The Metaphysics of Appearance in Republic X (596a5–598d7)

Abstract

Plato's Republic X attack on imitative poetry is based in the metaphysics of appearance, since appearances are the objects and products of imitation. I offer a new reading, showing that Plato's account coherently introduces appearances as a new type of item, distinct from Forms and sensible particulars, and applies beyond imitation to a broad range of appearances. Focusing on the importance of perspective to Plato's reasoning, I argue that an appearance is a relation that comes about between a material particular and an apprehending subject. Ordinarily, appearances are transparent: they confer determinate awareness on the subject, but are not the objects of our awareness insofar as we are appeared to. This reading resolves longstanding obscurities, grounds an improved account of imitation, and shows that Plato here presents the cornerstone of a general theory of appearance.

Introduction

plato's republic x attack on imitative poetry is based in the metaphysics of appearance. In the first stage of the argument (596a5–598d7), Socrates sets out from a restatement of the Theory of Forms to give a general account on which imitation has appearances (phantasmata) or images (eidola) as both its objects and products—what it seeks to imitate and what it makes.¹ Imitation is "far removed from truth" because appearances are themselves three removes from reality and truth (598b6, 597e3–4, 602c1–2).² Though scholars take Plato's ban on imitative poetry seriously, they often regard the metaphysics on which it is based as a theoretical kludge—a set of ideas jerry-rigged for the attack on imitative poetry, but which are incoherent or cannot sustain systematic application. Scholars are leery of the notion that Plato here introduces appearance as a new type of item in his ontology, distinct from Forms and sensible particulars, or that his characterization applies to appearings that occur outside the context of imitation.³

Some wariness stems from the impression that Plato's approach in this passage is anomalous. Socrates's presentation of the Theory of Forms seems to diverge from its standard formulation, leading some to doubt that the argument reflects Plato's settled view.⁴ Less explicitly, scholars seem to regard a metaphysical account of appearance with apprehension, since the topic is typically considered epistemologically. On this more common approach, appearance—phantasia—is a distinct type of cognitive state between sense-perception and reasoned belief that may include hallucinations, illusions, dreams, hedonic and affective states, and perhaps a range of propositional apprehensions lacking full rational assent. Questions germane to this intermediate position concern how appearance differs from sense perception, whether it involves conceptualization, whether the content of appearances can be propositional, and whether appearances constitute doxa (beliefs). Focusing on these questions, interpreters of Plato's theory neglect the opening stage of the book X argument, looking instead to the so-called "cognitive partitioning argument" (602c4–603b5), to the book VII discussion of summoners (523a1–524d5), or to later dialogues.⁵ Metaphysics, it seems, has little to teach us about appearance.

In any case, readers of the book X argument are primarily concerned with the aesthetic, ethical, and psychological dimensions of Plato's attack on imitation, and with the challenge of integrating this argument with other passages. Constrained by assumptions about the objects and products of imitation, these interpreters often tolerate obscurity in Socrates's presentation. For instance, there is an unexplained equivocation in Socrates's characterization of the product of imitation. Socrates sometimes speaks of the imitator's creations, and imitators themselves, in ordinary terms, for instance, as a couch (596e10) or a carpenter (598b9), respectively. By this direct characterization, he seems to designate the items that appear in imitative works—such as the characters in a play—and which are thought and spoken of directly as instances of their kinds. At other times, Socrates says that the imitator makes an appearance or image (phantasmatos [598b3]; eidolon [598b8]), terms that designate something distinct from the item of which it is the appearance, and insubstantial (smikron ti [598b7]) compared to it.⁶ The mere image of a couch, after all, is not a couch. But if the imitator has only one product, as most assume, these characterizations must refer to the same thing. Consequently, Plato seems to conflate distinct conceptions of the imitator's product and, by extension, appearance itself.⁷

Despite this obscurity, there is a standard view of what appearance is in this argument. This view starts from a conception of imitation as likeness-making. The work of the imitator is to produce a material particular artwork, such as a painting, which bears some resemblance to the item being imitated, which is also a material particular. Appearance is the crucial link in this process, the feature copied from the object being imitated and conferred on the item created by the imitator. Thus, interpreters arrive at what I will call the feature reading, on which an appearance is an intrinsic feature—a property or part—of a material particular.⁸ Such an account, however, does not clearly apply to forms of imitation that do not involve likeness-making,⁹ or to any type of appearing that depends on the perspective of the subject appeared to. Several appearings of this kind are invoked prominently in connection with appearance in the Republic: illusions (382a1–e11, 602c10–d4), subject-relative hedonic experiences (583b–584a, esp. 584a7–10), and affective states such as grief (603e–604c). In this way, the feature reading implies that Plato's book X account of appearance is not intended for application beyond imitative contexts. More problematically, the role of perspective is central to the argument by which Plato introduces appearances, as such, to the discussion. Besides offering an account that fits only a narrow range of appearings, a feature reading does not accommodate Plato's reasons for distinguishing appearances as items in their own right.

On the view presented here, considerations of perspective are key to understanding why Plato gives a metaphysical account of appearance. Perspectival appearing shows that we cannot explain how things appear just in terms of sensible particulars and their features, or souls and their states. We need to examine how these elements interact. Appearance is a relation that typically comes about between a material particular and an apprehending subject.¹⁰ This relation is content bearing, conferring determinate awareness of some kind of thing or other.¹¹ In this way, as we shall see, Plato's account builds on his established metaphysics. An appearance is not a sensible particular, nor a part or feature of one, nor a state of the soul, nor a Form. Nevertheless, an account of what an appearance is invokes them all.

Besides providing an account suitable to a broad range of appearings, this reading also illuminates Socrates's dual characterization of the products of imitation. Ordinarily, appearances are transparent—they are the conduits, not the objects, of our awareness insofar as we are appeared to. Anyone subject to the appearance of a couch experiences the awareness of, or as of, a couch present in their environment. Accordingly, we must distinguish what appears, or comes to be represented in the subject's awareness, from appearance itself, the relation by which that awareness has come about. Socrates's two characterizations of the imitator's products correspond to these distinct elements and are clearly distinguished by their place in the investigation. The appearing things created by the imitator—fictional items that appear in a work of imitation—are products of imitation as we experience them ordinarily. Appearances, by contrast, are content-bearing relations generated by the imitator, and through which we come to apprehend the items that appear in the imitative work. But we come to recognize appearances as such only through philosophical inquiry.

This interpretation will emerge from a close reading of Socrates's general account of imitation developed in the first part of the argument banning imitative poetry (596a5–598d7). Along the way, we will confront assumptions that have constrained prior interpreters and consider the ramifications of Plato's account of imitation. Though the paper's attention will be narrowly focused on these interpretive matters, I hope that my reading will support reconsideration of Plato's aims with respect to appearance. At issue, ultimately, is the place of appearance in the synoptic philosophical vision of Plato's Republic. Interpretations on which Plato's theory of appearance is incoherent or ad hoc prohibit consideration of a role for appearance beyond the attack on imitative poetry. In contrast, the reading offered here is motivated by the view that appearance plays a systematic role in Plato's account of prerational beliefs and affective states, experiences that Plato regards as forms of representational awareness that form spontaneously, passively, and without active reasoning on the part of the subject. Nevertheless, for Plato, the way that the world strikes us in these experiences is thoroughly subject-relative, shaped by the interaction between our perspective and the world. An account of these fundamental forms of awareness must therefore begin with a metaphysics of appearance as a relation.

1. the imitator's creations (596a5–596e11)

Socrates develops a general account of imitation by resolving a puzzle about the imitator: how can an imitator be the creator of every sort of thing? To pose this puzzle, Socrates frames the imitator's products univocally as things that appear in imitative works, fictional items that we speak and think about as we do the objects of ordinary experience. Though some find the notion of a mere appearance in this introduction, Socrates comes to speak of appearances and images as such only later, in the theoretical account that solves the puzzle of the imitator.¹²

Socrates first investigates what sort of craftsperson the imitator is, focusing this question on what an imitator makes. After a brief, controversial restatement of the Theory of Forms (596a5–b2),¹³ Socrates describes how an ordinary craftsperson makes many instances of an artifact by referring to the Form (596b4–8). He then introduces another craftsperson who can make "everything [panta . . . hapanta . . . hapanta]" (596c5–9; cf. 596d4), likening this amazing (thaumaston [596c3, d1]) creative ability to the way one holding a mirror could "quickly make the sun and the things in the sky . . . the earth, [oneself] and the other animals, manufactured items, plants, and everything else" (596b10–e3). Glaucon suggests that the mirror-holder makes these things appear, but not be in truth (phainomena, ou mentoi onta ge pou alêtheia [596e4; cf. e11]). Praising Glaucon, Socrates adds that a painter—henceforth his model imitator—creates the same sort of items (tôn toioutôn gar . . . dêmiourgon kai ho zôgraphos estin [596e5–6]).

How should we understand these creations? Prior to the mirror illustration, Socrates emphasizes two points. First, the remarkable craftsman is a creator. Socrates deploys a flurry of terms for craft producers (dêmiourgon [596b12, d3], cheirotechnôn [596c2, c5], poiêtês [596d4]) and production (poiei [596c2, c5, c6, d5, e1], ergazetai [596c7, c9]). Pace Stephen Halliwell, I see insistence, not irony, in this repetition: the amazing craftsperson brings something new into being.¹⁴ Second, Socrates stresses that the items created by the imitator are, in some sense or other, akin to those created by ordinary craftspeople. The amazing craftsperson "makes everything, all of the things each individual handicraftsman makes [hos panta poiei, hosaper heis hekastos tôn cheirotechnôn]" (596c2).¹⁵ In the mirror illustration, Socrates refers to the items created in the same terms used to speak of ordinary items: "You will quickly make the sun and the things in the sky . . . the earth, yourself and the other animals, manufactured items, plants, and everything else" (596e1–3). The amazing craftsperson creates a set of items spoken of directly and without qualification as instances of their kinds.¹⁶

This characterization is readily explained by reference to a familiar facet of our engagement with imitation. Looking at a painting, we often speak directly about the things that we see in it, saying, for instance, "That's a couch." Likewise, in engaging with literature or drama, we say many things directly about the fictional characters depicted. In both cases, we assign additional features that can belong to these items only as instances of their designated kinds. So, we might say what a couch in a painting is made of, where the same description cannot hold of the painting itself. That we engage with imitations in this way is vital to Plato's critique: Socrates later describes our response to events in tragedy as "enjoyment of other people's suffering" (606b5–8; cf. 393a8–b2).¹⁷ Anticipating this, Socrates frames the imitator's products initially as fictional items that we speak and think about directly, just as we do ordinary items.

This explains the controversial emphasis on language in Socrates's earlier remarks, which seem to assert that there is a Form corresponding to every meaningful general term: "We usually posit some one particular Form in connection with each set of many things to which we apply the same name [eidos gar pou ti hen hekaston eiôthamen tithesthai peri hekasta ta polla hois tauton onoma epipheromen]" (596a7).¹⁸ This is problematic, though, if each Form represents a real kind, a single property genuinely shared by many instances, since not every common general term signifies a real kind. But Socrates presents this as a posit appropriate to the beginning of an investigation (arxômetha episkopountes . . . tithesthai [596a5–7]). At the start of inquiry, we have no better guide to identifying real kinds than our ordinary linguistic terms. Each general term provisionally names a kind appropriate for further inquiry, though our investigations may discover that some terms do not designate real kinds.¹⁹ Thus, we need not take Socrates's formulation to say that there is a Form for every general term. Rather, Socrates's emphasis on language serves a different purpose: it prepares us to think of the imitator's creations as items that we speak and think about in the way that we do ordinary objects. Specifically, we predicate properties directly of these items, regarding them as instances or bearers of the properties named by our general terms.

Some commentators derive a conception of the imitator's products as mere appearances or images from the expressions that bracket the mirror illustration.²⁰ Introducing the illustration, Socrates says that the mirror-holder comes to be the creator of items in the mirror "in one way, but not in another way [tini tropôi genesthai an toutôn hapantôn poiêtês, tini do ouk an]" (596d3–4). Afterwards, Glaucon explicates this idea by saying that the mirror-holder makes these items "appear but not be in truth" (596e4). On a common reading, Glaucon here specifies that the mirror-holder produces the mere appearance of things, but not the things themselves.²¹ It is implausible, however, to take Glaucon to invoke a distinction of this kind, using this language, at this point in the discussion. For, whatever an appearance is, it is not an instance of the kind of item whose appearance it is.²² Accordingly, to the extent that we recognize an appearance as an item in its own right, distinct from a particular instance of a kind, we will not speak about it as such an instance. For example, we will not say of the mere image of a couch that it is upholstered in leather, or that three people are sitting on it, though we would say that about a couch depicted in a painting. Thus, taking Glaucon to say that the mirror-holder makes the mere appearance of things requires him to break from Socrates's characterization of the imitator's creations. Glaucon's terminology does not support this. He uses the neutral plural participle phainomena (appearing things), the antecedent of which is panta hosa nundê elegeto (596e3), a phrase that refers to the list of items created by the mirror-holder (596c4–9). Read literally, then, Glaucon says that the mirror-holder makes those things appear, but not be. A moment later, as if for emphasis, Glaucon use the same language to say that a painter makes an appearing couch (klinên . . . phainomenên [596e9–11]). It is farfetched to suppose that simply by this language Glaucon introduces a substantially different conception of the imitator's products as mere appearances or images, where these are a distinct category of items. More plausibly, Glaucon is working within the parameters of Socrates's initial characterization and grouping to signal the difference between the imitator's products and the material instances that we encounter ordinarily. Yes, the mirror holder makes animals, plants, and artifacts, but they are only appearing animals, plants, and artifacts. But it is not yet clear what that means.

It may seem that I am pressing a false dichotomy in insisting that one can speak of the imitator's creations as instances of their kinds or as mere appearances, but not both at the same time. Perhaps Glaucon's insight is that the creation of a fictional appearing thing involves nothing more than the generation of a mere appearance, as heat is analyzed as the mean kinetic energy of molecules. Plato's view will turn out to be something along these lines. The problem is that a reduction of this kind depends on a theory analyzing the entity construed pretheoretically into the theoretical entities that underlie it. So far, though, Glaucon and Socrates have not even mentioned appearances as such, much less explained how the things that appear in imitative works depend on or consist in them. In short, we must distinguish the characterization involved in presenting the philosophical problem from the one that grounds the solution. Socrates will eventually come to speak of appearances and images as such. In so doing, he will take himself to explain the imitator's creative abilities: "The reason that imitation produces everything . . . is that it touches only something small of each thing, and that is an image [dia touto panta apergazetai, hoti smikron ti hekastou ephaptetai, kai touto eidolon]" (598b7–8). But, as we will see, isolating appearances as entities in their own right is a theoretical achievement, the result of investigation into the problem Socrates is now posing. For this purpose, he frames the products of imitation univocally as appearing things, items depicted in imitation that we speak and think about just as we do the objects of ordinary experience. Before we can say what it means to be something that appears but is not in truth, we must first determine what an appearance is.

2. the object of imitation (597a1–598a1)

A preliminary discussion is needed. Socrates comes to speak of appearance in its own right through reflection on the object of imitation. In fact, Socrates speaks of what we might call the object of imitation in two distinct senses. The first refers to the familiar item that imitators take as their subject matter: a couch, a vase of flowers, a carpenter. I leave it open, for the moment, whether this item is an already existing particular or a generic representative of a kind; one can draw a chair without drawing an existing chair. I will call the object of imitation in this sense the item being imitated. The object of imitation may be spoken of more narrowly as that element in or of the item being imitated that the imitator specifically seeks to imitate (epicheirein mimeisthai [598a3; cf. 598b1–3]). I will call this narrower object the focus of imitation. Socrates says that the imitator seeks to imitate only the appearance and not the being of the item being imitated, illustrating this by reference to the way a couch appears differently from different perspectives. Thus, appearance is delineated in its own right specifically as the focus of imitation. But since the focus of imitation is itself an element in or of the item being imitated, we must first get clear on what that is. Most readers assume that the item being imitated is an already existing material particular and, consequently, that appearance must be a feature of a material particular. I will argue here that the item being imitated is not a particular, but a kind.

There is an important shift in Socrates's presentation of the imitator after the mirror illustration. Socrates proceeds to set out a hierarchy of three "sorts of couches." These are the Form of Couch, a material particular couch, and an appearing couch. Each is paired with its own maker: god, carpenter, and painter, respectively (596e–597b). This leads to a surprising new description of the imitator's work. For now, in contrast to the prior characterization of the imitator as a creator, Glaucon emphatically denies that the painter makes what the carpenter produces (oudamôs [597d10–12]). Socrates asks, "What would you say he is of a couch [Alla ti auton klinês phêseis einai]?" (597d13), and Glaucon answers that he is the imitator. Now, whatever item is referred to by the term 'couch' (klinê) here, it must be distinct from the one that the painter was previously said to create. Additionally, that item, which the painter imitates, must be the same as the item that the carpenter makes, and the item whose nature is created by the god. Thus, Socrates has redescribed the three makers of different couches as engaging in different kinds of work with respect to the same entity. With respect to that item, the god is the nature-maker (phutourgon [597d4]), the carpenter is the crafter or producer (dêmiourgon kai poiêtên [597a8]), and the painter is an imitator.

To identify this shared object, consider that Socrates will soon go on to ask whether the painter strives to imitate "each thing itself . . . in its nature . . . or the products of the craftspeople" (598a1–3).²³ Socrates's question is whether the imitator seeks to imitate the product of the god or the craftspeople, the Form or material particulars. This question would be redundant if it had already been specified whether the painter imitates material particular couches or the Form. So, in his earlier statement, Glaucon cannot be saying either that the painter is the imitator of some existing particular couch, or that he is the imitator of the Form. Rather, in asking what the painter is with respect to "a couch," Socrates refers to the kind Couch without differentiating the Form from its particular instances.²⁴ To be the imitator of the kind Couch is to engage in a different sort of work from either the god or the ordinary craftsperson. The god makes the nature of the kind, that is, the Form. The carpenter crafts the kind—he is a Couchmaker—insofar as he produces many material instances of that kind (cf. 596b4–8). And the painter imitates the kind by producing potentially many appearing instances of it.

Identifying the item being imitated as a kind does not require that we introduce a new entity to Plato's metaphysics. By 'kind' I mean nothing more committal than our ordinary conception of a category or class, whatever we take ourselves to be speaking of when we characterize a type of item generically: "A couch is a cushioned seat for two or more people." Plato centers the discussion on kinds to capture the imitator's orientation toward his or her object. Since imitators are not philosophers, their construal of a thing being imitated will be metaphysically innocent. In this way, Plato captures the fact that imitators primarily seek to depict an item of some type, not necessarily an already existing particular. It is consistent with this that an imitator sometimes seeks to depict specific fictional characters, such as Chryses,²⁵ or already existing individuals, such as Socrates in Aristophanes's The Clouds. For, even in these cases, the imitator is not interested in every feature of the person imitated or in representing the person under any description whatsoever. Rather, the aim is to depict that individual in a specific light, where this may involve a collection of properties. The imitator's goal is for the items depicted to be apprehended by the audience as items of some intended kind or kinds.

If the painter imitates the kind, how should we understand Socrates's remark that the imitator seeks to imitate "the works of the craftsmen" (598a3)? This phrase seems unambiguously to designate existing material particulars. The key point is this: kinds and material particulars are not mutually exclusive candidates for the item being imitated. When the imitator attends to existing material particulars, he regards them as models of a kind. This is shown by Socrates's repeated use of the term hekaston. Socrates asks, "Do you think it is what each thing itself is in its nature that he is trying to imitate, or the products of the craftsmen [potera ekeino auto to en tê phusei hekaston dokei soi epicheirein mimeisthai]?" (598a2–3). On a common reading, Socrates's use of hekaston, "each thing," ranges over the individuals a painter selects, such that the imitator imitates a single existing material particular on each occasion.²⁶ This reading is undermined, though, by Socrates's use of the plural in speaking of the "works of the craftsmen [ta tôn dêmiourgôn erga]" (598a3).²⁷ There is no reason why a painter would look to the works of multiple carpenters to imitate a single existing couch.²⁸ Rather, Socrates is asking whether, for any kind that the painter seeks to imitate, he attends directly to the nature of that kind, the Form, or to several instances of it that have been produced by craftspeople, regarding them as models of the kind to be imitated.²⁹

Accordingly, when Socrates next asks whether the painter imitates the works of the craftspeople "as they are or as they appear" (598a5),³⁰ the object of imitation has not shifted from kind to particulars. Rather, Socrates here advances a distinction between the shared being of instances of the same kind, and their shared appearing as instances. Does the painter seek to imitate particular couches with respect to how they are couches, or with respect to how they appear as couches? To explain this, Socrates points to the way a couch looks different as we move about it. We will consider this couch illustration in detail in the next section. For now, it is sufficient to note that Socrates employs the same use of hekaston immediately afterwards, asking again, "At what does painting aim in each case [hekaston]? To imitate what is as it is [to on, hôs echei], or what appears as it appears [to phainomenon, hôs phainetai]?" (598b1–3). Since hekaston ranges over the kinds that the painter seeks to imitate, Socrates's use of to on and to phainomenon in this question are incomplete and need to be filled in by the specific kind at issue on any occasion. Does painting strive, in each case, to imitate what it is to be F, or what it is to appear F?

This account illuminates why Plato reminds us of both Forms and their material particular participants at the start of the argument. Appearing, for Plato, is determinate representation; all appearing is appearing as some kind of thing or other. The imitator's products are characterized in every case as some specific kind of thing or other. Thus, every appearing must be classifiable by reference to some definite kind or category.³¹ Forms are the essences or natures defining these determinate kinds. But the appearance of a kind does not directly involve the Form. When a kind appears ordinarily—as in our recognition of a couch across the room—it appears to us in its material particular instances. That is, the kind F appears to us when we recognize material instances as Fs. Accordingly, an account of appearing must include reference to material particulars as participants in Forms, since they are the typical mediators of appearing. Imitators come to their work intending to imitate a specific kind of thing, the broad object of imitation. In considering this object, they ask, In virtue of what do instances of this kind come to appear as instances? The answer to this question is the focus of imitation: the appearance of F.

3. appearance as a relation (598a1–b8)

Let us now consider directly what an appearance is. Glaucon does not immediately understand the distinction between how things appear F and how they are F. To explain, Socrates offers what I will call the couch illustration: "If you look at a couch from the side or the front or from anywhere else, does it differ in any way from itself? Or, while not differing in any way, does it appear in different ways?" (598a7–9, translation adapted from Reeve). As a result of this illustration, Socrates concludes that the focus of imitation is an appearance (phantasmatos [598b3]), speaking of appearance in its own right for the first time in the argument. To emphasize this new terminology, Plato has Glaucon immediately repeat the term, and a moment later Socrates introduces an equivalent, eidolon, or image (598b5–8). Drawing on affinities between this example and a passage in the Theaetetus, I will argue that appearance is not an intrinsic feature in a material particular, as a feature reading proposes, but is instead a relation that comes about in between a particular and an observer.

In broad strokes, the reasoning of the couch illustration entails that there is a difference between what a couch is (i.e. its being) and how a couch appears.³² Since the former remains unchanged while the latter changes, they must be distinct. If this were the entirety of Socrates's reasoning, the appearance of the couch might be construed as a part or property of the particular distinct from the property that constitutes it as a couch. But Socrates's illustration also specifies that the couch appears differently as the observer moves about it, even though the couch does not change "in relation to itself [mê ti diapherei autê heautês]" (598a8; cf. 598a10). To see the significance of this detail, consider a similar distinction from the Theaetetus. There, Socrates asserts that the Protagorean measure doctrine entails that "there is nothing which in itself is just one thing," but instead, all things "are in process of coming to be" as a result of their interaction (Theaetetus 152d2–8).³³ To explain this, Socrates explicates the difference between what something is in its own right (auto kath' auto [Theaetetus 152d3]) and what it comes to be in relation to something else. He provides two illustrations, the first involving color (Theaetetus 153e–154a) and the second involving the relative size of groups of dice (Theaetetus 154c–155c). In each case, a subject comes to be different in some respect despite no change in itself (auto ge mêden metaballon [Theaetetus 154b3]; cf. auto kath' auto [Theaetetus 152d3, 153e4–5, 157a8]), but merely through "movement and change and blending with one another [phoras to kai kinêseôs kai kraseôs pros allêla]" (Theaetetus 152d7–8; cf. 154b1–6). Consequently, the respect in which the item is altered does not belong to the item by itself (Theaetetus 154b1–6). In this way, Socrates forwards a distinction between what holds of an item in its own right, or intrinsically, and what consists in the item's relationship to something else.³⁴

The couch illustration displays the same reasoning. Since the couch comes to appear differently without any change in itself, the ways that it appears, or its appearances, do not belong to it intrinsically, itself by itself. To see that a feature reading does not accommodate this line of reasoning, consider the most fully articulated account of this kind. According to Jera Marušič, the appearance of F is a property G distinct from and non-essential to being F, but associated with F in such a way that items possessing G are "constituted as representations or figures of F."³⁵ For instance, the property of moving one's arms and torso in a certain way is not essential to rowing, but is associated with rowing such that one moving in that way may represent or depict rowing to viewers. On Marušič's view, this motion just is the appearance of rowing, a feature held in common by actual rowers and those who impersonate them. A person can have this feature, however, without actually appearing to anyone as rowing. Imagine a person moving her body in this way consistently over a period of time, say, in a game of charades. The guesser sees other activities in the presenter's motion—"You are doing sit-ups! No, you are covering yourself with a blanket!"—before eventually guessing correctly. On a feature reading, since the presenter moves in the same way the entire time, she has the appearance of rowing the entire time. This motion is one of the respects in which the presenter does not change. By contrast, according to the reasoning of the couch illustration, this motion cannot be the appearance, since the motion persists while the presenter appears differently to the guesser.

If this is right, the couch's appearance cannot be identified with any feature that persists in the couch while it comes to appear in different ways. Thus, it cannot be identified with the surface of the couch, or stable superficial features, nor with the complex disposition to look thus to one viewer and so to another. Indeed, it is misleading to construe the couch as having a single appearance, different parts of which manifest themselves to viewers at different positions.³⁶ Socrates uses the plural, alloia, indicating that multiple appearances come about and pass away as one moves around the couch, a different appearance for each of the viewer's perspectives (598a9).³⁷ According to the reasoning of the couch illustration, appearance comes about between the couch and the viewer, through their interaction. Appearance is not an intrinsic feature of a material particular, but relational.

How does a relational account of appearance as the focus of imitation fit with Socrates's construal of the item being imitated as a kind? Plato's thinking, I take it, is as follows: a particular couch will not appear as a couch from every angle, or to every viewer. Most clearly, no item will appear as a couch to one who lacks the concept. But even for those who know what a couch is, there are spatial perspectives from which a couch will not be readily recognized as such. In stressing the different angles from which a particular couch can be viewed, Socrates emphasizes that one who intends to imitate the kind Couch must attend carefully to which perspectives of particular couches provoke recognition.³⁸ In fact, the imitator cannot consider this question without also considering to whom this display is to occur. That is, in asking how an item appears as an instance of a kind, the imitator must examine the interaction between selected features and aspects of the model and the epistemic dispositions of the viewer. This point is somewhat obscured by the example of a couch, which will appear as a couch from most angles, and in largely the same circumstances to all. It is brought into greater relief by the shift in Socrates's examples from artifacts to producers (598b8–c1). Which features most readily provoke recognition of a doctor (599c1–3), for example, depend on who is doing the recognizing.³⁹ The same point holds all the more when the kind at issue is ethical. What provokes one to recognize a person as virtuous or happy depends on one's ingrained tendencies for discerning these properties. That an item appears as an item of a specific kind is born out of the interaction between that item and the viewer's perspective, where this includes not just the viewer's spatial vantage point, but a range of epistemic dispositions.⁴⁰

This is why Socrates consistently includes reference to the intended audience in his description of imitative appearances: the painter will "make what seems to be a shoemaker to those who know . . . little about shoemaking [tois mê epaiousin]" (600e7–601a2; cf. 601a6–7, 602b1–3).⁴¹ Indeed, the fact that appearances are relational is central to the power of imitative poetry to corrupt. Imitative poetry seeks to depict fictional characters in an evaluative light, as individuals of specific types, performing actions of ethically significant kinds, and experiencing pleasure or pain, happiness or misery as a result (603c2–7; cf. 600e4–5).⁴² On Plato's view, the characters and events depicted will not appear as intended unless we, or some parts of us, are disposed to apprehend them in the intended light. Imitative poetry recruits our collaboration in the production of its depictions. Accordingly, Socrates describes the work of imitation in copulative terms, saying that imitation consorts with (prosomilei, suggignomenê) a part of us "to produce inferior offspring [genna]" (603b1–5). In his study of the item being imitated, the imitator is considering not merely the features of that item, but the way that those features interact with observers' perspectives to provoke recognition. That interaction, strictly speaking, is the focus of imitation. In productive work, the imitator's goal is to bring the audience and the artwork to interact in much the same way, yielding the same recognition.

These points shed light on the motivations for Plato's account of appearance, and the shortcomings of the feature reading. The ultimate aim of imitation is not to produce an artwork—a material particular produced or altered by the imitator's activity—but instead to produce a specific kind of recognition or awareness in the audience.⁴³ Plato develops a relational account of appearance because the primary explanandum of his theory is active appearing. To satisfy this aim, an adequate account of appearance must identify the element in virtue of which something is actively apprehended as F by some observer. This is why the appearance of F cannot be identified with any feature of an item that remains present in a material particular while it fails to be apprehended as F. If we identify appearance in this way, there will be an explanatory gap between the presence of (what is taken to be) appearance and the recognition that we seek to explain. If appearance is to explain active appearing, it must be construed as a relation that holds between object and observer.

To be clear, Plato does not need to reject the claims of the feature account entirely. We do sometimes refer to an intrinsic feature of an item, in virtue of which it regularly provokes recognition of a kind F, as the appearance of F. So, we might say that the shape and coloration of a boulder give it, or constitute, the appearance of an elephant.⁴⁴ While this manner of speaking is acceptable in ordinary parlance, it does not suffice as a complete theoretical account for two reasons. First, as we have seen, the presence of the feature in question is not sufficient on its own to generate the recognition in an observer necessary for active appearing. This requires an audience positioned in a certain way, interacting with the object. This leads to the second point, which is that the features in question are designated as the appearance of F only derivatively, precisely because they ground a capacity to generate appearings of this kind for certain audiences. So, to call these features appearances in themselves is at best a shorthand for a fuller explanation, the primary element of which is the relation between object and observer.

Proponents of the feature reading are constrained, it seems, by the assumption that our description of the product of imitation must center on the artwork. So constrained, they conclude that appearance in the strict sense is an intrinsic feature in the artwork. But Plato can acknowledge the importance of such features in generating appearances without identifying them as appearances. Near the end of the Sophist, the Eleatic Stranger divides the genus of appearance-making into two species, one that brings about appearances "through the use of tools," and another in which "the very maker of the apparition provides himself as the tool [organon tou poiountos to phantasma]" (Sophist 267a3–4).⁴⁵ The Stranger's formulation is potentially misleading. His distinction is not between a form of imitation that uses tools and another that does not. The distinction, rather, is between a form of imitation that employs an object other than the imitator as a tool, such as painting, and those in which the imitator is the appearance-producing tool, such as acting. In either case, though, the tool is not the appearance that it serves to produce. Nor are the features conferred on the tool appearances, except perhaps in the derivative sense described above, even if those features make the tool a likeness for the item being imitated.

Socrates's book III remarks on imitative poetry do stress the likeness of the imitator to what he imitates: "to make oneself like someone else in voice or appearance is to imitate the person one makes oneself like, isn't it?" (393c4–5). Here three points are important. First, just prior to this description, Socrates has stated that the goal of the poetic imitator's likeness-making is "to make us think" that he is the imitated character himself (393a8–b2). Even here, the fashioning of a likeness is undertaken for the sake of producing recognition in the audience.⁴⁶ Second, Socrates's emphasis on likeness-making in book III is explained by his narrow concern, in that discussion, with the effects of impersonation on the soul of the performer, especially children preparing to become Guardians.⁴⁷ Socrates emphasizes shared features, then, because impersonation does involve making oneself a likeness of the impersonated person. Third, it is therefore especially noteworthy that in book X, Socrates never calls the product of imitation a copy,⁴⁸ or characterizes the imitator's product as something that resembles the item imitated.⁴⁹ Moreover, despite using painting as a model of imitation, Socrates does not so much as mention a material particular artwork, or the processes by which these items are produced, as he does elsewhere (tauta ha plattousin te kai graphousin [510d]). That is, in giving a general account of imitation (cf. 595c8–9), Plato studiously avoids a construal of imitation as a process of likeness-making, centered on the production of an artwork.⁵⁰ He does so, I contend, because he does not want to mistake appearances in the strict sense with the intrinsic features of material particulars through which appearances are sometimes generated.

So far, I have emphasized that likeness-making intrinsic features of an artwork neither constitute appearances in the strict sense, nor suffice by themselves to bring them about. In fact, such features are not even necessary for bringing about appearances. Seeing this is important not only for understanding Plato's account of imitation, but also for grasping the broader application of his theory of appearance. Plato recognizes in the Sophist that some appearance-making proceeds without making a likeness, as in the case of sculptures whose proportions are distorted to accommodate the perspective of the audience (Sophist 235c9–236b3). Though this consideration is not raised explicitly in the book X characterization of the products of imitation, it is implied in two ways. The first concerns the way that appearances are generated for an ignorant audience. Plato's concern, I take it, is that characters depicted in drama will register as happy, say, on the basis of features or events incompatible with actual happiness. A drama that elicits the audience's celebration of a revenge murder provokes the appearance of success or fulfillment in the completion of a vicious act. But if happiness requires virtue, then such depictions do not offer genuine likenesses of happy people or beneficial events; they merely appear happy or beneficial through interaction with the distorted perspective of the audience.

The same holds for the nonmimetic experiences that Plato classifies as appearings in book X and elsewhere in the Republic. These include misperceptions and illusions of various kinds, including distortions caused by water or distance (602c7–d1); trompe l'oeil effects in painting (602d2–4; cf. 583b5); and a state of calm appearing pleasant when it follows pain (583c10–584a10). In each case, some item comes to appear a certain way not solely in virtue of its own features, but because of its spatial or temporal context and the way that this interacts with the perspective of the apprehending subject. And in each case, there is no feature in the item that appears thus-and-so that is suitably associated with the property that it appears to have, so as to explain the appearing on its own. To be clear, an explanation of the appearing will refer to some intrinsic features of the object. But the features referred to in this way may be incompatible with the properties whose appearance they generate! The stick in the water appears bent because it is in fact straight; the coin viewed at an angle looks oval because it is in fact circular. This point holds more broadly since the spatial perspectives of the couch illustration stand for a much broader range of epistemic dispositions. Accordingly, the instances of illusory appearings based in spatial or temporal perspective illustrate a much broader range of subject-relative experiences. This is most easily seen in the case of pleasure and pain. The illusory pleasure taken in relief from pain exemplifies, for Plato, a wide range of pleasures taken in physical indulgence or honor seeking (586a1–d2). The capacity for such pleasures depends on the character of the agents; these activities are pleasant for certain types of people (Hoi . . . phronêseôs kai aretês apeiroi [586a1], to philokerdes kai to philonikon [586d4–5]). In this way, the model of illusory pleasure taken in the relief of pain grounds an explanation of the phenomenon with which Socrates begins his book IX investigation of pleasure, namely, the fact that appetitive, spirited, and rational people find their own lives to be most pleasant (581c9–582a1).⁵¹

Appearing occurs when a recognition-provoking relation, or appearance, comes about typically between an apprehending subject and some external object. There may be many different ways that such relations come about, allowing for the systematic delineation of different kinds of appearing. It is beyond the scope of this paper to detail how the account of appearance is applied in each of these specific cases. Nevertheless, there is reason to believe that Plato sees appearance at work in a number of common experiences and forms of awareness, including pleasure, affective states, and the level of cognition designated as eikasia in the Divided Line.⁵² For now, it is enough to see that a feature reading, which construes appearance as something that exists in the world, residing stably in the material objects that appear to us, cannot accommodate Plato's reasons for delineating appearances in their own right. Nor can it accommodate the range of appearings that Plato invokes in book X and elsewhere in the Republic. This breadth of application requires metaphysical innovation on Plato's part, to delineate an appearance as something that comes about between the mind and the world.

4. the transparency of appearance (598b8–d7)

It may seem that the relational account cannot accommodate a crucial aspect of appearances, namely, that they bear content. One subject to an appearance thereby comes to be in a state of awareness that consists, one might suppose, in apprehending the appearance. That is, one might think that appearances or images are the direct objects of our awareness insofar as we are appeared to, and it is unclear how a relation can occupy this role. In fact, the two main parts of this characterization come apart. The conception of appearance as a relation is consistent with the idea that appearances have determinate content. But to say that appearances have determinate content does not require that they are the objects of our awareness insofar as we are appeared to. It is important to Plato's view, in fact, that they are transparent—the conduits, not the objects, of our awareness insofar as we are appeared to. Getting clear on how appearances are content bearing will yield an improved account of the sense in which an appearance is insubstantial, smikron ti (598b7).

The coming about of an appearance necessarily results in a determinate awareness of an item of some kind or other. This leaves open whether our awareness, insofar as we are appeared to, is assertoric or not. By 'assertoric,' I mean apprehension of an item or state of affairs as occurring, typically in one's immediate environment.⁵³ Non-assertoric awareness, by contrast, occurs when we merely imagine or envision an item or state of affairs. Assertoric awareness of an F declares, as it were, that an F-thing is present. Non-assertoric awareness makes no such declaration; one merely entertains the idea of such a thing. Now, this way of putting things makes it seem like non-assertoric awareness is basic, while assertoric awareness is achieved by adding something like assertion or assent to one's non-assertoric awareness.⁵⁴ Plato's view, I shall argue, is the reverse. The basic awareness brought about by appearance is assertoric; non-assertoric awareness, when it involves appearance, is achieved by qualifying or contextualizing our reception of appearance with additional awareness of the context in which one is appeared to.

To see this, let us revisit the question of how imitation is deceptive. In a controversial remark just after the introduction of appearance, Socrates states that "a good painter, by painting a carpenter and displaying him at a distance . . . might deceive children and foolish adults into thinking that it truly is a carpenter" (598c1–4). Interpreters are often reluctant to take this remark at face value since audiences of imitation are not regularly fooled into thinking that what is depicted is really there. Consequently, some think the deception at issue concerns only the imitator's knowledge and not the reality of his or her products.⁵⁵ Against this, first, Socrates specifies that the deception in question is uncommon, limited to children and the foolish, and more likely to occur when they view a painting at a distance.⁵⁶ Second, in thinking that only one kind of deception can be at issue, these interpreters introduce a false choice. Socrates may be ultimately concerned with the capacity of imitators to deceive audiences regarding their expertise; he is about to consider Homer's supposed knowledge of crafts depicted in his works, with emphasis on the craft of ruling.⁵⁷ But this is consistent with claiming that imitators sometimes deceive children and the foolish into thinking that their imitations are real. Indeed, the fact that imitators deceive these audiences may be part of the way they deceive others regarding their expertise.⁵⁸

For our purposes, understanding how the first deception occurs is most important. Why are children, the foolish, and those who view paintings at a distance tricked? It is because, in some way or other, they lack understanding of the circumstances in which they are appeared to. Viewing a painting from a distance makes it more likely to escape our notice that what we see is just a painting, as with trompe l'oeil scene painting viewed from the back row of the theater.⁵⁹ Similarly, young children may be less likely to realize that they are looking at a painting. In short, we are not deceived when we are aware of a painting as a painting, and this is more common the older and more sophisticated we are, or the closer we are to the painting itself. But this awareness is a second awareness, additional to the one conferred by the appearance of the carpenter itself. To capture this duality, Socrates later assigns these different forms of awareness to distinct parts of the soul (603a1–7).⁶⁰ The nonrational part of the soul forms beliefs (doxazein [602e8], doxazon [603a]) on the basis of appearances. The rational part, by reliance on calculation, formulates opposing beliefs. Crucially, our reason-based understanding does not eliminate the initial "childlike" apprehension, but only enables us to contextualize it, qualifying the impression, say, that a carpenter is present in our physical environment. It is the nonrational part's persistent receptivity to appearance that causes us to see a stick in the water as bent even after we have concluded through measurement that it is straight. Similarly, this persistent receptivity grounds our emotional responses to poetic imitation, leading us to feel for dramatic characters as if they were real. In short, the presence of reasoned judgments contextualizing what appears does not alter how things strike the nonrational parts of the soul (606a3–b8). The state of awareness one occupies insofar as one is subject to an appearance is the same in every case: one receiving an appearance of F comes to be aware of an F-thing as present.

Accordingly, we must distinguish appearance itself from what appears. When we receive an appearance, the appearance itself is not the object of our awareness. This, I take it, is one of the reasons why Plato uses an instance of ordinary appearing, in the couch illustration, to isolate appearance as an item in its own right. The aim of Socrates's reflection on the couch is to examine how material particular couches appear. The item we come to be aware of in this instance is just the couch, as a couch. Since, according to Socrates's reasoning, the couch is distinct from its appearances—the latter change while the former does not—it follows that we do not thereby come to be aware of the appearances themselves.⁶¹ Rather, in the case of ordinary appearings, we come to be aware of some material particular through the appearance. Becoming aware of appearances as such involves philosophical reasoning of the sort performed by Socrates and Glaucon in the couch illustration; this is why the antidote to imitation consists in knowledge and argument (595b6–7, 608a3–4). Ordinarily, and in prereflective contexts, appearance is transparent—it is the conduit of our awareness, not its object.

To capture this, it is tempting to suggest that Plato characterizes appearance in the Republic as a kind of motion or transmission. In the Theaetetus, Socrates's model for the interaction between the appearing thing and apprehending subject classifies a range of experiences—perceptions, pleasures, pains, desires, and fears—as motions that pass from the object apprehended into the perceiving subject (pheretai gar kai en phora . . . hê kinesis pephuken [Theaetetus 156d1–2]).⁶² This construal enables Socrates to distinguish the motion by which awareness is conferred from the object of awareness (156b7). In addition, it allows Socrates to capture the passivity that we experience as subjects of appearance, that is, the fact that we are struck by what appears to us (156a6–7, 157a4–5). Several details in the Republic suggest that Plato is entertaining this idea. The first is the use of proteinein (to extend) to ask whether the gods send misleading appearances (phantasma proteinôn [382a1–2]). The second is the claim that pleasure is a kind of change (kinêsis tis [583e9–10]) shortly before classifying certain pleasures as phantasmata (584a9). A third is the language of becoming applied to appearances in book X (gennêma [597e3–4], genna [603b5–6]). These hints are not enough, I think, to conclude that Plato construes appearance in the Republic specifically as a kind of motion. Appearance is a relation of some kind that obtains between an observer and an object, the transparent conduit of representational awareness. More work is needed before Plato can make this characterization more precise.

5. three degrees from reality: appearance and what appears

We should not mistake lack of specification on this point for a more general lack of commitment to the metaphysical innovations in this argument.⁶³ These are trumpeted in the repeated characterizations of the imitator's object and product as something insubstantial, removed from reality and truth (597e3–4, 598b5–7, 599a1, 599d1–2, 602c1–2, 603a11; cf. 597a10–11, 600e4–6, 605a9–10, 608a6–8). On the basis of the reading developed so far, we are in a position to explicate these claims in detail and, returning to the imitator's creations, explain what it means to be something that appears but is not in truth. Though the imitator's creations are distinct from appearances as subjects of thought and talk, they depend on appearances for their coming to be: they exist just insofar as they appear. Consequently, they inherit the metaphysical deficiencies of appearances, specifically their mind-dependence and what I will call their ontological thinness.

The deficiency of appearances consists in two related characteristics. The first concerns the conditions in which they come about. As relations, appearances depend on the relata between which they come about. Since one of these is a soul, appearances are mind-dependent. No appearance occurs unless someone is appeared to. Moreover, which appearance comes about—what content is conveyed—is determined by the interaction between the features of some object in the world and the perspective of the subject. Far from a being that exists stably and independently, such as an intrinsic feature of a material particular, an appearance is essentially dependent on the interaction of more substantial entities and is, for that reason, often fleeting, like an electrical spark. This is why, presaging the Theaetetus, Socrates uses the language of becoming to characterize appearances (gennêma [597e3–4], genna [603b5–6]).

The mirror illustration characterizes the imitator's creations in similar ways. In listing what one might create with a mirror, Socrates repeatedly uses the word tachu (quickly) (596d8–9, e1–2), indicating that these items come to be in the instant that they come to appear.⁶⁴ It follows, however, that they cease to exist when they disappear from the mirror. Any reason that we might give to claim that they persist after they disappear would be a reason to claim that they exist before appearing. But if they exist before appearing, then they are not created in the moment that they come to appear. The objects seen in the mirror are made by being made to appear.⁶⁵ The mirror illustration introduces the imitator's creations as mind-dependent items that come about and exist just insofar as they appear to or for someone.

The second aspect of metaphysical deficiency depends on the mind-dependence of appearances and concerns their content. The coming about of the appearance of F coincides necessarily with some subject's awareness of an F thing. But the conditions on a kind's realization in a subject's awareness are less stringent than those on bringing about a mind-independent instance. The appearance of a kind F does not require the appearance, much less the material instantiation, of conditions necessary for a mind-independent instance of the property. Thus, a determinable kind can appear without any of its proper determinates: it may appear that there is a tree in the distance, but remain unclear what species of tree is there.⁶⁶ More importantly for Plato's critique of imitative poetry, there is no requirement that the appearance of F include or supervene on the appearance of properties necessary for actually being F. Thus, characters in a drama may appear happy to some audiences despite failing to display any features necessary for actual human happiness, such as virtue. Indeed, for some audiences the appearance of happiness may depend on the appearance of traits incompatible with happiness. As a conduit of awareness, the content of an appearance is limited strictly to what is conveyed. Thus, the appearance of F may come about separately from the conditions necessary or constitutive of the being of the kind.

This is the meaning of Socrates's remark that imitation can make everything "since it engages with something small of each thing, and this is an image [hoti smikron ti hekastou ephaptetai, kai touto eidolon]" (598b7–8). This remark, perhaps more than any other, has encouraged the view that an appearance or image is an intrinsic part or feature of a material particular.⁶⁷ But there is no mention of a material particular here. Socrates's use of hekaston, following those before it, refers to the kind being imitated. Moreover, there is no part language. Finally, ephaptetai does not denote literal touching, since the subject of the verb is mimetikê, the practice of imitation. The sense in which the appearance of a kind is insubstantial consists in a comparison of its content to the essence of the kind, the Form of F—and the necessary features of any mind-independent F. With respect to any kind, F, the appearance of F itself is ontologically thin, containing virtually nothing of the being or truth of the kind.

The appearing things created by the imitator inherit the ontological thinness of the appearances on which they depend. Just as the appearance of F does not require the co-appearance of other kinds necessary for F, so a fictional item can be made to appear without specification of features necessary in an actual instantiation. It is indeterminate whether a couch in a painting has a wood or metal frame. Glaucon's qualification that the mirror-holder can make all things appear, but not be in truth, points to the different requirements on creating mind-dependent and mind-independent instances of a property. To create an F item in truth is to bring about all the conditions necessary for its mind-independent existence. An imitator, by contrast, need not bring about all of these conditions. The imitator must only bring her intended audience to recognize an item as an item of the desired kind. This is why the painter can "paint us a cobbler, a carpenter, or any other craftsman, even though he knows nothing about these crafts" (598b8–c1).

Should we say that appearing things are reducible to appearances, so as to be eliminated on the best description of what is? I am reluctant to ascribe this view to Plato, in part because he recognizes the possibility of truth-bearing discourse about things that appear. The fact that fictional items are mind-dependent, ontologically thin items does not entail that they are private, such that a different thing appears to each viewer of an imitative work. Plato's critique assumes that a work of imitation can generate the same appearance for a large audience. His account can explain this by reference to regularities in the interaction between an audience's shared perspective and the features of the artwork. These regularities make it possible for an audience to witness the same characters and the same events in an imitative work. To be sure, some aspects of imitative work may be subject to truth-bearing discourse, even as other aspects of these works remain open to creative interpretation. Acknowledging discourse about what appears is all the more important to the extent that the account of appearance is applied beyond mimetic contexts to a broader swath of human experience, as indicated by references to appearance in the Divided Line and the Cave (509d10–510a3, 516a6–b7).⁶⁸ An eliminativist position would have Plato say that in such discourse we are thinking, talking, and having feelings about things that do not exist at all. It seems better to say, instead, that the imitator's fictions are distinct, as objects of thought and talk, from the appearances on which they depend.

In light of its broader significance, the sequence by which Plato develops the account of imitation is particularly noteworthy. In moving from our initial, prereflective reception of imitative fictions to the theoretical account of appearance by which they are explained, Socrates's progress exemplifies an ascent in which countenancing entities of progressively greater truth and being, through philosophical inquiry, illuminates our own prior experience. The result, in this case, is not to show that the objects of our prior experience are unreal, nor to deny their existence entirely, but rather to reveal how far from reality and truth they are, because they come about from our interaction with the world.⁶⁹