[15] This goes for primary as well as secondary qualities. This creates a problem of translation for the poly-modality-thesis regardless of whether it is the nature of primary or of secondary qualities that we want to assess. It is evident that no light-wave numbers can explain what a red patch should look like, but we hardly fare better with the primary qualities. If 'figure' is supposed to be perceivable through two senses, sight and touch, we are still confronted with the task of specifying exactly how the data from touch are to be translated into those of sight, or the other way around.[16] Tactile data are processed by causal connections between the senses and the objects producing them, whereas visual and audible data stem from rather distinct kinds of distantial processing. Let me give an example: when I watch my son brush his teeth, I hear the sound of the brush and see it going up and down, and common sense tells us that this indeed is how brushing one's teeth shall look and sound. However, I also see the tiles on the wall glistening from being polished, I see the colour of my son's face and clothes, none of which do I also hear; instead, I hear the ticking of the clock in the adjacent room, a car passing by outside, and none of these things I see. What I see is in this room, and much of it remains inaudible, whereas what I hear is inside and outside this room, and is only partially visible. Now if these two senses structure the world in so incomparably different ways, on which grounds do we come to think of the sight and sound of teeth brushing as hanging together, and as forming the phenomenal appearance of a singular event? Seen from this angle the distinction between primary and secondary qualities appears to be born from a misunderstanding of the incompatibility of data produced by the various senses. We may conclude that poly-modality is of no help once we try to explain the exact nature of the relevant qualities, and therefore, for an assessment of their nature primary qualities are as dependent upon perceptual states as secondary qualities are supposed to be. Regarding both types of qualities then existence can be proven once some sort of operationalizable polymodality is installed, but with or without such polymodality no explanation of their specific nature is forthcoming. If there were properties that would confront us with a total lack of poly-modal accessibility, however, we would most certainly have to be antirealists regarding their ontology, since we would be incapable of proving their existence. Neither could we conceive of being in error about their specific nature. However, it would be their ontological peculiarity that would mark them off from primary and secondary qualities, but this cannot be understood without the epistemology involved, as we saw above. Well, aesthetic properties are like that.

5. Aesthetic properties.

The existence of aesthetic qualities, in contradistinction with primary and secondary ones, is not poly-modally assessible. Not with our more natural faculties, nor with publicly accessible instruments. We possess no relevant theories, because in art no universalizable problems are being addressed,[17] nor can we refer to available samples,[18] because it isn't the secondary qualities that make up the art work that found our aesthetic judgement, but the absolute singularity in which these secondary qualities are related to one another (and to other works). Such singularity can only be appreciated by an acquaintance with the relevant object, and this acquaintance, as a correlary of the singularity of its object, must be conceived as subjectivist in nature. It is my thesis that to perceive an aesthetic property one must realize that it is not merely primary and secondary qualities that must be taken in, but first and foremost such properties' experiential dimension: one shouldn't merely watch intently, but empathize instead. Indeed, secondary qualities are response-dependent regarding the specifics of their nature, but not in the 'aesthetic' sense of forcing us to be anti-realist about them. 'Frightening', to name an example referred to in the literature, may be response-dependent, but it certainly is not a secondary quality: whether some situation or other is frightening depends on one's imaginative associations with what is perceivable in it.[19] Neither, however, does 'frightening' automatically function as an aesthetic property term, although it is, in the sense I am about to explain, a tertiary quality.

6. Empathic imagination.

Let us, therefore, look deeper into the nature and ontology of the aesthetic properties correlating with descriptive terms such as those addressed by Sibley. Aesthetic properties cannot be proven to exist by using samples, scales, or other objective, operationalizable, categorizations. We do seem to possess exemplary art works, but these are essentially different from a sample of colour or textile, or from the tone scale: art masterpieces are unique and singularly exemplary. Here we find a type of properties of which it is not altogether clear from the start how natural their ontology is. We seem unable to prove their existence notwithstanding our having an idea of their nature. This I propose to explain as follows: regarding aesthetic properties none of the senses provide sufficient evidence of their existence, let alone of their nature. What is needed in addition is empathic imagination. Imagination is the faculty of using the absent to structure and inform what is present to the senses, and it can be seen as coming in two guises, one instrumental, the other empathic.[20] We imagine instrumentally when trying to accurately perceive and understand objective and mechanicist implications (from past, present, and relevant future). It is instrumental imagination that is used for the application of concepts (to recognize that this or that object can be subsumed under such and such a concept; Kant's "schematism" and his "analogies of experience"), but also for the recognition of possibilities or impossibilities with regard to behaviour. For example, a cat uses his instrumental imagination when, after recognizing the sound of a mouse he starts stalking. In contradistinction, when we empathize we inform all this with an experiential dimension as well. We imagine what it would be like to experience this or that situation.[21] Why else would we want to think of a fast moving car as 'frightening' than because we imagine ourselves to be the victim of its speed. Such imaginative anticipation involves instrumental and mechanicist imaginations as well, in that one imagines the splashing of a soft object under the wheels of the vehicle regardless of the kind of object. Such imagining becomes empathic not by introducing, say, a human corpse as being run over (although this will stir some thoughts as well), nor even by merely introducing a living human being, but by introducing the experience to be had by such a human being at being run over. In art appreciation we can distinguish both kinds of imaginative activity as well: for hearing a sound healthy ears suffice, to hear how they form a melody, instrumental imagination would have to come in, but to hear that the melody is sad we shall need empathic imagination on top. It forms part of art's experiential functionality that it enhances our empathic imagination. 'Aesthetic' properties, I propose, express the experiential implicatures of the valued object.[22] One who evaluates aesthetically must not only project imagined events of some general nature onto the object, but experiential implicatures specifically. Aesthetic evaluation presupposes this reversal, typical for tertiary qualities, of the direction in which information is being exchanged between object and the subject's experience, and the experiential naturte of the projections involved. This explains why we should argue so much about aesthetic judgements. It also explains why we should be subjectivists in this area.[23] If aesthetic arguments could be settled simply by pointing at the primary or secondary qualities present to the senses then such arguments would be as redundant as they would be uninteresting. A property then is a tertiary quality if its existence and nature is assessed by way of informing the perceptual, primary and secondary, qualities it supervenes upon with experiential associations (memories and anticipations) provided by the beholder's imagination; his or her empathic imagination brings in a narrative structure, a temporal order, and experiential implications.[24] Thus tertiary quality terms specify the human perspective on the object or event under consideration. Because, in the words of David Wiggins, such properties really are property-response-pairs,[25] this introduction of idiosyncrasies should not pose any further problems for the theory: their imaginative subjectivity is basic. And, following his line of argument, the appropriateness of some tertiary quality term's attribution can and must be settled intersubjectively by discussions. My provisional defence against relativism is that imaginative considerations notwithstanding their idiosyncratic origins derive their relevance for the awareness of tertiary qualities only insofar as these tertiary qualities can be recognized from a third person perspective, i.e., by other people's imaginative associations. They must be shareable and arguably appropriate.[26] The imaginative associations are functional in enabling people to perceive the aesthetic properties, and only in this functionality are they relevant for our aesthetic judgements. It is the shareability[27] of our imaginative perceiving that explains why we should find aesthetic properties valuable.[28] Epistemologically, tertiary qualities are properties of a special kind in that they can be recognized and explained only through empathically imaginative activities: no mere pointing to what is available to however many of our senses shall ever suffice. Tertiary qualities are truly response-dependent. Ontologically we confront great problems in proving the existence of some tertiary quality, because there are no ways independent of imagination with which to establish exactly which property is there to be recognized. We hold no samples or theories, only individual psychology (and what universalizable elements thereof we allege exist). Here, more than anywhere else, argument is of the essence, and, therefore, we cannot be realists. This position I tag "imaginativist subjectivism". But tertiary qualities are not automatically also aesthetically relevant, i.e. they do not automatically coincide with aesthetic properties. It is, I think, at least a well-educated quess that tertiary qualities that are being judged in the light of their behavioral and material effects relative to standards of the good life, are viewed as ethical properties, and their judgement regards moral values.[29] The answer of imaginativist subjectivism to the open question argument held against Sibley is that such tertiary qualities as are recognized by empathic imagination are aesthetic properties, or reasons, if and only if their 'force' of occasioning such recognition is being judged by taste.[30] So aesthetic properties are not perceived by taste, but by a species of imagination; taste is what judges these properties' experiential efficacy (pace Sibley).

References

Gaut, B. (1994). "Metaphor and the Understanding of Art." : Paper presented at the Dutch-British Conference on aesthetics, Oxford, September 1994.

Gerwen, R. v. (1995). "Kant's Regulative Principle of Aesthetic Excellence: The Ideal Aesthetic Experience." Kant-Studien 86: 331-345.

Hacker, P. M. S. (1987). Appearance and Reality: A Philosophical Investigation into Perception and Perceptual Qualities. Oxford, Basil Blackwell.

Hacking, I. (1983). Representing and Intervening: Introductory Topics in the Philosophy of Natural Science. Cambridge, Cambridge Univ Pr.

Johnston, M. (1989). "Dispositional theories of value." Aris Soc, SUPP 63: 139-174.

Lewis, D. (1989). "Dispositional theories of value." Aris Soc, SUPP 63: 113-137.

Locke, J. (1690). An Essay Concerning Human Understanding. London, Printed for Tho. Basset.

Lorand, R. (1994). "Ethics and Aesthetics are not One: Aesthetics as an Independent Philosophical Discipline." : Personal Communication.

McDowell, J. (1985). "Values and Secondary Qualities". Morality and Objectivity Ed. T. Honderich. London, Routledge and Kegan Paul. 110-129.

Pettit, P. (1991). "Realism and Response-Dependence." Mind : 587-626.

Scruton, R. (1983). Public Text and Common Reader. The Aesthetic Understanding: Essays in the Philosophy of Art and Culture Scruton, R. London, New York, Methuen.

Sibley, F. (1959). "Aesthetic Concepts." In: International Philosophical Review 68: p. 421-50.

Sibley, F. (1963). "Aesthetic Concepts: A Rejoinder." Philosophical Review 72:

Wiggins, D. (1987). A Sensible Subjectivism? Needs, Values, Truth: Essays in the Philosophy of Value Wiggins, D. Oxford, Blackwell. 185-214.

Wollheim, R. (1980). Art and Evaluation. Art and its Objects Wollheim, R. Cambridge, Cambridge U.P. 227-240.

Wollheim, R. (1993). Correspondence, Projective Properties, and Expression in the Arts. The Mind and its Depths Wollheim, R. Cambridge (Mass.), London (England), Harvard University Press. 144-158.

Zemach, E. M. (1993). "The Ontology of Aesthetic Properties." Iyyun 42: 49-66.

The investigations were supported by the Foundation for Philosophy and Theology Research (SFT, which is subsidized by the Netherlands Organization for Scientific Research (NWO).