The Elasticity of Perception: Undermining the (Non-)Conceptualism Debate

In the current philosophy of perception, a debate about whether concepts permeate perceptual states in constituting the perceptual object or not has been widely discussed. Analytic philosophers and phenomenologists participate in this debate likewise, but it is also a debate in Kantian scholarship since the conceptualists’ thesis goes back to Kant’s Criticism and neo-Kantians already discussing such theory against any philosophy of immediate experience long before Wilfrid Sellars had started his attack against the so-called myth of the given. In light of this historical panorama, the article reconstructs Ernst Cas-sirer’s views on perception in order to systematically reject both Conceptualism and Non--Conceptualism. It can be shown that both positions are uncritical stances which make claims to either the absoluteness of language or perception and that much-discussed arguments such as the fineness of grain argument rely on a category mistake. The proposed solution is a view that upholds the criticism against the myth of the given, but replaces the idea of a conceptual mediation of perceptual experience with a symbolic mediation. As a consequence, perception must perform a paradoxical feat and has thus to be elastic.


Introduction
The philosophy of perception is the perfect field for bringing Neo-Kantianism, especially that of Ernst Cassirer, into conversation with analytic philosophy.To this end, the paper's aim is to undermine the (Non-)Conceptualism Debate within this field by drawing on some of Cassirer's conceptual tools regarding perception.It can be shown that this debate rests on at least one category mistake and a not sufficiently complex concept of the concept.This leads to the effect that Conceptualists absolutize such an inadequate way of looking at concepts whilst non-Conceptualists absolutize the given by unnecessarily targeting this false understanding of concepts.Conceptualists hence have to defend an overly rationalist form of cognitivism whereas non-Conceptualists wrongly tend to think that any form of cognitivism is intellectualistic.Cassirer's concept of the symbol can correct those views and in light of it the whole debate can be rejected as an impasse.
The article starts with a reconstruction of the (Non-)Conceptualism debate as discussed in current analytic Kantianism and with particular consideration of John McDowell's views on perception (section 1).I will then present Cassirer's way of distinguishing thing and expressive perception (section 2) in order to gain a more encompassing conception of perception that can be termed 'elastic perception' (section 3).In the last section I will show how to identify the erroneous assumptions on both sides of the debate and debunk the fineness of grain argument as a category mistake.

Kantian (Non-)Conceptualism
In the Critique of Pure Reason, Kant gives us various definitions of perception that at first glance seem to contradict each other.In the deduction of the pure concepts of the understanding according to the A edition it says: "The first thing that is given to us is appearance, which, if it is combined with consciousness, is called perception" (KrV, A 119f).In §22 of the deduction according to the B edition, however, he defines perception only as "representations accompanied with sensation" (KrV, B 147). and specifies in the paralogism chapter that a given sensation, "which, if it is applied to an object in general without determining it, is called perception" (KrV, A 374).To elucidate and conciliate these definitions, it is worthwhile to look at Kant's model of representation, which he explains at the beginning of the Transcendental Dialectic as a progression (Stufenleiter): "The genus is representation in general (repraesentatio).Under it stands the representation with consciousness (perceptio).A perception that refers to the subject as a modification of its state is a sensation (sensatio); an objective perception is a cognition (cognition)" (KrV, A 320).Furthermore, Kant defines intuitions as "singular" and "immediately related to the object", while concepts "by means of a mark" relate mediately to what "can be common to several things". 2 Kant so calls objective perception cognition.But our perception does not only deliver cognition, already the further term "perceptual experience" points to that. 3 The phenomenon perception is not only related to epistemic claims, but is first of all an experience sui generis, which is often described to have a certain content.How would this content be described and methodically grasped if it is -as we want to assume -more than a propositional content, such as a thought?It is along this problem of content or purport that the debate between conceptualists and non-conceptualists in contemporary Kantianism and the philosophy of perception runs.Strictly speaking, it is about similarities and dissimilarities of the 2 All quotes ibid.3 Perception and perceptual experience are terminologically distinguished today.The former tends to be the focus of theories of perception, which ask about the status of perceptions in contrast to hallucinations and illusions.The latter takes into account the fact that perceptions normally give us reason to beliefs about the external world and therefore focuses on the nature of the relationship between perceptions and the totality of our experience.
Studia Kantiana | vol.20 n. 2 | ago.2022 The Elasticity of Perception structure of the content of concepts on the one hand and perceptual experiences on the other hand, and one asks primarily with reference to the dissimilarities of both, how they interact.Many interpreters, who orientate themselves on the concept of experience rather than on the strictly terminological concept of perception, overlook the fact that, as soon as they follow their intuition concerning the content structure of perception, they have skipped a more decisive question: For Kant asks about the conditions of objectivity of the intentionality of perceptual experiences.He asks what conditions must be met at all in order to have representational content.From a logical point of view, this precedes the question about the nature of the content of perceptual experiences, since first it has to be clarified what it means objectively to have content in a perceptual experience at all.Conceptualists and non-conceptualists now agree on one thing: Perceiving is more than having sensory impressions; perception structurally is a relation between a perceiving subject and a perceived object.Perception thus has objects as its content.Impressions are merely instantaneous and perspectival.Objects, on the other hand, systematically unite perspectival impressions into a spatiotemporal unity.Consequently, the perception of objects in this respect of content is an awareness of unities.This brings us back to Kant's definitions at the beginning of this chapter: The content of perceptual experience are appearances or "objects in general", which we first sense by affection, however, as Kant also says, "without determining them".The determination of objects, on the other hand, is done by concepts, that is, by the understanding.
I would now like to briefly present the heterogeneity of sensibility and understanding and then collect the arguments that non-conceptualists adopt with reference to Kant. 4 The difference between thinking and perceiving can be made plausible on the basis of several aspects of discursivity and intuition.A basic idea of discursivity states that thinking is judging and operates conceptually.Concepts are thus (1) general representations, (2) classificatory, they instantiate something in its generality and (3) logically articulated as inferences.A standard view of intuitions is that they are perceptions in space and time.Perceptions are thus (1) singular intuitions, (2) completely determined in their presence and (3) spatiotemporally structured, i.e. they have objects in space and time as their content.From this follows directly the heterogeneity of sensibility and understanding, which Kant described in a prominent passage as follows: "Thoughts without content are empty, intuitions without concepts are blind" (KrV, A 51/B 75).Now, basic problems for conceptualists, that is, claims supporting the non-conceptuality of perception, are primarily the (a) fine-grainedness of perception, (b) the situation-dependence of perception, and (c) the commonality with animals and humans of pre-linguistic age with respect to perception.(A) states that concepts are more rough-grained than anything that takes place in perceptual experiences.This thesis allegedly goes back to Gareth Evans and Christopher Peacocke, 5 but can already be found in Schopenhauer 6 and Nietzsche 7 to name but two earlier philosophers.It can be easily illustrated if we imagine that we can perceive, e.g., color gradations in a much more differentiated way than actual concepts would be available to us.(B) states that in perception, content is presented rather than represented by actual perceptual instantiation.(C) states that children who do not yet have fully developed rational standards and language, or animals who do not have them in the first place, share at least a basal layer of perception with rational creatures, since they can also individuate objects in their environment, though not necessarily classify them further.Non-conceptualists such as Robert Hanna now endeavor to show, with reference to Kant, that the ability to locate objects in space and time can do without concepts. 8Then again, in Enactivism, which also is part of this type of theorizing perception, one speaks of abilities and embodiment.Object perception is primarily understood as a purely sensory ability.Authors like Alva Noë try to describe this by including cognitive abilities, but the essential point is that these are non-conceptual abilities.9 If one refers these theories back to Kant, one is tempted to consider the Transcendental Aesthetics independent of the Transcendental Logic.But what point of view does one gain by this at all, apart from the assertion that also animals can identify objects of their environment and that humans are animals, too? 10 It could be denied at first that non-conceptualists touch Kant's problem at all.Kant wants to make existential or actuality statements and ultimately asks about the epistemic functionality of perceptual contents. 11Kant does not ask for the concrete object of perception, but first wants to secure, against skeptical objections, knowledge transcendentally by the idea that the formal object of perception is identical with the formal object of thinking.Because of this interest in knowledge, he is concerned with the question of in which way perception and thinking are made for each other, how perception and concept can 'fit together'.From the metaperspective, again -for all the difficulty of showing such fitting -the problem arises that if the object of perception and the object of thought matched only contingently, there would be no epistemic function in a true sense. 12So that if objects can be content of knowledge at all, they must also be content of judgement.The sense of object here is: formal object of thought.If now, under a non-conceptualistic sign, the unity of the object is transferred to pure intuition, one needs an additional step to show that the objective unity of intuition and the objective unity of thinking are the same.Kant, however, asks directly for the conditions of possibility of this identity and determines thereupon the object of cognition.A Kantian non-conceptualism thus fails to ensure the objective unity of intuitions. 13Without the involvement of concepts, intuitions cannot play an epistemic role; intuitions remain blind.
John McDowell has adopted this form of argumentation and thus objected to Gareth Evans' theory of informational states in Mind and World and most recently to Tyler Burge's theory of perception from Perceptual Entitlement in Perception as a Capacity of Knowledge. 14McDowell's early 'hard propositionalism', however, also brings with it problematic theses that one would not want to follow easily.Relying on Kant, he wants to show that intuition has the same structure as judgment, or turned differently: that sensible synthesis, which Kant calls apprehension, is an act of judgment.McDowell later abandoned this form of hard propositionalism.He now only speaks of the fact that intuition must be conceptually accessible to judgment, but is not itself already conceptually articulated. 15However, he argues the controversial thesis "that the world is conceptually structured" (McDowell, 2013, p. 144, note 18).McDowell is able to show convincingly that he is advocating a form of idealism that is at the same time an empirical realism in the Kantian sense.It is prima facie quite plausible that the world can only be objective in the sense in which we can conceptualize and articulate it.If one reconstructs the world on the basis of the idea that it is everything that is the case in the Wittgensteinian sense, one does not, after all, claim that it is a projection of subjective cognitive structures.Rather, one defends the idea that the objective reconstruction of the world must be conceptual in order to meet the claims of objectivity to the extent that one is expressing truthful propositions about the world and not a purely holding-to-be-true of a subject.But the question also arises to what extent one can refer to Kant with such a 'factual world' at all, for whom the world is considered a regulative ideal.The world in the Kantian understanding always shows -even if not in the sense of nonconceptualism -a surplus, which we cannot catch up by judging.In this sense, the result of the Critique of Pure Reason is not only a limitation of our powers of understanding, but primarily the insight that our understanding of the world reaches further than its cognition sensu stricto. 16rnst Cassirer developed this "surplus" or "more" on the basis of his theory of perception and in the course of this the debate about conceptuality and non-conceptuality of perception can be further undermined and ultimately rejected.

Cassirerian Expressive Perception and the Objectivity of Expression
Ernst Cassirer, as a philosopher of science, addressed the question of the theoryladen nature of observations in scientific experimentation, and in the process anticipated the epistemological holism associated with such names as Quine, Duhem, Hanson, Kuhn, and Davidson.As a result, in his early work Substance and Function he advocates a theory of perception in the sense of a hard propositionalism à la early McDowell. 17According to this, the thesis that intuition always stands under the categories would mean that all content of perceptual experiences is completely conceptually structured.But in contrast to Kant, concepts, for Cassirer, are functional relations that arrange particulars into a conceptual series.Concepts are hence functional concepts or series concepts.For example, we do not judge "The bird flies" because in observation flying is conceptually abstracted from the bird (like for Aristotle), but because we can assign quite different things to the concept 'flying' such as birds, bees, bats, helicopters etc. 18 On the early Cassirer's view, perceptual content always is suchlike structured.But just as McDowell, in the transition from Mind and World to his later writings, went over to loosening this determination in such a way that the contents of perceptual experiences must be merely conceptually accessible and thus linguistically articulable, the Cassirer of The Philosophy of Symbolic Forms analyzed perception in particular in the light of the symbolic forms of language and myth, which in his case led to the abandonment of the thesis of a purely conceptual synthesis.Understanding concepts functionally and no more along the lines of the copula, was just a first step to conceive of concepts, and concept formations symbolically.From this, Cassirer reconstructs the pluralization of world views such as the mythical, the empirical (i.e., the modern every day), and the scientific world view here in the starting point of the distinction between perception of things and of expression.The objectivity of the latter will be the subject of the following.
To say it in a nutshell, Cassirer holds that perception has a double countenance and that its most primitive layer is the perception of expression.Such expressions are gestalts of a figureground organization that are necessarily immersed in moods, feelings and emotions and that stand in kind of a magical or to say the least unstable relation to the perceiver.But not only that: 16 Cf.Hogrebe, 2016, p. 55.  17 Cf.Cassirer, 1923, p. 111, p. 134, p. 245f.18 Cf.Matherne, 2021, p. 53f.  the relationship between perceiver and perceived is above all also interpersonal in nature, which is why Cassirer also calls it thou perception.The objects of expressive perception are hence rather persons than things, although such lines, as we will see, are fluid.The more advanced layer in perception is thing perception.It is constituted by language and hence conceptual.Its objects are stable entities that stand in empirical, i.e., causal relation to each other and the perceiver.Now, in order to link perceptual experiences to forms of knowledge that are located below the (neo-)Kantian judgment of experience and the natural object, Cassirer must show that not only thing perception but also expressive perception has an objective dimension in addition to the subjective component.Against this Cassirer sees two kinds of intellectual opposition, which generally deny the question "is something other than "thing perception" possible at all -is there such a thing as "expressive perception" as an originary-valid function of cognition?"(Cassirer, 2004, p. 70)  19 : "In philosophy: physicalism [,] in psychology: behaviorism" (Ibid.).For both camps it is a pronounced consensus that the experience of colors, feelings, other minds, i.e., the whole experience of the inner life of persons, are purely private, thus subjective, processes.Observable and thus objective are only behavioral patterns and physically measurable things.These assumptions have to be refuted in the following.With Cassirer, on this, we first bring to mind the difference between cultural and natural objects: The entity "Goethe" can, for example, be described as the product of a certain "disposition" [,] this can again be traced back to a certain "hereditary disposition" -in addition to the 'external' conditions, especially the social conditions -grown up in a Frankfurt patrician home, Goethe as a poet of a 'bourgeois culture' (Thomas Mann) and this "bourgeois culture" can again be traced back to certain economic moments -e.g. the development of capitalism -That has its full right -without all this the appearance "Goethe" would not have been possible but even if with it the "causes" to this appearance would be completely shown, if a "Laplacian spirit" would be able to see through it completely, so the "sense" of Goethe, the sense of Buddhism, the "sense" of the Greek language or mythology would not be grasped with it yet -this rather constitutes itself in another mode of cognition -and with another means than that of the "thing perception".[b)] We denote this new mode of cognition by the name of expressive perception -We claim that it is expressive perception that results in a new dimension of reality -and that this "dimension" is characteristic and determinative for all cultural objects -(ECN 5, p. 67f.).
The perception of expression is the starting point in the functional construction of all knowledge and understanding.Even if it opens up a 'new' dimension of reality (i.e.'new' for philosophical reflection) compared to the perception of natural objects, namely that of purport, it is important to note that the natural object, despite its 'empirical reality', in knowledge is also reconstructed via the dimension of meaning mentioned by Cassirer.The intentional orientation towards the object is merely such that the cultural sense is eliminated.The cognition of the natural object is connected with the perception of a real thing, but cognition remains a cultural practice and as such is integrated into a system of meanings.The understanding of meaning, starting off expressively, hence is prior to the cognition of natural kinds.Nevertheless, every cultural object is also a natural object, because the meaning of e.g., Buddhism has to be realized materially by means of scriptures, practices, ritual actions etc. Cassirer thus consistently places the perception of purport alongside the perception qua sensibility.Presentation and representation are intertwined, but precisely not in a conceptual way, but in a symbolic way.And symbols, for Cassirer, are functions of different types, such as expressive, (re-)presentational, and relational types whereas the former is the most elementary type.Meaning in the broadest sense and signification in the strictest sense thus stand on two ends of a spectrum.Genealogically speaking, every cognition hence once has had to have started expressively in perception.The Elasticity of Perception can become a sign and hence understandable in the broadest sense.The skeptical denial of the expressive function cannot be refuted in a strict sense.Referring to David Hume, Cassirer asserts, "All reality is always mediated to us by a-theoretical functions (by 'experiences')" (ECN  5, p. 105).He extends this basis to the expressive experiences and at the same time counters Hume that the latter is subject to a fallacy when he infers from the theoretical unprovability the pure subjectivity of perception, the external world and experiences of other minds. 20The "objectivity claim of "perception" and the experience of expression" is given and therefore "a moment of certainty that goes beyond all pale proof" (ECN 5, p. 107).This is not to say that perception is always objective or a receiving of truth21 , but that the perception of expression directly testifies to a contact with the real, while the objectification of the perception of things leads to entities that are capable of objectivity (and thus fallible), namely propositions such as sentences of natural language as well as mathematically representable laws.The perceptual experience of a material relation, e.g., the tasting of wine together with cheese, is on the one hand a subjective experience (it tastes good to me or not) and on the other hand objective, e.g., in the sense that I can come to the conclusion that wine tastes more intense with cheese, because cheese partly consists of fat, which intensifies the taste.The objectification of thing perception aims consequently at regularities of the 'external world' in the sense of empirical causality.
To the question "Is there an analogon of these general laws also for the expressive function?"Cassirer answers surprisingly at first: "The question seems absurd -because the expressive function seems to be just that which belongs to the circle of the pale subjectivity, the inner world, the only-psychic reality" (ECN 5, p. 108).But the question just seems absurd, "because the opposition: subjective-objective (in the transcendental sense) is a "dignity"opposition -which has nothing to do with the subject area as such -which therefore does not coincide at all with the opposition "physical-psychic"" (ECN 5, p. 108f.).Rather, it is an "error of a naive copy theory" to equate the "opposition inside-outside [...] with the opposition subjective-objective" (ECN 5, p. 109).On the one hand, this is due to the fact that the inner life of human beings is not a solipsistic private event, but expresses and forms itself intersubjectively in social life.The perception of expression is consequently connected with the laws of the intersubjective world of purport: of culture: "In the construction of the "inner world" we also must distinguish sharply between what is objectively-valid and only "subjectively-valid"" (Ibid.).We can illustrate this very clearly with colors: Phenomenally, colors are subjective, objectively, however, they are describable as wavelengths -but not only: For the mentioned reason it would be a fallacy to deduce from their appearance as red, as blue, etc., that they are mental in the sense of sense data and could not be found on external objects.Cassirer concludes: "there is physical-subjective and physical-objective [-] so there will also be psychic-subjective and psychic-objective"(ECN 5, p. 110) -and in contrast to the psychic-subjective he determines the "area of the psychic-objective" (Ibid.)as "the whole field of the so-called "spiritual", of what can be grasped, objectified in "spiritual" acts, what can be represented by language, art etc" (Ibid.).The psychic-subjective is the "non-representable, the mere given 'to me' and known to me alone" (Ibid.)banished by physicalism and behaviorism.In an overly strict sense, a sceptic observer will never be able to judge whether another person really feels pain.Such skepticism cannot be refuted on the grounds of knowledge.But it does not follow from this that we cannot communicate intersubjectively about inner phenomena.Pain, for example, can be represented in art, be it pictorial or literary, and can even be perceived directly in the face of another person.When we speak about color, we speak about colored objects, not about mental sense-data, though we do not know if blue appears blueish to other minds.
Along the phenomenon of expression "a different degree of certainty" (ECN 5, p. 116) can be shown.Cassirer distinguishes "reflexive expression" (ECN 5, p. 117) such as blushing or heartbeat from the perception of more specific signo-cultural expressions.Quite obviously, we interpret the former as "signs (symptoms) of another mind" (Ibid.)and orient ourselves to them in the "interaction from person to person, in a social group" (Ibid.), because we understand them "commonly in all their finest nuances" (Ibid.).Arguably, however, this form of certainty, which according to Cassirer we see "unhesitatingly" as "testimonies of psychic processes" (Ibid.) in the expressive movements of other subjects, is not sufficient to objectively demonstrate the world of intersubjectivity, because they are modes of passive expression.Reflexive or passive expression leaves doubts about a meaningfulness, since it can be proved e.g., also for the tropisms of the plants, to which, however, we want to ascribe the access to the space of the meaning still much less than to the animals.But cultural or active expression, on the other hand, is geistig in the sense that "I am related to another "being", "living" by representation" (ECN 5, p. 125), which is why "the reference to another mind [...] "given" in the pure experience of expression [...] finds its confirmation, its objective "proving" [...] in the construction of a 'cultural world' (language, art ... ) which is 'common' to all [...]" (Ibid.).The objectivity of the cultural world lies in the cognition of supra-individual nexuses, based on realized works in the intersubjective space of cultural exchange.That is why Cassirer also uses the term 'objective spirit' when he speaks of culture.The objectivity of the inner world, one might say, lies in its cultural expression, which is formable and perceptible.And the products of culture are above all objective, because they are realized in material things and consequently possess "an empirical substantiality, which must not be mixed up or confused with any metaphysical substantiality" (ECN 5, p. 126).Cassirer has also coined the term 'symbolic pregnance' for the objectivity of expressive perception, which gives a great deal of insight into what a 'symbolic form' is supposed to be: The manifestation of a cultural, and hence any, object lies not primarily in what constitutes it as a physical object, but in its form, which is objectively perceptible and intersubjectively formable."We do not find and do not seek here any other 'objectivity'" (ECN 5, p. 125).

The Elasticity of Perception
As we have seen, Cassirer's turn to expressive perception is a turn away from his early writings and hence from Marburg neo-Kantianism.One could say that in neo-Kantianism just as in any contemporary form of Conceptualism perception forfeits its status as an experience sui generis since it is investigated in regard to knowledge in a strict sense only.Undeniably, the concept of perception has been an integral part of every theory of knowledge, experience, and cognition since antiquity.But perception as something given is no less suspected of being a myth for Cohen, Natorp, and the early Cassirer than it is for Wilfrid Sellars or John McDowell.Their stance, according to which all conscious perception stands under the categories and thus exclusively has an object for a knowing consciousness, is expressed overly clear by Natorp as follows: "No object is given to us in any other way than in knowledge" (Natorp, 2015, p. 169, my  emphasis).Cassirer breaks with this assumption at the latest in The Philosophy of Symbolic Forms, but then programmatically as I have shown in section two of this paper.In this section I want do deepen the determinations of thou-and thing-perception, present a transcendental argument for adopting the former and combine those views in a higher order definition of perception.To do so, I have to repeat some things already said and remind the reader of the general outlook of Cassirer's The Philosophy of Symbolic Forms.
I have shown elsewhere (Endres, 2020, pp.36-57.) that the Phenomenology of Cognition, the original title of all three volumes of Cassirer's magnum opus, is a Phenomenology of Perception, or rather that those two projects are two sides of the same coin.Two main theses go along with this view, which Cassirer explicitly defends: (1) First, all forms of perceiving, intuiting and thinking are symbolically formed.Cassirer's so-called thing-perception stands thus admittedly under the categories of language and thought.However, there is a deeper layer in perceptual consciousness, which Cassirer calls perception of expression or thou-perception and which does not aim at enabling true propositions about the object of cognition, but which forms its objects emotionally, gestalt-like and truth-indifferently. (2) Secondly, there is a genealogical connection between perceiving, intuiting and thinking as expressed in the symbolic forms myth, language, and science.Perception is the most basic form of objectification, even the expressive perception.Pure thinking by contrast, as expressed in modern science, has completely detached itself from perception.The constancies of perception are no longer a prerequisite for formation of invariants in, for example, modern geometry or physics.Nevertheless, in the creation of pure relations on which these theories are based, thinking has had a model in perception and has overcome it exactly where neither perception nor language can no longer provide indications for modern scientific thinking. 22Cassirer put this thesis most sharply as follows: The function of simple sensation and perception is not merely "combined" with the basic intellectual functions of comprehen sion, judgment, and inference but is already such a basic function -it implicitly contains what emerges there in conscious forming [Formung] and in independent configuration (Cassirer, 2021, p. 278).Now, I would like to show how Cassirer justifies the concept of perception as thingand thou-perception transcendentally and also how the transcendental method leads him to a more global determination that encompasses them both.First, we need again to look at what expressive perception is and what justifies its adoption.Expressive perception is a depth layer of perception that is intentionally directed not to things, properties, and empirical causality, but to moods, affects, emotions, figures, and mythic-magical cause-effect relations.Cassirer gains this determination from the analysis of myth, whose characteristics he determines as (1) polysynthetic, (2) fluid, (3) metamorphic, and (4) magical in terms of causality.In short, this means that in principle everything in myth can hang together with everything and that in principle everything can be caused by everything.Our today's everyday understanding of genuine truth and falsehood, of being and appearance, of waking and dreaming, and so on does not find any application in myth.
The main argument for deducing thou-perception from myth is: If empiric and scientific thinking would not be an expression of a special direction of perception -namely of thingperception -and would myth not originate from thou-perception -if there would be only thingperception -then the way of thinking, the way of intuiting and the way of living myth would not only violate our empirical thinking, but also the world of perception of humans living in myth.The theoretical contradiction becomes a practical one.For example, a member of the Dieri tribe in South Australia, sees in the head of the tribe the incarnation of a certain plant.A Huichol Indian sees a deceased ancestor in a deer.This does not mean, of course, that the perceptual performance of man on the level of mythical consciousness has no stabilizing functionality.Of course, he can individuate plants, animals and other human beings, i.e. perceive them as entities existing for themselves.However, they are not distinguished genus-wise for him.The noticing of an animal can mean the encounter with a deceased ancestor, and according to Cassirer's presuppositions this also means that the phenomenon of perception appears to the human being in exactly this way.The boundary lines that we draw through our empirical concepts of genus and species shift and evaporate to man in myth.This quality of the ephemeral would not be comprehensible, if already the direct perception, as it is present in expressive perception, before the formation of things through language, would include "the division and partitioning of the world into fixed classes" (Cassirer, 2021, p. 71).Ethnological, linguistic, and religious studies suggest, for Cassirer, that perceptual consciousness cannot be limited to fixed genera and classes in the sense of linguistically formed perception of things.If one hence wants to be able to explain the real actions of people who live in myth, one must consequently describe their perception as 'mythically pregnant' along different than linguistic categories for the sensible synthesis and as accessible to expressive perception.According to this, the objects of thou-perception are gestalts, which by themselves reveal different expressive values.On the one hand, they are hence fixed in the sense of the supersummativity of a melody, the figurebackground contrast, outlined forms, or in general: as individuated wholes.On the other hand, they burden human experience with the continuous collapse of stability by the life of affect.In myth, a change of direction or intensification of feeling is accompanied by the possibility of a change in the objective view.If the emotional ground of the subject changes, then the entire atmosphere in which the objects are shrouded changes at a stroke.A certain place may therefore be accessible by day, but off-limits by night.Just as in our perceptual world things take on a different color as soon as they move into a new illumination, so in the magical-mythical view objects change their nature when they are struck by a different ray of affect.This dynamic is only gradually pushed back by thing-perception.In the earliest forms of social life, man establishes rules to provide stability to the dynamics of affect beyond the perceived individuals.These limits provide a relative firmness, as long as man moves in the circle of everyday life.Furthermore, the development of language plays an essential role for tightening this firmness.
The objects of thing-perception, on the other hand, reveal in the light of language, instead of affect, their determining characteristic.Thing-perception sets certain limits in order to fix them as such.This does not mean that those objects would be fixed once and for all by being named for the first time.Cassirer's point here is that, in contrast to the hardly controllable thou-perception, thing-perception knows a law of continuity which is the key to understanding the connection between perception and concept: the new perception does not appear abruptly and, as it were, eruptively; it joins the earlier ones and tries to balance itself with them.That is, thing-perception does not make a break with thou-perception in the construction of the empirical object, but establishes its boundaries in relative agreement with the object of thouperception.This process can be conceptualized as an equilibrium of drawing and dissipating boundaries.From the relative agreement of the double countenance of perception the image of the empirical object is built up.This leads to the following conclusion and definition: Perception must perform a double and, at first sight, paradoxical feat by allowing the impermanence of expressive experiences and bringing it into balance with the stabilizing accomplishments of thing perception.For this, it must be, as it were, fixed and mobile at the same time.It overcomes this contradiction by an elasticity peculiar to it.In creating constancies, perception is thus elastic.

Undermining the Debate
This fourth and concluding section takes, from Cassirer's views on perception, consequences for the (non-)conceptualism debate in contemporary Kantianism and the current philosophy of perception.As we have seen in section one, this debate mostly draws on the heterogeneity thesis, i.e., the distinct status of concepts and perception in regard to (1) their 'fitting', (2) different roles or functions they do fulfil, such as representing singular or general content.My aim is to undermine and reject the entire presentation of the problem by showing that, in relating aspects of non-conceptualism to the perception of things instead of that of expression, it is ill-posed.
As we have seen, there are good arguments for even hard propositionalism, that is, the (idealist) view that the world is essentially a factual world and concepts are ultimately real because they structure the world as it is, as well as its perception.If we assume that there is objectivity only in the sense of true statements about the world, propositionalism seems correct with respect to the philosophy of language and perception.Starting off from the objective givenness of perception, the counterposition cannot show how perception and concept can correspond functionally, i.e., epistemically.Propositionalism, again, must assume that the content of perception is at least potentially conceptual, and it leaves no room for aspects of our world that escape the rationality of judgment and concept in the empirical sense.
With Cassirer, we can, on the one hand, catch up with the conceptualist thesis that every perception, as McDowell puts it, must be structured cognitively "all the way out" (McDowell,  1996, p. 69) in order to be a conscious perception about something, i.e., to have content or purport.On the other hand, with Cassirer, we can also claim that this way of structuring perception is not necessarily conceptual, but symbolic.With the latter claim we open the road for non-propositional aspects of perception that do not run either into the problem of what Thomas Pendlebury calls accidentalism, i.e., the problem of not being able to show that there is a necessary connection between the way we perceive things, if correctly perceived, and the way we bring this experience into concepts, nor into the problem of non-conceptualism, i.e., the problem of not being able to show how perception is accessible to concepts.
If we assume that both the perception of things and that of expressions are perceptual experiences sui generis, then the whole (Non-)Conceptualism debate takes another shape.We can then identify the (a) fineness-of-grain argument as an erroneous attempt to apply non-conceptual aspects of expressive perception to thing-perception and thus reject this attempt as a category mistake.If we understand perception in contexts of empirical considerations, i.e., that we for instance want to know under which empirical conditions, regarding illumination, perceptual mechanisms etc., a tomato appears light or dark red or even black, then it does not need to be a problem if we concede to the conceptualist claim that perception in this experimental scenario is conceptually structured; if and only if we also concede that there is another mode of perception where all this does not apply and where perception rather is as what it appears to be.In the latter case one can simply state 'The object appears reddish', or 'The object appears threatening', or 'The object appears to tell me something I don't know what', or 'The object appears to tell me that the sun is rising soon' etc.The point would be that empirical thinking, the presupposed link between perceptual states and knowledge claims has to be relativized in favor of a view that makes room for perceptions that stand in a larger context, i.e., in a context of understanding, of feeling, of interpreting the world rather than of sensu stricto knowing it.From this follows that perception has purport only in that its object is present in relation to a mode of representation, or, as Cassirer also says, symbolic formation.The Non-Conceptualists' either-or regarding (b) objects being presented or represented in perception is just another false alternative.And with Cassirer we can furthermore and finally address objection (c) in saying that (some) animals and humans in pre-linguistic age do master symbols at the pre-linguistic stage which makes the claim for a non-symbolic layer of perception obsolete.
Perception can tell us a lot.It gives us a feeling of situatedness, it allows us to notice the finest grammar of our world, it enables us to grow our senses into language and it is itself a model for objective knowing.In all that lies no contradiction since perception is elastic.Through perception we have access to the vaguest forms of noticing and understanding change in our environment and in the expressive behavior of our fellow human beings and also in that of animals.We undeniably can get into other minds and understand those and ourselves through perception in its interplay with our cognitive faculties, particularly with the symbolic ones that are at interest here.This entire field of human experience cannot be reduced to facts or propositions, since it exceeds the space of true or false statements like 'I perceive that the apple is red'.Conceptualism has no conceptual means to address those phenomena.Cassirer on the other hand, by beginning with the larger concept of the symbol, can encompass aesthetic expressions, linguistic expressions, purely conceptual expressions etc.His genealogy of symbolic expression beyond that makes the important point that our ordinary concepts of everyday language and scientific concepts such as 'field', 'electron', 'force', 'matter' etc. are not at all on the same line, but still in some way linked.To be precise, concepts of modern physics, e. Cassirer conceives of as being purely relational, having originated from perception as a model of objectivity, but presupposing nothing at all that we can find in perception.This is a problem that we find never addressed in current philosophy of perception and in Conceptualism.We can therefore say that Conceptualism absolutizes concepts in a manner that neither does justice to non-propositional experiences such as 'simple seeing/hearing…', the perception of moods, vague meanings etc., nor to concepts in a demanding sense, since no one within the philosophy of perception, as far as I can see, investigates the differences, but also nexuses between distinct concept formations.Non-Conceptualism on the other hand absolutizes perception as some kind of pure givenness that rightfully has been criticized more than once.No perception can be attributed to myself as having this or that perception if it is not by some means 'intellectually' mastered, i.e., with Cassirer, symbolically structured and hence an objective experience for a subject.Furthermore, the three objections against Conceptualism raised by non-Conceptualism amount to (a) a category mistake in the sense that non-conceptual perceptual experiences are measured against a mode of perception that is constituted conceptually, to (b) a conflation in connection with the relationship between being present and being represented, and finally to (c) an unfounded idea of a brute, non-symbolic perceptual layer in infants and animals.

Conclusion
Cassirer's philosophy of perception provides the conceptual tools to undermine an important strand within the current state of discussion in the philosophy of perception.The debate between Conceptualists and Non-Conceptualists is an impasse with no way out since the myth-of-the-given-objection absolutizes the concept of the concept in a manner that does not encompass all our symbolically structured experience and since the fineness-of-grain-objection absolutizes the senses in their givenness.Cassirer's distinction between expressive and thing perception, on the other hand, gives rise to elastic perception which can deal with a much larger phenomenal domain than Conceptualists and Non-Conceptualists thematize and by bringing very different perceptual experiences in perspective.