6. Imagining Images in Chains

In this chapter, I approach Plato’s scenario of the Cave, starting with some narratological observations and then considering the sequence that moves from the Good through Sun and Divided Line to the Cave. It will take some preparation to reach the Cave itself, but an important part of my argument goes against the typical and long-standing view of the Cave as a sort of fictional narrative. I will suggest that the passage is best considered as a hypothetical scenario, embedded in a sequence of theoretical speculation, and not as an allegory, myth, or literary simile. Insofar as the Cave is a kind of trap for readers, it seduces them into the fixed confines of allegorical interpretation, in which the procedure of making likenesses leaves nowhere else to go. [1]
After situating the Cave scenario in the peculiar narrative discourse of the Republic, I make three main points about it. First, I explain the implications of viewing Socrates’ description of the Cave as a hypothetical scenario. By scenario, I mean a theoretical (and not fictional) sequence of events. While the Cave scenario reflects, and is subordinate to, the passages that precede it—in which Socrates enlists the Sun and the Divided Line to make his notion of the Good more accessible—the scenario itself stands as a chain or stack of hypothetical possibilities and probabilities within the larger sequence, building from one nonasserted premise to the next. It offers a conditional chain of argument that, however outlandish, is rational and literal rather than figurative. My second point is that the Cave is a moment of self-reflection in the Republic. The Cave is itself an image, an example of eikasia, that tends to hold readers captive. Instead of taking us closer to the idea of the Good, the sun-line-Cave sequence leads several steps away from it. The turning or conversion of prisoners in the Cave, culminating in the murder of the philosopher, is yet another in the series of digressions that structure the entire book so that it avoids reaching a linear endpoint. Hence my third and final point: in its self-referential aspect, the Cave scenario is essentially a paragon of self-critical reflection, contributing to the undercurrent of caution and equivocation that runs through the entire Republic, both as written by the author Plato and as dominated by the character Socrates. The Cave contributes to the methodological thread of thought articulated in language that, in surprising ways, gives preference to ongoing movement as opposed to stopping at an endpoint.
Although the Cave itself is a tomblike torture chamber where prisoners like us would murder their enlightened benefactors, it is nevertheless possible to believe that “the Cave is Plato’s most optimistic and beautiful picture of the power of philosophy to free and enlighten.” [2] For it is inspiring to imagine ourselves, drawn toward the Good, ascending on “the rough, steep path that leads out of the Cave, up the Line, and into the light of the Sun.” [3] But this path reverses and backs out of Socrates’ chain of images, in which the Cave is a low point that follows from Socrates’ discussion of the Good, which is the highest ideal in Socrates’ proposed hierarchy of ideals, paradigms, and Forms. The description of the Cave is a theoretical argument, and its conclusion is a bad one: if these events were to occur, the philosopher would be murdered. And the Cave is explicitly presented as a likeness (eikōn) in a context where likenesses and mimesis are subjected to radical criticism. [4]
In explaining these points and anchoring them in the specific language of the text, it is my hope to demonstrate the scenario’s extraordinarily self-­referential quality. As a powerful example of rhetorical recursion in the book—and a major counterweight to the central image of the sun—the Cave contributes to the pervasive methodological countercurrent that I have been describing, countervailing the idealized version of investigation that also runs through the Republic and which is, furthermore, often presumed to be pursued in the service of the Platonic goal of Plato’s persuasive efforts. The narrative of the Republic has the structure of a journey that begins with a digression and ends with an afterlife. And the conversation throughout the book moves from digression to digression, guided by floating and shifting, buoylike goals rather than by fixed beacons reached as firm conclusions. The Cave counterbalances the sun as a central perspective on human seeking, ultimately raising—but not answering—the question: how could human beings possibly come close to articulating their ideals?

The Cave in Socrates’ Fictional Narrative

While the term “narrative” is often used quite loosely, it will be best to reaffirm a definition before turning to the Cave itself. For theorists such as Gérard Genette, a verbal narrative recounts a sequence of events in the voice of a more or less prominent mediator, a narrator who is structurally distinct from the author. Narrative, in this view, exhibits a special temporal structure in which the teller portrays the tale (usually comprising past events) as a connected series of events that have already happened. [5] Less typically, a narrative may report events as currently happening (as when a sports announcer describes a game in progress) or going to happen (as in a prophecy). As concerns fictionality, I follow Dorrit Cohn in maintaining a clear distinction between fictional and historical narrative. [6] Abiding by these categories, I have been calling the whole of Plato’s Republic a fictional narrative. Nonfictional narratives, such as history books, are recounted by their authors. And however strange the Republic may be, it does not raise the kinds of postmodern problems that might in some fundamental way call the adequacy of classical narratology into question. [7] The Republic unambiguously presents Socrates telling a story about the previous day, and the fact that Socrates, who historically would have been dead at the time, serves as narrator of Plato’s text earmarks the text as fictional.
The length of the Republic also makes it unlikely as a verbatim transcript. [8] And the chain of evidence whereby Plato reports what Socrates reported that Socrates and others said on a previous occasion is contrived. Readers are unlikely to construe this narrative as verifiable documentation of the events described. We therefore have good reasons for calling the Republic a fictional work of the imagination, featuring the historical Socrates transformed by the author Plato into a fictional character not unlike Virgil in the Divine Comedy. Calling the Republic a fictional narrative is by no means the same as calling it a novel, but the pretense of fictional narrativity established at the beginning introduces the book as fundamentally mediated and modalized, marked by what narrative theorists call distance. I have already linked this distance with the notion of modalization, by which I mean an equivocal mode of discourse that renders thematic statements as possibilities rather than facts, truths, or certainties. Metaphorically expanding grammatical terminology from the level of the sentence to that of discourse, we may describe the global mood of the Republic as oblique and conditional.
At the same time, one of the main things that makes this dialogue (and others) such a strange kind of narrative, and quite different from a novel, is that the action is dominated precisely by dialogue, the narrative being almost entirely in a mode that Genette identifies as “narrative of words” as opposed to “narrative of events.” [9] Instead of mainly recounting various actions performed by characters, Socrates the narrator in the Republic primarily reports on what characters said. On the one hand, the utterances presented through reported discourse never quite escape the boundaries of Socrates’ narration, retaining their tags of attribution (“he said,” “I said”) throughout and never quite shifting into unmediated, dramatic speech. One may ignore these nagging reminders of mediation, as C. D. C. Reeve does for reasons of “readability and intelligibility” in his translation of 2004, but this is at the cost of significant texture. As long as our goal is to gather what Plato said in his writings, we must stick to what he actually wrote. Particularly because of the Republic’s structure of sequential subordination—whereby Socrates’ fictional narrative of events takes us into a fictional narrative of words—I have argued for observing the work as a globally fictional narrative.
On the other hand, the main events in Plato’s dialogues are acts of speech and argument; and in the case of the Republic such arguably uneventful elements take over after Book 1, with notable exceptions being the embedded narratives concerning Gyges and Er, in which characters do lots of things. [10] Keeping in mind the discussion of dominance in my previous chapter, I would also note that almost all fictional narratives contain plenty of nonnarrative, expository elements—static descriptions, comments, claims, arguments, and the like. [11] But the Republic seems overwhelmed by expository elements, to the point where its identity as a fictional narrative is threatened. Just as the Republic portrays an ideal of ongoing movement (made explicit in Socrates’ reluctance to take a stand or position) that struggles to envisage paradigms of discovery, perfection, and knowledge, so too does the book exhibit a struggle in the presentation of ideas in verbal form. These struggles allow us to see standards (conventionalized ideals of perfection) against which to measure both formal features such as digression and thematic concerns such as justice, ideals, and method.
Many readers assume, and some propose—notably Hans Vaihinger—that all imaginary constructs are fictions. But it is useful and commonsensical to observe the difference between theoretical speculation and fictional art, particularly when dealing with literary questions. In Dorrit Cohn’s words, summarizing the view of Käte Hamburger, “novels present us with a semblance or illusion (Schein) of reality that we don’t take in a conditional sense, but that we accept as a reality so long as we remain absorbed in it.” [12] The Cave scenario, as I will elaborate below, is presented precisely in a conditional sense. Theoretical discourse, which is a brand of expository argumentation, strives for a degree of abstract objectivity that is alien to fictionality and to literature, although it can certainly be found in, and even dominate, works that are otherwise fictional and literary.
Like the Republic itself, the Cave is a rather extreme clash between fictional and theoretical discourse. But here my previous concerns are inverted. Whereas an allegory is typically an extended metaphor taking the form of a fictional narrative, the Cave is none of these things. [13] And while the term “allegory” indicates that the text says something other than what it says, allegorical characters and events are understood to translate into transparently stable equivalences, so that an allegorical text has only one correct reading (per allegorical level). Occupying the ultimate position in a sequence of interdependent segments of theoretical speculation, the Cave invites readers to look for this sort of stable meaning, which might stand as a graspable antidote, a cure for our blindness vis-à-vis the Good. In this sense, the Cave holds the temptation of ultimate knowledge by posing as the retrospective key to this puzzle of the Good. [14]

The Greatest Studies

As mentioned in this book’s preface, it seems to me that the greatest strangeness of the Republic comes mainly from the tension that it builds between posing questions and answering them, by leading readers to seek answers that are not necessarily there. At the same time, the book addresses this tension thematically, as a methodological topic. One of the most striking examples of answers being turned into questions may be seen in the unusual way that the text dwells on the topic of education while continuing to evade—but also to consider and not simply ignore—the human desire for a systematic didactic agenda. That is, even when the thematic focus (the what) of the book rests on statements and proposals about education, the methodical treatment (the how) of the topic restlessly continues to raise questions about it, and so this raising of questions is highly self-reflective. Socrates’ treatment of the Good is the most profound instance of evasion in the Republic. He brings up the topic (505a) and then declines to discuss it directly (506e). As Paul Shorey puts it, “The idea of good is nowhere defined, but its supreme importance and all of its meanings are symbolized in the images of the sun and the cave” (1937:2.xxiii). Shorey’s assurance notwithstanding, I would add that the Good is not simply lacking a clear-cut definition in the Republic. This supremely important topic is expressly evaded, in a rhetorical move that turns all that follows (by which I mean, in particular, the sequence of Sun, Line, and Cave) into a digression from the Good. In this sense, the Good is an unexpressed, nonasserted premise at the head of a sequence, followed by degraded, metaphorical, subordinate, displaced copies of it. Thus the Good is presented in the text as the ultimate goal and answer—and as a conspicuously unanswered question. [15]
Addressing “The Good As the Supreme Object of Knowledge,” R. L. Nettleship sees philosophical education as an important thread of continuity in the Republic, linking Books 2–4 with Books 5–7 and also drawing together Books 2–9 (1901:212–213). Nevertheless, “so great an advance” occurs in Books 5–7 “that it looks as if Plato were beginning all over again, and had forgotten or ignored what seemed in the earlier Books to absorb his whole attention” (212). The great digression (which turns away from the absorbing topic of justice at the start of Book 5 and swerves toward the greatest studies) may thus be understood as a corrective revision in the conversation, self-consciously returning to improve on the earlier discussion. But by coming this far and then going back, the conversation only seems to grow longer. As suggested above, this return to the longer way also means revising, again, the benchmarks that were reached when defining justice in the city-state and in the individual. Meanwhile, even more in this revision than before, new priorities seem to lead backward.
Of course, there is nothing unusually recursive about ideas getting revised during a conversation. But Plato’s verbal articulation of the conversation’s progress forms several interesting perspectives that have little to do with straightforward improvement. For one thing, revising an ideal—an ideal of absolute perfection and completeness—by declaring that it was incomplete after all, leads to the idea of a “better” ideal. [16] Socrates raises this problem explicitly when he reassesses the earlier efforts, in another doubly negative formulation (504c). The problem may be mapped out in the methodological domain of traveling. A foregrounded image of method as a path occurs when Socrates draws closer to the idea of the Good. Reflecting on the method of heuristic conversation pursued thus far, Socrates now wishes to hew strictly to a method that he identifies as a “longer way round [makrotera … periodos]” (μακροτέρα … περίοδος, 504b). The “longer way round” has been mentioned and avoided before, when the group settled for a merely adequate account of justice in the soul (435d). Socrates now considers their previous findings “lacking in precision” (504b). According to Adeimantus, however, what was said seemed at the time “within measure” and sufficient. Socrates then dismisses this idea of sufficiency and takes a hard line:
ἀτελὲς γὰρ οὐδὲν οὐδενὸς μέτρον. δοκεῖ δ’ ἐνίοτέ τισιν ἱκανῶς ἤδη ἔχειν καὶ οὐδὲν δεῖν περαιτέρω ζητεῖν.
“For nothing incomplete [ateles] is the measure of anything. But sometimes it seems [dokei] to some people to be already sufficient [hikanōs ekhein] and to require no further seeking [zētein].”
In the context of discussing the earlier discussion, Socrates’ swipe at those unspecified people who sometimes compromise too soon leaves room for his fellow investigators to save face by revising their past acceptance, while Adeimantus takes the additional step of deploring a weakness that afflicts some unspecified people.
But the lack of specificity in Socrates’ complaint extends beyond the people involved and allows us to glimpse a more general problem. By dwelling on some evocative words here, we may better understand the implications of Socrates’ criticism (which is, obliquely, a criticism of the conversation he has been leading). The first of the two sentences just quoted leaves no room for mere sufficiency: according to Socrates, an approach that is atelēs ‘not (a-) complete (teleios)’ is worthless when it comes to the highest studies. It would leave promising conditions unfulfilled. In this part of the Republic, Socrates is turning his attention to the Good as a telos (‘completion, end, fulfillment’, e.g. 504d, cf. 505d–e), and his revised standard of completeness and perfection casts a dubious light not only on earlier findings (such as definitions) but also on earlier standards (such as paradigms) that were beacons guiding the investigation. Some people, ignorant of the incompleteness of their knowledge and complacent in their opinions, see no need for further seeking. The Greek expression translated as “to be already sufficient,” combined with an idiom that means something like “it seems” or “they think,” conveys two aspects of the problem. As an adverb used with the verb ekhō ‘to have, hold’, hikanōs ‘sufficiently’ describes a state of affairs. Deriving from the verb hikanō ‘to come to (a place)’, hikanōs here designates an approach that has reached an endpoint, a static holding pattern of sufficiency. One might also note the connotation of possession introduced by ekhō ‘to have, hold), which becomes explicit in an upcoming contrast between the everyday possession of goods and the philosophical grasp of the Good (505a–e).
And right before declining to give an account of the Good, Socrates suggests that “men who opine something true without knowledge” are like “blind men traveling [poreuomenōn] the right road [hodon]” (ἢ δοκοῦσί τί σοι τυφλῶν διαφέρειν ὁδὸν ὀρθῶς πορευομένων οἱ ἄνευ νοῦ ἀληθές τι δοξάζοντες; 506c). Glaucon, understanding that Socrates is on the verge of confessing his own blindness, urges Socrates to “go through” (dierkhomai) the Good: “Do not stand away [apostēis] when you are, as it were, at the end [telei]” (ὥσπερ ἐπὶ τέλει ὢν ἀποστῇς, 506d). The length and circuitousness of the way toward the Good, which is a periodos ‘round, around (peri-) + way (hodos)’, emphasizes its forbidding distance, which is also reflected in the verb aphistēmi (apo- ‘away, far from’ + histēmi ‘stand’).

A Series of Explanatory Comparisons

Unwilling to discuss the Good per se, Socrates instead embarks on a series of comparisons. He starts by likening the Good to the sun (506e–509b), but in making this comparison, which is often called a simile, Socrates embeds two other metaphors within it. “I’m willing to tell what appears [phainetai] a child [ekgonos] of the Good and most similar [homoiotatos] to it” (ὃς δὲ ἔκγονός τε τοῦ ἀγαθοῦ φαίνεται καὶ ὁμοιότατος ἐκείνῳ, λέγειν ἐθέλω, 506e), says Socrates. Only a little later will it dawn on Glaucon that Socrates is talking about the sun (508a). In my opinion, simile is a misleadingly literary term for this comparison, given that Socrates is offering a theoretical explanation. Similarly, if a physicist were to tell us that electrons in an atom are like moons orbiting a planet, we would be unlikely to call the comparison a simile.
While delaying the appearance of the sun, however, Socrates teases his listeners with a sort of riddle: the Good is most like something that appears to be a child of the Good. The similarity of Good to sun is thus complicated by a metaphor in which the Good shines like the sun (phainō literally = ‘comes to light’; “often of the rising of heavenly bodies” [LSJ]) but appears like a child. So the child initially stands in for the sun just as the sun will stand in for the Good. The metaphor also explains the similarity as a kind of dependence: the Good is to the sun as a parent is to a child. Glaucon pushes the metaphor further by calling the narrative of the father a financial debt that will have to be paid later, while Socrates punningly replies that the “interest” (tokos, which can also mean ‘child, offspring’) will have to do for now (507a). The financial metaphor expands on the familial one; and with each turn of metaphor the argument entails a further image (an offspring or copy), a step of subordination, degradation, digression, and debt that takes us a greater distance from the original. [17]
For the sake of greater clarity, it is worth reviewing a basic problem of metaphor that figures prominently in Socrates’ elaboration of similes about the Good. Our traditional notion of metaphor becomes inadequate when Socrates ascribes literal existence, a higher Reality, to invisible objects, objects that we today call metaphysical, because they are conceived of as existing beyond the physical realm. In order to explain the idea of the Good, Socrates makes a well-known distinction between two fundamental kinds of objects that can be investigated by the soul: real things, which are “seen [horasthai] but not intellected [noeisthai],” and ideas ‘Forms’, which are intellected but unseen realities (καὶ τὰ μὲν δὴ ὁρασθαί φαμεν, νοεῖσθαι δ’ οὔ, τὰς δ’ αὖ ἰδέας νοεῖσθαι μέν, ὁρᾶσθαι δ’ οὔ, 507b–c). By this point in the Republic, the discussion assumes a familiarity with this distinction, and it is implicit in earlier comments made by Socrates as well. Scholars are understandably apt to say that Socrates must resort to metaphorical or mythical language in order to talk about metaphysical Realities such as the Idea of the Good and the Form of Justice. (Idea and eidos both mean ‘form, kind, appearance’, although Socrates hints that the Good is in some ways beyond being entirely [509b].) [18] Yet the burden of language in the dilemma of describing metaphysical entities is not to replace them with figurative versions but rather to describe a theoretical Reality. It is for this reason that I would prefer not to call the comparison between the Good and the sun a literary simile. Socrates is saying quite literally that, of all the things in the physical world, the Good is most like the Sun. He could hardly say otherwise, given that the notion of seeing saturates the entire epistemological vocabulary of the Greeks, making the sun a profoundly overdetermined image.
In Socrates’ comparison, he stresses that the sun is superior to sight; Glaucon calls the sun the “cause” of sight that is “seen by sight itself” (αἴτιος δ’ ὢν αὐτῆς ὁρᾶται ὑπ’ αὐτῆς ταύτης, 508b). [19] Thus he starts to describe the proportions of a complex hierarchy—in which effects are subordinated to their causes—that will soon inform the Divided Line. Difficulties of interpretation notwithstanding, we may agree that the visible “region” (topos ‘place’) is subordinate to the intelligible “region”; and in the Divided Line, likenesses within each of these two realms will be subordinate to originals. Socrates explains the sun as “the offspring of the Good, which the Good begot in a proportion [analogon] with itself” (τὸν τοῦ ἀγαθοῦ ἔκγονον, ὃν τἀγαθὸν ἐγέννησεν ἀνάλογον ἑαυτῷ, ὅτιπερ αὐτὸ, 508b). The hierarchy is growing increasingly complicated here, with Socrates using the metaphor of the child (which, to repeat, is already, as a stand-in for the sun, embedded in the comparison of the Good to the sun, in which the sun is a stand-in for the Good) in order to nest analogous categories inside one another: the dependence of the visible on the sun is itself subordinate to the dependence of the Forms on the overarching Idea of the Good.
And in the larger sequence of explanations following in the footsteps of the Good, the Divided Line depends on the proportions used to explain the image of the Sun. Having exclaimed that the Good, in Socrates’ account, is a profound “excess,” Glaucon must encourage Socrates to continue “going through [diexiōn] the comparison [homoiotēta] with the Sun” (τὴν περὶ τὸν ἥλιον ὁμοιότητα αὖ διεξιών, 509c). Meanwhile, the theoretical comparison that Socrates is making is coming to be more and more about resemblances, and these resemblances are primarily methodological. Each region (topos ‘place’)—the realm of the visible and the realm of the intelligible—is part of a map that describes the world as it can, in practice and in theory, be known. (In fact, the Good and the sun are now both said “to rule as a king” [509b] over their respective regions, a small touch that emphasizes the vastness of this hierarchy, where even philosopher-rulers would themselves submit to ever higher powers and their states would owe allegiance to a yet higher order.)
Using the Divided Line, Socrates explains the relationship between the two regions in terms of likenesses (509d–510b). As indicated in Socrates’ earlier remarks about the sun, the visible region is, proportionally, a version of the intelligible region. Each region has two parts. The first part of the visible region contains “images” (eikōn) such as shadows and reflections, and each such visible image “is like” (eoika) a physical thing in the other part. Socrates introduces a human soul into his description of the intelligible region:
ἧι τὸ μὲν αὐτοῦ τοῖς τότε μιμηθεῖσιν ὡς εἰκόσιν χρωμένη ψυχὴ ζητεῖν ἀναγκάζεται ἐξ ὑποθέσεων, οὐκ ἐπ’ ἀρχὴν πορευομένη ἀλλ᾽ ἐπὶ τελευτήν.
“In one part of it, a soul, using as images [eikosin] the things that were previously imitated, is compelled to investigate [zētein] on the basis of hypotheses, making its way [poreuomenē] not to a beginning [arkhēn] but to an end [teleutēn].…”
For the purposes of my analysis, three things about this difficult passage deserve mention. The first is the recursive interdependence of the two regions. The same things that were physical objects in the visible region are here merely images. Hence in the larger hierarchy the shadowy and reflected images are like the physical objects; the physical objects are like the intelligible objects; and the intelligible objects are like the Forms, which are intelligible by virtue of the Good—all in a theoretical hierarchy of likeness, which is being described by Socrates in a hierarchy of verbal likenesses.
Second, when a human soul enters the picture, we are now talking about methods of investigation as well as the relationships between objects of knowledge. Which brings us to a third point: what ultimately interests Socrates here is the journey of investigation. [20] In the part of the intelligible region just described, the soul must proceed by an inferior route: in the final part of the diagram, the soul is “making [poioumenē] its methodical journey [methodon]” (τὴν μέθοδον ποιουμένη, 510b) by relying solely on the Forms. Socrates emphasizes that this superior manner of investigation does not reach an end (teleutē) but goes “toward an unhypothesized [anupotheton] beginning [arkhēn]” (ἐπ’ ἀρχὴν ἀνυπόθετον, 510b). This paradoxical phrase articulates an endpoint as a starting point, thereby summing up the Republic’s paradox of Socratic investigation in a single ideal toward which one travels, from secure suppositions, at the same time that one travels from it toward secure conclusions.
Socrates and then Glaucon also associate superior intellection with a special method of “dialectic” (dialegomai ‘to converse, discuss’), a procedure whereby “discourse” (logos) can understand intelligible objects by moving beyond the hypotheses or assumptions that ground opinion (511b–d). The participants in the current conversation do not seem to view their own dialogue as an example of this special dialectic, but their own investigation of the Good is nonetheless a version—an all-too-human reflection, perhaps—of the greatest studies. By investigating the topic of investigating the Good, the conversation recalled by Socrates in the Republic passes its highest point of knowledge just when it approaches a depth of recursion that leads into the Cave. The spatial orientation of Socrates’ sequence of images, going from the light of the sun to the darkness of a Cave where the sun will be lost from view, is confirmed in Socrates’ overview of the faculties in the soul that correspond to the four types of knowledge portrayed in the Divided Line. Intellection (noēsis) is located in the “highest” (anō, superlative) quarter of the line; whereas the faculty called eikasia, which comprehends those images most remote from the Good, is at the bottom of the line (511d–e).

Self-Reflection in the Cave

The Cave, in my analysis, is a paragon of self-reflection and rhetorical distancing. Like the makers of political paradigms (472d–e), the prisoners in the Cave are described by Socrates in a sequence of hypothetical statements. The human figures in the sequence (prisoners, puppeteers, visiting philosophers) are said to do things, and the description becomes an extended scenario. As it was with the regime painters, the building of the scenario is punctuated by interruptions from other speakers, frustrating our desire to see the scenario as a smooth and gapless moving picture.
Socrates introduces the initial image of the Cave after summing up the proportions of the divided line.
μετὰ ταῦτα δή, εἶπον, ἀπείκασον τοιούτῳ πάθει τὴν ἡμετέραν φύσιν παιδείας τε πέρι καὶ ἀπαιδευσίας. ἰδὲ γὰρ ἀνθρώπους οἷον ἐν καταγείῳ οἰκήσει σπηλαιώδει, ἀναπεπταμένην πρὸς τὸ φῶς τὴν εἴσοδον ἐχούσῃ μακρὰν παρὰ πᾶν τὸ σπήλαιον, ἐν ταύτῃ ἐκ παίδων ὄντας ἐν δεσμοῖς καὶ τὰ σκέλη καὶ τοὺς αὐχένας, ὥστε μένειν τε αὐτοὺς εἴς τε τὸ πρόσθεν μόνον ὁρᾶν, κύκλῳ δὲ τὰς κεφαλὰς ὑπὸ τοῦ δεσμοῦ ἀδυνάτους περιάγειν, φῶς δὲ αὐτοῖς πυρὸς ἄνωθεν καὶ πόρρωθεν καόμενον ὄπισθεν αὐτῶν, μεταξὺ δὲ τοῦ πυρὸς καὶ τῶν δεσμωτῶν ἐπάνω ὁδόν, παρ᾽ ἣν ἰδὲ τειχίον παρῳκοδομημένον, ὥσπερ τοῖς θαυματοποιοῖς πρὸ τῶν ἀνθρώπων πρόκειται τὰ παραφράγματα, ὑπὲρ ὧν τὰ θαύματα δεικνύσιν.
“Next then … make a likeness of [apeikason] our nature in its education and want of education, likening it to a condition of the following kind. See [ide] human beings as though they were in an underground cavelike dwelling with its entryway [eisodon], a long [makran] one, open to the light across the whole width of the cave; they are in it from childhood with their legs and necks in bonds so that they are fixed [menein], seeing [horan] only in front of [prosthen] them, unable because of the bond to turn their heads all the way around; their light is from a fire burning above [anōthen] at a distance [porrōthen] behind [opisthen] them; between the fire and the prisoners there is an upward path [epanō hodos], along which see [ide] a wall, built like the partitions puppet handlers set in front of [pro] the human beings and over [huper] which they show the puppets.”
So Socrates’ Divided Line leads into an elaborate and extended “image” or “likeness” (eikōn, 515a, 517a]). Increasing the rhetorical complexity of image-making here, Socrates uses an imperative, asking his audience to make the likeness, as though each viewer should make a version or copy of the image that Socrates is describing. The Greek verb apeikazō ‘to form a likeness, copy, compare with’ (from the same family as eikazō ‘to make a likeness’ and eikōn ‘likeness’) strongly links the conversational procedure of likening to eikasia, the just-mentioned kind of knowledge that occupies the lowest part of the Divided Line.
It is crucial, I believe, that this comparison is to be made between “our” human nature and the condition or experience of the prisoners found in the Cave. Even as we do the likening, we human readers, along with Glaucon and Socrates himself, constitute a term in the comparison. This is stressed when Glaucon says that the image (eikōn), and the prisoners, are strange (atopos ‘not [a-] a place [topos]’), and Socrates replies that the prisoners are like (homoios) us (515a). We are therefore implicated, involved, and reflected in the unfolding events that ensue from the initial image.
Many of the details described in the image evoke concerns that have been developed over the course of the dialogue, and the elaborate orientation of elements in the Cave echoes the disorientation observed in the Republic’s opening scene. Relevant themes implicated in methodological questions include the use of images as likenesses, images as shadows and reflections, the length of the path and the distance from the light, the theme of sight, and the struggle between moving and remaining in place. Examples of recursion are rife: use an image to liken us to prisoners who are captivated by images; see prisoners seeing; investigate the path of investigation.
A single sentence, complex and meandering, sets the scene. The aorist imperative “see” (ide, 514a) is followed by a string of seemingly subordinate participles that at times characterize the inhabitants and at other times the Cave. The description floats in a kind of timeless present, without finite verbs or distinct tense markers. Although translators are inevitably compelled to use numerous finite verb forms in English, in Greek the elements of the image are simply there. How these sights depend upon our seeing is emphasized by the repetition later in the sentence of the aorist imperative (ide, 514b); the result is a strange concurrence of linguistic indirection and imaginary, timeless presence before our eyes. When Socrates pauses, Glaucon replies in the present indicative, “I see” (horō). While the tenses and moods are perfectly fitting, we should note that the image is not (yet) conditional, unreal, or even nonreal; Glaucon is, in theory, seeing the picture constructed by Socrates. That is, although he exists here in a realm of historical fiction, Glaucon is responding as a listener-reader to the nonfictional, ahistorical image. Accordingly, there is no grammatical indication of unreality. Instead, the grammatical form is indirect in both syntax and mood.
The successive segments of the sentence weave back and forth between the people in the Cave and the Cave itself. And the sense of an unstable focus pervades the whole sentence. In the middle of the sentence, we find an explanatory result clause containing two infinitives, “remain” and “see.” More participles follow until the end of the sentence, where a simile between the little wall in the Cave and partitions used by puppeteers contains two finite verbs. Though the cave image is itself introduced as the simile’s qualified metaphorical vehicle, a vague and qualified condition; the puppeteers’ partition is a metaphorical vehicle used to describe the Cave. As such, the puppeteering wall, presumably a reality familiar to the Greeks and therefore described with the gnomic present tense, becomes a secondary or subsimile; an internal wall, fittingly wedged into a convoluted syntactical construct.
Socrates is presumably using similes because a familiar image can help us to grasp an abstraction. And this eikōn ‘likeness’ of the Cave is inviting us to practice eikasia, even though Socrates has recently denigrated the use of images. Thus the methodological rhetoric (and vocabulary) of the Line is embedded in the Cave. But if the puppet partition helps to explain the peculiar cave wall, it puts us at a double remove from the aspects of human nature that this whole complex of imagery is supposed to illuminate. The sentence begins with the narrator’s telling us what was said in the past, and ends with a comparison in the gnomic tense of timeless theoretical truth. This progression hardly provides a definitive model for what Plato is doing in the passage as a whole, but it does exhibit a peculiar imbalance between what is fiction posing as fact (as in “I said this yesterday”) and what is fact presented in a figurative amplification (like puppeteers’ partitions).

Conditions in the Cave

After Socrates has asked his listeners to make a likeness of (apeikazō) the image or likeness (eikōn) of prisoners in a Cave, he explains why he thinks the prisoners are like (homoios) us: “For in the first place, do you suppose [oiei] such men would have seen [heōrakenai] anything of themselves and one another other than the shadows … ?” (τοὺς γὰρ τοιούτους πρῶτον μὲν ἑαυτῶν τε καὶ ἀλλήλων οἴει ἄν τι ἑωρακέναι ἄλλο πλὴν τὰς σκιὰς, 515a). Socrates’ image of prisoners in a cave works something like a mirror insofar as it gives its viewers an opportunity to see an image of themselves; the prisoners in the image are like its viewers (us). Yet what makes the people in the image like us is, strangely enough, their inability to see themselves, which suggests that we are going to have trouble seeing the image as it is. It is shadowy, ominous, and chthonic. Socrates explains his comparison between prisoner-viewers and us-viewers by asking Glaucon to treat the initial image of the Cave as a hypothetical assumption—based on the situation described, further conclusions can be drawn. The image thus acts as an understood conditional clause (which is in fact supplied in Glaucon’s response: “If they should have been compelled [optative mood] to keep their heads motionless,” εἰ ἀκινήτους γε τὰς κεφαλὰς ἔχειν ἠναγκασμένοι, 515a), while the verb oiomai ‘to suppose’ introduces a conclusion. [21] And with this conditional question, Socrates establishes the underlying grammatical and rhetorical structure of the Cave scenario, which may be viewed as a long chain of future-less-vivid (“should–would”) conditions, moving from one supposition to the next. The entire chain of reasoning constitutes a thought experiment. [22] Through conversation, the speakers develop a speculative account of what would happen, in theory, if the imagined state of affairs were to exist. (And strangeness notwithstanding, Glaucon fully entertains the possibility that such a cave full of prisoners could, in theory, exist; if only for the sake of conversation.)
Socrates continues to ask questions and draw conclusions, and in each instance he seeks Glaucon’s confirmation: If the prisoners should be able to converse (dialegomai, 515b), do you not believe they would hold such and such? If the prison should have an echo, do you suppose the prisoners would believe such and such? Then surely they would hold such and such … (515c). This is how Socrates articulates the entire scenario (514a–517a); a fuller summary would be tedious. The prisoners do things and have various things done to them, and they even converse while deep in their Cave, which is embedded deep in the conversation recounted by Socrates in the Platonic dialogue. But unlike the tale of Er, in which voices proliferate and echo forth as reports from beyond the grave, the long scenario is not a recounting of anything. It is an account propelled by calculative thinking. Neither Socrates’ companions nor we readers are invited to believe that these events have happened, are happening, or will happen; instead, the scenario has us imagine and consider what would happen under certain specified conditions. The events described are striking and memorable, but the cognitive process holding them together is one of rational acceptance, subsequent events in the scenario being entirely dependent on conditions that were accepted previously.
All of these suppositions, each one subordinated under the last, recede toward a devastatingly profound conclusion: “And if they [the Cave dwellers] should be somehow able to get their hands on and kill the man who attempts to release and lead upward [anagein], wouldn’t they kill him?” (καὶ τὸν ἐπιχειροῦντα λύειν τε καὶ ἀνάγειν, εἴ πως ἐν ταῖς χερσὶ δύναιντο λαβεῖν καὶ ἀποκτείνειν, ἀποκτεινύναι ἄν; 517a). This question is nothing like the conclusion of a fictional narrative, and this central conclusion’s theoretical power—the likelihood that something so shocking could possibly happen in practice—is anachronistically confirmed by allusion to the historical execution of Socrates. With such an extraliterary allusion, this hypothetical scenario in a theoretical sequence in a fictional text draws attention to itself as authored by Plato, and Plato’s fictional character Socrates draws attention to himself as a version of a historical figure who is here theorizing by means of comparisons. The self-awareness of the Cave scenario closes in on itself, emphasizing that Socrates, unable to reach the Good in conversation, has instead led us to a potentially fatal depth of theoretical likenesses. Having predicted his own murder as a theoretical likelihood, Socrates tells Glaucon to connect “this image [eikona] as a whole with what was said before” (ταύτην τοίνυν, ἦν δ’ ἐγώ, τὴν εἰκόνα, ὦ φίλε Γλαύκων, προσαπτέον ἅπασαν τοῖς ἔμπροσθεν λεγομένοις, 517a–b). In order to move forward, Socrates looks backward.
The Cave scenario radically tests the conceptual boundaries of discourse and conversational practice. Socrates introduced the Sun as a way to get around and past the ineffable and elusive Idea of the Good, but this detour has led downward, away from the Sun and under the ground. Located in a theoretical sequence in a fictional depiction of a heuristic conversation, and itself a conversational detour embedded in the larger, ongoing conversation, the scenario is a sublime mise-en-abîme, where methodical discourse questions itself. [23] The scenario may well serve the ends of the conversation, helping us to imagine a path out of our world of shadows. But it is hardly a pretty picture and it is far from the end of the book.


[ back ] 1. Much of my account of the Cave as a metatextual “trap” and a “mise-en-abîme” is anticipated by Laird 2003, who also notes the “recursiveness” of visual vocabulary in Plato’s rhetorical presentation of the scenario.
[ back ] 2. Annas 1981:253.
[ back ] 3. Reeves 1988:91.
[ back ] 4. See Smith 2007, Laird 2003:16–21; Annas 1981:256–257.
[ back ] 5. Genette 1980:33–34.
[ back ] 6. Cohn 1999.
[ back ] 7. “Classical narratology” is a standard designation for the theories of Genette and his peers.
[ back ] 8. See Ferrari 2010.
[ back ] 9. Genette 1980:169–175, under the heading of “mood” and the subcategory “distance.”
[ back ] 10. On eventfulness as a criterion for narrativity, see Schmid 2003.
[ back ] 11. See, e.g., Cohn 1999:14.
[ back ] 12. Cohn 1999:6, see 1–17; Hamburger 1993:36–37; and Vaihinger 1924.
[ back ] 13. Cf. different views in McCabe 1992; Morgan 2000:202–207; Nightingale 2004:94–98.
[ back ] 14. The bibliography listed in Ferarri 2007:501–503 may serve as a starting point for the literature on Sun, Line, and Cave. Cross and Woozley 1964:196–228 provide a helpful overview of key debates. On the Good, see esp. Ferber 1984.
[ back ] 15. Cf. Derrida on “the ellipsis of the sun” (1982:230–244).
[ back ] 16. Cf. Annas 1981:295 for a comparable observation.
[ back ] 17. Cf. Derrida 1983:81–83.
[ back ] 18. So too, the main Greek verb for knowing, oida ‘to know’, is a perfect-tense form of horaō ‘to see’.
[ back ] 19. Brann observes: “What is characteristically Socratic about the Sun Image (on which the Divided Line is a gloss) is that it is reflexive, an image of imaging. But more than that, in presenting the sun as an image of the principle of wholeness, Socrates shows not only how imaging itself comes about but also how that particular kind of philosophical imaging which makes the Whole appear as embedded within itself is possible—how we can see the Good from within recursively, as a likeness of itself. This aspect of the imagery of the Republic is reflected also in the central visual image of the closing myth, the Myth of Er” (2004:206, my emphasis).
[ back ] 20. Cross and Woozley: “We might say that the Line is a map of the country through which the human mind must travel as it progresses from a low degree of intelligence to the highest, while the allegory of the Cave pictures to us the actual journey through the country mapped out in the Line” (1964:208).
[ back ] 21. The verb form translated “would have seen” is an infinitive (heōrakenai) accompanied by the modal particle an, which indicates the optatival sense of the infinitive.
[ back ] 22. “A ‘thought experiment’ is an attempt to draw instruction from a process of hypothetical reasoning that proceeds by eliciting the consequences of an hypothesis which, for aught that one actually knows to the contrary, may well be false. It consists in reasoning from a supposition that is not accepted as true—perhaps is even known to be false—but is assumed provisionally in the interests of making a point or resolving a conclusion” (Rescher 1991:31).
[ back ] 23. On the Cave as mise-en-abîme, cf. Laird 2003, esp. 27.