(4th tetralogy : The Soul - 2nd dialogue of trilogy)

Foreword: this page is not the translation into English of its counterpart in French. It is an html version of my translation into English of this section of the Republic directly from the Greek found in the pdf file Plato (the philosopher) : User's Guide accompanied by comments mostly taken from this work and a few notes written for this version but it doesn't include a translation of the numerous notes found in the French version. That will come later.

[531c] [...]
Anyway, I do think, said I, that if the path we have followed till the end through them all (the objects of learning) [531d] arrives at the community and kinship [they have] with one another and gathers through reasoning (sullogisthèi) why they all are of the same family, the hard
work on them will somehow bring us where we want and it won’t have been unprofitable labor, but otherwise [it will be] unprofitable.
I too, he said, so presage. But [it is] a huge task you are talking about, Socrates.
[Is it] of the prelude, said I, or of something [else that] you are talking about? Or do we not know that all these are preludes to the song/law (nomos) itself that must be learned? For, I suppose, the experts in these [matters] don’t seem the least to you [531e] to be dialektikoi?
No, by Zeus! he said, except maybe a very small number of those I happened to meet.
But then, I said, will those [who are] unable to give and receive a logos ever know something of what we say must be known?
No again, he said, to that too.
[532a] Thus, I said, Glaucon, is not this now the song/law itself that to dialegesthai executes / fulfills? The one which, despite its being intelligible, the power of sight would mimic, which we described as attempting to look first toward the living [creatures] themselves, then toward the stars themselves and in the very end toward the sun itself. And so, when someone, by means of to dialegesthai, attempts, without all the senses, through logos, to rush toward that itself which each [thing] is and doesn’t give up until [532b] he might have grasped by thought itself that itself which is good, (1) it reaches the limit of the intelligible itself, like the one earlier that of the visible.
Undoubtedly indeed, he said.
What then? Don’t you call this journey “dialektikèn”?
Yes indeed.
But then, said I, the release from chains and the turning around from shadows toward the likenesses (eidola) and the light and the ascent out of the subterranean [place] toward the sun, and there, regarding the living [creatures] and plants and the light of the sun, inability to yet [532c] look [at them], but regarding the reflections (phantasmata) on waters, habituation, (2) and also [regarding] the shadows of the [things] that are, but not the shadows of images (eidolon) cast by a light [which is] another such [image] when judging in comparison with the sun, all this hard work on the arts we have gone through has this power of elevation of the best [part] of the soul toward the contemplation of the best among the [things] which are, as, at the time, [the power of elevation of] the clearest [part] in the body toward the [contemplation] of the brightest [thing] in the corporeal and [532d] visible place.
I at least, he said, accept [that it is] so. And yet, it seems to me that they are very hard to accept [opinions / propositions / statements /…], but again, from another standpoint, hard not to accept. But nevertheless — for [it is] not only at the present time [that] we have to hear these, but we’ll have also to return back [to it] again many times — assuming those [things] to be as is now said, let us move now to the song/law (nomos) itself and go through it as we have gone through the prelude. Tell [us], then, what [is] the manner [of working] of the power of to dialegesthai, and [532e] and also in how many kinds it is divided and, further, what paths [it follows], for, presently, these [paths], it seems, could be those leading to that very [place] where, once reached, it would be sort of a resting place on the path and the end of the journey.
[533a] No, said I, my dear Glaucon, you will no longer be able to follow [me], though on my part at least, there would be no lack of good will, and [it would] no longer [be] an image (eikôn) of what we are talking about that you would see, but the truth itself, that’s at least what it seems to me, but if [that’s] really [the case] or not, [it’s] no longer worth exhausting all our strength about this; but at least that it be seeing something of the sort, it must be strongly asserted. Isn’t that so?
Surely.
Then also that the power of to dialegesthai alone could reveal [this] to [someone] being experienced in what we have gone through, but [that] otherwise, [it is] in no way possible.
This too, he said, is worth being strongly asserted.
[533b] This, at any rate, said I, nobody will argue with us when we say [it]: that, about each [this or that] itself, what each one is, some other approach (methodos) attempts, in a [specific] way (hodôi), to grasp [it] about everything. But on the one hand, all the other arts (technai) either are toward opinions and desires of men, or are turned toward creations and assemblies, or toward tending [things] that grow and also all those that have been assembled; on the other hand, the remaining ones, which we described as grasping something of what is, geometry and those following it, we have seen that they dream [533c] about what is, but that it is impossible for them to see [as if] in a waking state so long as, while making use of foundations (hupothesesi), they let them unmoved, unable to give an account of them. For where a principle that one doesn’t know, an end [result] and the intermediate [steps] coming from what one didn’t know have been twined together, what contrivance could ever transform such an agreement into knowledge?
None, said he.
Then, said I, the dialektikon pathway (methodos) alone proceeds this way, doing away with foundations (hupotheseis), up to the (leading) principle (archèn) itself [533d] in order to secure itself, and the eye of the soul really buried in some sort of barbaric muddy swamp, it gently draws [it] and leads [it] upward, using as coworkers and assistants in the turning around process the arts (technai) we have been reviewing, which we often called “sciences” (epistèmai) out of habit, but would require another name, connoting more clearness than “opinion” (doxa), but more obscurity than “science/knowledge” (epistèmè); “thought” (dianoia), I think, [is how] we defined it earlier, but it is, it seems to me, not [533e] a controversy about the name, on [concepts] as broad as those about which an inquiry is set before us…
Certainly not indeed, he said, but saying with clarity what may only be plain in the soul as a consequence of its habit of mind!  (3)
It is satisfactory then, said I, like before, to call “science” (epistèmè) the first part, the second one “thought” (dianoia), the third [534a] one “belief / faith” (pistis) and “conjecture / imagination” (eikasia) the fourth one, and those two together “opinion” (doxa) and the two other together “intellection” (noèsis); and opinion (doxa) one the one hand [is] about becoming (genesis), intellection (noèsis) on the other hand about ousian; and what ousia [is] with regard to becoming (genesis), intellection (noèsis) [is] with regard to opinion (doxa), and what intellection (noèsis) [is] with regard to opinion (doxa), science (epistèmè) [is] with regard to belief / faith (pistis) and thought (dianoia) with regard to conjecture / imagination (eikasia); but the relation of analogy between what those [are] about, and the division into two parts of each one of these two, opinable (doxaston) and intelligible (noèton), let us drop [that], so that it doesn’t fill us full with discussions many times longer than those having preceded.
[534b] But for sure, for me at least, he said, the rest at least, insofar as I am able to follow, seems good too.
And will you call dialektikon too the one grasping the logon of the ousias of each [being]? And the one not being able [to do this], will you say that, to the extent that he is not able to give a logon to himself and others, to that extent he doesn’t have intelligence (noûs) about this?
How indeed, said he, could I say [he does]?
Thus also about the good, same thing; [the one] who would not be able to distinguish clearly thru logos, by separating [it] from all the other [things], the [534c] idea of the good (hè tou agathou idea) and, like in a battle, going all the way through all refutations, eager to refute them not according to opinion (doxa), but according to ousian, [would not] in all these find its way through with an unfailing logos, you will say of the one behaving this way that he knows neither the good itself (auto to agathon) nor any other good, but that, if somehow he grasps some image (eidôlon) [of it], [it’s] to grasp through opinion (doxa), not knowledge (epistèmè), and, [after] wandering like in a dream and dozing in his present life, arriving in Hades [534d] before waking up here, to fall asleep forever.
Yes, by Zeus, said he, I will indeed say all this most strongly.
But of course, your own children at least, that you rear and educate in the [present] logos, if one day you should rear them in deeds, you would not allow them, I think, if they were irrational (alogoi) as lines (4), by leading the city, to be masters of the greatest [things / affairs /…]?
Well, no indeed, he said.
Then you will make a law for them to receive more than anything else this education by virtue of which they will be able to ask and answer questions most knowledgeably?
[534e] I will make this law, he said, with you indeed.
Then, doesn’t it seem to you, said I, that, like a capstone for studies, dialektikè lies from our standpoint at the top and that no other study could rightly be put higher than it, but that [535a] by now the [matters] of studies have [reached] completion?
I do indeed, he said.

(1) The Greek text which I translate as “that itself which is good” is auto ho estin agathon, which echoes the auto ho estin hekaston of the previous line, which I translate as “that itself which each [thing] is”. What is the reason for breaking the parallelism between the two Greek expressions and not translate the second expression as “that itself which the good is” or an equivalent expression, that is, read both hekaston (“each [thing]”) and agathon (“good”) as subjects of esti (“is”)?
The answer to this question is simple: it is the absence of the article in front of agathon. For agathon to be understood as subject, Plato should have written auto ho esti to agathon. Yet, all the translators I referred to, except possibly Bloom, translate as if it is what Plato had written: Jowett paraphrases as “the absolute good”; Cornford expands as “the very nature of Goodness itself”; Shorey (Loeb) translates “the nature of the good in itself”; Gube and Reeve interpret as“the good itself”; Reeve 2004 translates as “what good itself is.” Bloom translates as “that which is good itself”, leaving in doubt how we should understand these words: they can be understood as equivalent to “that which good itself is” (“good” subject) or as “that which is good in itself” (“good” predicate), so that his wording is more a stylistic device to give the impression that he stays close to the Greek than the mark of a proper understanding of the Greek of Plato. It seems that all these translators, anxious to read what they expect to read rather than what is actually written, either forgot or never read the Hippias Major, whose dramatic motivation is the difference between ti est kalon (“what is beautiful?”) and ti esti to kalon (“what is the beautiful?”) (see Hippias Major, 287d2-e1).
Indeed, in Greek, both the subject and the adjective used as predicative expression are in the nominative but, except in very specific cases, the adjective used as predicative expression, as opposed to subject, must not be preceded by the article. In our expressions, ho is the neuter singular nominative or accusative of the relative hos (“who”) and similarly, ti is the neuter singular nominative or accusative of the interrogative tis (“who?”). In an expression with the verb « to be », which is the case here, it can only be a nominative, but may as well be subject as predicative expression. Thus, the question is: what is the function of agathon (and of hekaston) in the expression from the Republic, of kalon in the expressions from the Hippias Major? The only way of deciding in the case of the adjectives agathon and kalon is, as in English, the presence or absence of the article, which determines whether the word is subject or predicative expression, that is, whether the adjective remains an adjective and is used as a predicative expression, or is used as a substantive with the article and thus becomes the subject. This is what Socrates shows in the Hippias Major, explicitly making a difference between ti esti kalon (“what is good?”, “good” adjective”) and ti esti to kalon (“what is the good?”, “good” substantive with article), a difference which Hippias doesn’t understand, allowing Socrates to stress the fact that, for him (and thus for Plato holding the pen), each one has a different meaning. When we move from the Hippias Major to the Republic, what was in the form of a question in the Hippias Major, introduced by ti, becomes an affirmative clause in a relative proposition in the Republic, introduced by ho, and then, aside from the presence or absence of the article, the order of the words changes in English: ho estin agathon is translated “[grasp] what is good” and ho esti to agathon is translated as “[grasp] what the good is”. This being so, it would be quite surprising on the part of Plato, so careful about what he writes and who, in the Hippias Major, showed us that he was fully aware of the difference between the two expressions, that, in the Republic, he suddenly decided not to abide by the rule requiring the article before the subject but not before the predicative expression, and wrote ho esti agathon rather than ho esti to agathon with the idea in mind that agathon was subject rather than predicative expression. The comparison with the parallel expression of the line before, ho estin hekaston (“that which each [thing] is”), doesn’t challenge this analysis since hekaston (“each [thing]”) is a pronoun and not an adjective, which doesn’t call for an article and can only be subject.
And the presence of auto (« itself ») in the expression (auto ho estin agathon) doesn’t change the analysis. When Plato wants to talk about “the good itself”, he uses the expression auto t(o a)gathon, potentially elided as auto tagathon, as can be seen at 506d8-e1 (auto men ti pot’ esti tagathon, “what on earth the good itself might be”) and 507a3 (ton tokon te kai ekgonon autou tou agathou, “this yield and offspring of the good itself”). And the fact that, a few lines later, he uses the espressions auto kalon and auto agathon without article (507b5) cannot be used as a counterexample to those I just mentioned since the context is quite different and there, Socrates is not really talking about “the beautiful itself” or “the good itself”, but is recalling the genesis of these notions through the move from multiple beautiful [things] (polla kalla, 507b2), for instance a beautiful picture, a beautiful horse, a beautiful summer evening, and multiple good [things] (polla agatha, id.), for instance a good meal, a good movie, a good argument, to the unique idea (507b6) corresponding to each one of these notions by asking ourselves the question : “in all these expressions where we use the words "beautiful" or "good", what is "beautiful" itself (auto kalon), or "good" itself (auto agathon)?” At this point, he is only referring to the words kalon and agathon, not yet to what they (might) refer to. But Plato had no quotation marks at his disposal, as I have here in English. The context is not the same in the expression from the Republic we are focusing on, where we are definitely in the intelligible only from beginning to end, both with auto ho estin hekaston (“that itself which each [thing] is”) and with auto ho estin agathon (“that itself which is good”), as can be seen from the context, which talks about “attempt[ing], without all the senses, through logos, to rush toward that itself which each [thing] is”, “not giv[ing] up until grasp[ing] by thought itself that itself which is good” and “reach[ing] the limit of the intelligible itself”.
This properly understood wording, in a sentence recalling the allegory of the cave, is an implicit confirmation of my remarks on the sight of the sun at the end of the ascent of the freed
prisoner: to contemplate the sun itself, that is, “grasp by thought itself that itself which the good is/what the good itself is” (the decoding of the allegory which would be provided here, had Plato written to agathon), is not “the limit of the intelligible itself” for it is not possible for human beings, logos allowing them only to describe relations, not to reach the “[things] themselves (auta)”; what is possible for them, and recommended, is to “grasp by thought itself that itself which is good”, that is, to look at “human beings (anthrôpoi) themselves and the other things themselves” in the light of the sun, that is, in the “light” of the good (agathon). The "limit of the intelligible itself” is the relation of each “thing” to the good, the understanding of the way each “thing” is good for us, not the understanding of the good itself in the abstract, which would teach us nothing so long as we don’t consider practical situations, in the same way looking at the sun itself would teach us nothing about what it lights.
But it is normal that scholars and translators fascinated by the magnificent image of the freed prisoner at last contemplating the sun after a long and painful ascent from the depths of the cave expect to retrieve here the contemplation of the sun and “read” what they hope to find rather than what Plato actually wrote, which is not open to doubt since textual criticism offers no alternante reading of these words. (<==)

(2) “But then, said I, the release from chains and the turning around from shadows toward the likenesses and the light and the ascent out of the subterranean [place] toward the sun, and there, regarding the living [creatures] and plants and the light of the sun, inability to yet look [at them], but regarding the reflections on waters, habituation, and also [regarding] the shadows of the [things] that are, but not the shadows of images cast by a light [which is] another such [image] when judging in comparison with the sun, all this hard work on the arts we have gone through has this power of elevation of the best [part] of the soul toward the contemplation of the best among the [things] which are, as, at the time, [the power of elevation of] the clearest [part] in the body toward the [contemplation] of the brightest [thing] in the corporeal and visible place” translates the Greek “hè de ge, èn d' egô, lusis te apo tôn desmôn kai metastrophè apo tôn skiôn epi ta eidôla kai to phôs kai ek tou katageiou eis ton hèlion epanodos, kai ekei pros men ta zôia te kai phuta kai to tou hèliou phôs eti adunamia blepein, pros de ta en hudasi phantasmata theia kai skias tôn ontôn, all' ouk eidôlôn skias di' heterou toioutou phôtos hôs pros hèlion krinein aposkiazomenas--pasa hautè hè pragmateia tôn technôn has dièlthomen tautèn echei tèn dunamin kai epanagôgèn tou beltistou en psuchèi pros tèn tou aristou en tois ousi thean, hôsper tote tou saphestatou en sômati pros tèn tou phanotatou en tôi sômatoeidei te kai horatôi topôi”.
All manuscripts (followed by most modern editors: Adam for CUP, Burnet for OCT, Shorey for Loeb, Chambry for Budé, Slings for the new OCT edition of 2003) unanimously give the text ta en hudasi phantasmata theia, where theia is the accusative neuter plural of the adjective theios,“divine”, applied to the phantasmata (“reflections, images”) formed “on waters” (en hudasi). As can easily be seen, the first part of the long sentence in which these words are found is a reminder of the allegory of the cave and the mention of reflections / images on waters refers to 516a7 where Socrates talks about en tois hudasi ta te tôn anthrôpôn kai ta tôn allôn eidôla (“the images on waters of men and the other [things]”) in a sentence reminiscent of his description of the subsegments of the visible in the analogy of the line which immediately precedes the allegory, where he
explains what he means by eikones (“images”) in his description of the first subsegment, saying that he has in mind “first shadows, then reflections in waters…" (prôton men tas skias, epeita ta en tois hudasi phantasmata...) (509e1-510a1), and also of 516b4-5 where he mentions, about the sun, en hudasi... phantasmata autou (“reflections of it on waters”). But the “divine” character of these reflections on waters is not mentioned in the allegory. It is only in his commentary that Socrates uses the word theiôn (“divine”) about the “contemplations" (theôriôn) of the freed prisoner at the end of his ascent, just before returning to the cave (517d4-5), but what he then qualifies as “divine” is the direct contemplation of the sun, the ultimate step in the progress in the intelligible, assuming it to be possible, while here, he is referring to the first step in the investigation of the intelligible, that in which the prisoner is not yet able to look, not even at what is in the heavens, but at the creatures themselves (human beings and the other things) populating the surface of the earth, but only at their shadows and reflections in waters, and besides, the word theios doesn’t apply to “images”, whatever the word used to name them, eikôn, eidolon or phantasma, but to “contemplations” (theôria). To find the word theios associated with a word meaning “image”, several translators refer to Sophist, 266b-d, where the Elean Stranger opposes images (eidôla, phantasmata, skiai) produced by natural phenomena to man made images such as paintings, qualifying at 266c5 the former, those produced by nature, as theias in the same way that of which they are images is, precisely because they are not made by human beings. But the context is not the same in both cases and here, what is at stake is the difference between images and that of which they are images, concerning things which are themselves images of something else since we are in an allegory.
If the word, though ill fitted here, is nonetheless not impossible under the pen of Plato, there remains that the sentence as found in the manuscripts poses a problem of global construction which led some editors (Ast and Apelt, see note ad loc. in Shorey’s edition of the Republic in the Loeb Classical Library, where he argues for the reading of the manuscripts, reproduced below) to suggest replacing theia (“divine”) by thea (“contemplation”, nominative feminine singular), a word which no longer qualifies the phantasmata, but plays a role in the sentence which can be illustrated by the parallel between the two parts of the clause including the words ta en hudasi phantasmata theia organized around men… de… (“on the one hand…, on the other hand…”):

What this arrangment in columns of the first part and of the two versions of the second part shows is that, with the reading of the manuscripts, nothing answers adunamia blepein, so that a translation respecting the text would look like this: “and there, regarding on the one hand the living [creatures] and plants and the light of the sun, inability to yet look [at them], regarding on the other hand the divine reflections on waters and shadows of the beings, but not the shadows of images cast by a light [which is] another such [image] when judging in comparison with the sun…” that is, a phrase which is incomplete on the side of the “on the other hand”, nothing answering the “inability to look” on the side of the “on the one hand”, which the reader is expected to supply from memory of the allegory. This is not absolutely impossible to accept, since the whole reminder of the allegory, occupying the first part of the long sentence we are analyzing till hôs pros hèlion krinein aposkiazomenas (“cast… when judging in comparison with the sun”) is but a sequence of juxtaposed nominals (“the release” (hè lusis), “the turning around” ([hè] metastrophè), “the ascent” ([hè] epanodos), “[the] inability to look” ([hè] adunamia blepein) detailing in advance what is referred to synthetically later as “this power of elevation” (tautèn tèn dunamin kai epanagôgèn...) and pointed at by the demonstrative tautèn (“this”), these nominals being meant to evoke summarily the various steps of the progress of the prisoner with a leading noun in each case taking the place of a verb, rather than to retell in detail each one of these steps, leaving it to our memory to reconstruct the whole image. There remains that the reading of the manuscripts increases the missing parts in the sentence that have to be supplemented by the reader in an unnatural way from a grammatical standpoint: if we may assume that blepein (“look at”) is implied in the second part, this is no longer possible for adunamia since what should be implied is precisely its opposite. In fact, all the translators who adopt the reading theia I had access to except Reeve in his 2004 translation add in their translation words in English which have no counterpart in Greek to answer the clause adunamia blepein (“inability to look”) by something which is no longer an “inability”, as can be seen below (the added words are underscored):
- Jowett: “But the release of the prisoners from chains, and their translation from the shadows to the images and to the light, and the ascent from the underground den to the sun, while in his presence they are vainly trying to look on animals and plants and the light of the sun, but are able to perceive even with their weak eyes the images in the water [which are divine], and are the shadows of true existence (not shadows of images cast by a light of fire, which compared with the sun is only an image)—this power of elevating the highest principle in the soul to the contemplation of that which is best in existence, with which we may compare the raising of that faculty which is the very light of the body to the sight of that which is brightest in the material and visible world—this power is given, as I was saying, by all that study and pursuit of the arts which has been described.” Jowett seems to have doubts on theia, whose translation he puts within brackets, but doesn’t hesitate to supplement much more than is required as the counterpart of adunamia blepein, which he “translates” “they are vainly trying to look on”.
- Shorey (Loeb): “And the release from bonds, I said, and the conversion from the shadows to the images that cast them and to the light and the ascent from the subterranean cavern to the world above, and there the persisting inability to look directly at animals and plants and the light of the sun, but the ability to see the phantasms created by God in water and shadows of objects that are real and not merely, as before, the shadows of images cast through a light which, compared with the sun, is as unreal as they–all this procedure of the arts and sciences that we have described indicates their power to lead the best part of the soul up to the contemplation of what is best among realities, as in our parable the clearest organ in the body was turned to the contemplation of what is brightest in the corporeal and visible region”, with a note on “created by God” reading: “"theia" because produced by God or nature and not by man with a mirror or a paintbrush. See crit. note and Class. Review, iv. p. 480. I quoted Sophist 266 B-D, and Adam with rare candor withdrew his emendation in his Appendix XIII. to this book. Apelt still misunderstands and emends, p.296 and note.” Shorey argues for the reading theia, but is forced to supplement the words “the ability to see”, which don’t exist in the Greek text, as the counterpart of “eti adunamia blepein” which he translates as “the persisting inability to look directly at” (“persisting” rendering eti in an unusual way).
- Cornford (Oxford UP): “There was also that earlier stage when the prisoner, set free from his chains, turned from the shadows to the images which cast them and to the fire-light, and climbed up out of the cavern into the sunshine. When there, he was still unable to lookꞏat the animals and plants and the sunlight; he could only see the shadows of things and their reflections in water, though these, it is true, are works of divine creation and come from real things, not mere shadows of images thrown by the light of the fire, which was itself only an image as compared with the Sun. Now the whole course of study in the arts we have reviewed has the corresponding effect of leading up the noblest faculty of the soul towards the contemplation of the highest of all realities, just as in our allegory the bodily organ which has the clearest perceptions was led up towards the brightest of visible things in the material world.” Cornford doesn’t translate, but rewrites to break up the long phrase written by Plato and replaces the sequence of nominals by separate sentences with verbs. He keeps the theia of the manuscripts and must supply “he could only see” which is not in the Greek to answer “he was still unable to lookꞏat”.
- Bloom (BasicBooks): “Then, I said, the release from the bonds and the turning around from the shadows to the phantoms and the light, the way up from the cave to the sun; and, once there, the persisting inability to look at the animals and the plants and the sun' s light, and looking instead at the divine appearances in water and at shadows of the things that are, rather than as before at shadows of phantoms cast by a light that, when judged in comparison with the sun, also has the quality of a shadow of a phantom-all this activity of the arts, which we went through, has the power to release and leads what is best in the soul up to the contemplation of what is best in the things that are, just as previously what is clearest in the body was led to the contemplation of what is brightest in the region of the bodily and the visible.” Reading theia and “looking instead” added.
- Grube & Reeve (Hackett): “Then the release from bonds and the turning around from shadows to statues and the light of the fire and, then, the way up out of the cave to the sunlight and, there, the continuing inability to look at the animals, the plants, and the light of the sun, but the newly acquired ability to look at divine images in water and shadows of the things that are, rather than, as before, merely at shadows of statues thrown by another source of light that is itself a shadow in relation to the sun-all this business of the crafts we've mentioned has the power to awaken the best part of the soul and lead it upward to the study of the best among the things that are, just as, before, the clearest thing in the body was led to the brightest thing in the bodily and visible realm.” Grube and Reeve not only add “the ability to look” which is not in the Greek, but further add “newly acquired” which is even less in the Greek.
- Reeve (Hackett 2004): “Then the release from bonds and the turning around from shadows to statues and the light; and then the ascent out of the cave to the sun; and there the continuing inability to look directly at the animals, the plants, and the light of the sun, but instead at divine reflections in water and shadows of the things that are, and not, as before, merely at shadows of statues thrown by another source of light that, when judged in relation to the sun, is as shadowy as they-all this practice of the crafts we mentioned has the power to lead the best part of the soul upward until it sees the best among the things that are, just as before the clearest thing in the body was led to the brightest thing in the bodily and visible world.” Here Reeve still keeps the theia, but doesn’t supply a counterpart to “the continuing inability”, leaving it to the reader to understand the mere “but” as implying the opposite of the whole clause “the continuing inability to look”, helped in this more by remembrance of the allegory than by grammar.
From a grammatical standpoint, the leading word of the sequence starting with pros men, is the noun adunamia (“inability”), which continues the series of nouns lusis (“release”), metastrophè (“turning around”), epanodos (“ascent”), and blepein (“look at”), verb at the infinitive, is subordinate to it. So, the reading of the manuscript has two defects: it breaks the series of names corresponding each to a step of the ascent and it requires assuming that what is implied after pros de is the opposite of what is explicit after pros men, which is bad grammar, to say the least.
My take is that the sentence requires a counterpart to adunamia blepein, that this counterpart was there in Plato’s original text and that a copying mistake dating back from a long time ago corrupted the word used by Plato to give birth to the reading theia. Aside from the grammatical reason, another reason pushes me to challenge the reading theia: the first part of this long sentence follows closely the allegory, reusing many terms used there, or, if not the terms themselves, terms of the same family (the noun derived from the verb, for instance) or close in meaning, and, in this perspective, theia would be the only “foreign body” in this summary, and not an insignificant one! For Plato to casually introduce in a mere summary such an attribute for images when, in the Sophist, it is highlighted by the Eklean Stranger, puzzles his interlocutor and leads to an explanation, seems surprising to me on his part, especially in a sentence where attention is distracted by an awkward construction and the fact that what is called for by adunamia blepein never comes.
But the emendation thea (« contemplation »), which indeed provides a counterpart to adunamia blepein, doesn’t seem acceptable to me. For sure, this word may be related to theasaito at 516a9 and to theasasthai at 516b6 in the allegory (conjugated forms of the verb theasthai,“to contemplate”, built on the root thea), but, in the allegory, theasthai is used, not about the first step of the prisoner outside the cave when he can only look at shadows and reflections on waters, but about the last step, the one where he has become capable of “contemplating” celestial bodies, the stars and the moon at night, and eventually the sun itself, which stand in the allegory for abstract “beings” having no “image” inside the cave, that is, in the visible realm, and the most important of them all, the idea of the good, pictured by the sun. But what would be “contemplated” here would be, on the contrary, what represents in the allegory the very first access to intelligible beings, even before the prisoner becomes able to look at “the living [creatures] and plants and the light of the sun (ta zôia te kai phuta kai to tou hèliou phôs)”, that is, their shadows (skias) and reflections (phantasmata). At this stage of the educational process, contemplation is not yet in order, and the prisoner is rather tediously making efforts to try to “see clearly” (kathoran) what is only so far principles of intelligibility of “beings” in becoming (“living [creatures] and plants”). Besides, from a paleographic standpoint, it implies that a copyist would have added a letter, the iota needed to transform thea into theia, with the result that a sentence formerly grammatically correct was now incomprehensible. Such a “mistake” might be possible on the part of a pious but hardly scrupulous monk prone to seeing the hand of God everywhere, but if that were the case, it would date back at most from the Early Middle Ages, which makes it unlikely that it could have propagated to all the manuscripts that have come down to us.
To help us find a more acceptable word than thea as counterpart of adunamia blepein, I suggest to go back to the vocabulary of the allegory, and more specifically to the first word of the sentence this part of the summary refers to, the one describing the stage where the prisoner can only look at shadows and reflections on waters of men and other creatures outside the cave, which is sunètheia (516a5), which I have translated as “habituation”. Sunètheia is the substantive formed with an ending in –ia, frequent in Greek, after the adjective sunèthès, “living together, habituated, accustomed”, itself derived from èthos, “custom, usage”, by addition of the prefix sun- introducing an idea of relations with something else or community between persons, and it is used by Socrates to refer to a process rather than to the result of this process (hence my translation by “habituation” rather than “habit”). As I said already, in the summary of the allegory here made by Socrates, he uses nouns which are for the most part nouns of action referring to the various stages in the allegory where these actions were then described by verbs (let us not forget that the allegory describes a process of education rather than states resulting from this process), lusis (“liberation”) echoing lusin (noun at the accusative singular) at 515c4 and lutheiè (a form of the verb luein, from which lusis is derived) at 515c6, metastrophè (“turning around”) echoing periagein ton auchena (“turn the neck around”) at 515c7, epanodos (“ascent”) echoing anabaseôs (“ascent”) at 515e7, adunamia (“inability”) echoing horan oud' an hèn dunasthai tôn... (“unable to see a single one of the [things]…”) at 516a2, all feminine and sharing the hè at the beginning of the sentence. To continue this list, the word which would perfectly fit to describe the stage being reminded here is simply sunètheia (“habituation”) used in the allegory itself, which is also a feminine noun of action, as are all the other nouns of the list. But it seems that here, Plato’s Socrates wanted to clearly recall the allegory, but without staying prisoner of the very words he had used in it, even at the cost of using rare words, possibly neologisms coined by him, as if to show by way of example that a dialektikon process requires precisely not to be a slave of words, but to be able to see beyond words and images what they point to. Thus, he returns to the word phantasmata to refer to the reflections on waters, used in the analogy of the line, at 510a1 in the explanation of what he means by eikones (“images”) when talking of the first subsegment of the visible, but replaced in the allegory of the cave at 516a7 by the word eidola (plural of eidolon), a word he uses in our sentence, at 532c6, to refer to the sculpted statues whose shadows the chained prisoners see; and one half of the nouns of action in the above list are rare words, possibly even words coined by Plato: metastrophè (“turning around”) occurs only twice in the dialogues, once a few pages earlier, at 525c5, and once here, and these are the only two examples given in the Bailly Greek-French Lexicon, while the LSJ adds a third example, in the sense of “turn of events”, taken from the Septuagint, the translation in Greek of the Holy Bible made by Jews of Alexandria in the third century B.C., that is, about one century after Plato); epanodos (“ascent”) occurs only three times in the dialogues, at Phaedrus, 267d4, a few pages before in the Republic, at 521c1, and here, and the LSJ gives no examples of use of this word earlier than Plato, except possibly one taken from the Hippocratic Corpus (an example is taken from a letter of Euripides, but these letters are probably apocryphal). Of the word epanagôgè (“elevation”), found in the second part of the sentence, which occurs only there in all the dialogues, the LSJ gives only two examples, this one and one in Thucydides (Histories, VII, 34, 6) in the sense of “naval attack”. In this perspective, it might be quite possible that Plato, to echo in this new context the sunètheia mentioned in the allegory without reusing the same word, decided to discard the prefix sun- and merely keep the root ètheia, whose intuitive meaning, almost identical to that of sunètheia, was probably clear to all Greeks of the time, be it a neologism coined by Plato or an ancient word no longer in use having left no traces in the extant Greek literature, since, as I said, the ending -ia is common in Greek and besides, in the present case, they could find help in the derivatives sunètheia, used in the allegory at the point Socrates was echoing there, and aètheia (“lack of habituation, inexperience”), its opposite, used at 518a7 in the commentary of the allegory (a close example of this in English would be someone who, after using the words “benevolence” and “malevolence”, would use the neologism “volence”, which doesn’t exist alone, to refer to a “disposition”, either good or bad).
There might even be a more profound reason, having to do with what the allegory is about, inviting Plato to drop the prefix sun-: this prefix introduces the idea of some activity (in our case, getting used to something) conducted in cooperation between several persons; but, strictly speaking, in the intelligible realm, there is no possibility of cooperation so long as we are there, since such cooperation implies dialogue (in the intelligible realm there are only words and the ideai some of these words point at by means of eidè) and dialogue between persons requires sound, which is available only inside the cave (in the visible/sensible realm)! Indeed, as I mentioned in my commentary of the allegory, the only dialogues mentioned in it, the dialegesthai between the prisoners to name what they see and the short dialogue when the prisoner is freed and forced to turn toward the fire, take place inside the cave. Once outside, the only sun- something which is possible, is the sullogizesthai (516b9) mentioned about the sun as cause of everything at the end of the ascent, and it is a gathering of ideas inside one single “head”, not of people. True, there are other anthrôpoi outside the cave (in fact the same as those inside, only seen from a different standpoint), but they are there only as objects of study and, as such they are capable of producing sounds (phtheggesthai) only when they are inside the cave as bearers. Thus, Plato’s rationale might have been the following: the first time around, in the allegory, he uses the word sunètheia, know to everybody (especially since one of its meanings has sexual connotations), to be sure to be understood, but when summarizing the allegory at the end of an ascent toward hè dialektikè as the highest science to be learned and practiced, assuming that his interlocutors and listeners will remember the allegory, to be more accurate, he drops the prefix sun- to face them with their individual responsibilities: in the realm of the intelligible, you are alone and nobody can do your homework for you, which is one good reason to return to the cave if you need help… But “help” doesn’t mean thinking in your place.
Thus I suggest to read ètheia in place of theia, that is, taking into account the elision of the ending alpha of phantasmata before the initial eta of ètheia, the sequence phantasmat' ètheia in place of phantasmata theia. To explain the corruption of the text, probably ancient, we must remember that, in the time of Plato, words were written in capital letters only, without stresses, breathings and punctuation signs (thus without a mark of elision), and without spaces between words, as a continuous sequence of letters, so that the sequence phantasmat' ètheia looked like this:
ΦANTAΣMATHΘEIA, while the sequence phantasmata theia would have looked like this:
ΦANTAΣMATAΘEIA. The only difference is the replacement of an eta (written in Greek capital letters as our “H”), the initial eta of ètheia, tenth letter from the left in the sequence of Greek letters reproduced above, by an alpha (written in Greek capital letter as our “A”), the ending alpha of phantasmata, which no longer needs to be elided if the ensuing word is supposed to be theia, starting with a consonant, the consonant theta. As we can see, all that was needed for the current reading to appear was that the two vertical lines of the eta (“H”) be somewhat inclined toward one another rather than strictly parallel for a copist having to choose between an unknown word and a deformed letter to opt for an alpha (“A”) with lines slightly disjoined rather than an eta (“H”) with inclined lines.
If we accept this emendation, we have a counterpart of adunamia (“inability”), but not of blepein (“to look at”). But this is not a problem since the complement of ètheia, which precedes it in the sentence, is introduced by the preposition pros (“regarding”) and the sequence of words pros ta en hudasi phantasmat’ ètheia (“regarding the reflections on water, habituation”) is grammatically correct and perfectly understandable as such, without having to add that this“habituation” is an habituation to looking at (blepein) these reflections, which should be quite obvious, especially for those who heard the allegory moments ago and with the reminder in the first part of the opposition (the pros men clause) that what we are talking about “regarding (pros)” what is listed in each clause is blepein (“look at”). And if Plato organized these two clauses the way he did, starting each one by the preposition pros and rejecting what pros was complementing, adunamia blepein (“inability to look at”) in the first case, ètheia (“habituation”) in the second case, at the end of the clause (or of its first part for the second one), it is to invite the listeners and readers to understand pros in a general sense, the same in each case, even though what it complements might suggest a slightly different understanding in each clause: to transpose the problem into English, a more natural way of organizing these clause would be “inability on the one hand to look at the living [creatures] and plants and the light of the sun, habituation on the other hand to the reflections on waters…” and, with this arrangement, the prepositions are not the same (“at” in the first case, “to” in the second). In Greek, pros with accusative (which is the case here) is perfectly correct with blepein (“look [at]), but somehow awkward with ètheia (“habituation”), but in the order chosen by Plato it becomes perfectly understandable, as is the translation I adopted with “regarding” in both cases and keeping in English the order of the Greek. In the end, the balance of the emended phrase looks like this:

The parallelism is strict, each side of the opposition listing two series of “things”, terrestrial (living creatures and plants) and heavenly (sun) “things” on one side, reflections and shadows on the other side, the only difference being that on the side where the prisoner cannot look immediately at what is offered to his sight, all these “things” are listed together before the inability to look at them is mentioned while on the side of what the prisoner gets used to the list is split into two parts by the word ètheia (“habituation”), which allows the word skias (“shadows”) to introduce a long comment without rejecting ètheia too far, a comment which is a sort of return to the cave with the comparison between what the shadows outside are shadows of, namely, “beings”, and what the shadows inside the cave are shadows of, namely, “images” (eidolon, the statues). (<==)

(3) The words following ephè (“he said”) in Glaucon’s line at 533e3 pose a problem. Editors and translators differ on whether they should be attributed to Glaucon or Socrates, and consider them incomprehensible as they stand, whichever variant is retained. Most of them consider them a late interpolation, probably of Stoic origin and some translators don’t even translate them. Let us see, to begin with, what the Greek text and its variants are and how they are translated by various translators before examining the pros and cons of the various options and offering my contribution to this debate.
In order to do this, it is necessary to put these dubious words in their context, starting toward the end of Socrates previous lines, at 533d7.
- The manuscripts (identified by the letters A, D, F and M) give the following text, which I reproduce without punctuation or attribution to either Socrates or Glaucon (we must remember that, in the time of Plato, a written text was a sequence of capital letters without stresses, breathings and punctuation signs and without spaces between words) underscoring the dubious words:
esti d' hôs emoi dokei ou peri onomatos [hè (F)] amphisbètèsis hois [tosoutôn (A F M) / tosouton (D)] peri skepsis [hosôn (A F M) / hoson (D)] hèmin prokeitai ou gar oun ephè [all' ho (A M) / allo(F D)] an monon dèloi pros tèn hexin saphèneiai [legein (F M) / legei (A D)] en psuchèi areskei [oun (F D) / goun (A M)] èn d' egô hôsper...
that is (words differing from one manuscript to the other are in bold):
- manuscript A: esti d' hôs emoi dokei ou peri onomatos amphisbètèsis hois tosoutôn peri skepsis hosôn hèmin prokeitai ou gar oun ephè all' ho an monon dèloi pros tèn hexin saphèneiai legei en psuchèi areskei goun èn d' egô hôsper...
- manuscript D: esti d' hôs emoi dokei ou peri onomatos amphisbètèsis hois tosouton peri skepsis hoson hèmin prokeitai ou gar oun ephè allo an monon dèloi pros tèn hexin saphèneiai legei en psuchèi areskei oun èn d' egô hôsper...
- manuscript F: esti d' hôs emoi dokei ou peri onomatos hè amphisbètèsis hois tosoutôn peri skepsis hosôn hèmin prokeitai ou gar oun ephè allo an monon dèloi pros tèn hexin saphèneiai legein en psuchèi areskei oun èn d' egô hôsper...
- manuscript M: esti d' hôs emoi dokei ou peri onomatos amphisbètèsis hois tosoutôn peri skepsis hosôn hèmin prokeitai ou gar oun ephè all' ho an monon dèloi pros tèn hexin saphèneiai legein en psuchèi areskei goun èn d' egô hôsper...
The variants outside the underscored section, use of the article (hè) in manuscript F before amphisbètèsis absent in the other manuscripts, accusative singular tosouton... hoson in manuscript D when the other manuscripts read tosoutôn... hosôn, genitive plural, and hesitation between oun and goun toward the end don’t change the overall meaning and have no bearing on the problem here considered.
- Adam (CUP) gives the following text (in this quotation as in all the following ones, in Greek or in English translation, I reproduce the punctuation and typographical disposition of the quoted text, which indicate the changes of interlocutor (the first line is always Socrates talking); besides, in the quotations of the Greek text, I use bold characters for words different from those read in one or another of the four manuscripts quoted above, that is, words not found in any one of them):
Esti d', hôs emoi dokei, ou peri onomatos amphisbètèsis, hois tosoutôn peri skepsis hosôn hèmin prokeitai.   Ou gar oun, ephè.   Areskei oun, èn d' egô, hôsper...
Adam drops the dubious words, rejecting in the critical apparatus the reading of manuscript A (all' ho an monon dèloi pros tèn hexin saphèneiai legei en psuchèi), adding “ineptum glossema damnavit Schneider” (Schneider rejected this improper/unsuitable gloss). In the appendix on this section, after reviewing various emendations of his predecessors, he examines the reasons why Schneider considered them an interpolation, giving weight to one based on the fact that Plato never uses the preposition alla to positively add arguments to the statement which preceded after an answer starting with ou gar oun, and he concludes: “The sentence is evidently an attempt to say that we should be content if the words we use express our meaning clearly. In legei (and still more legeis) en psuchè, we may detect an allusion to the Platonic theory of thought as the conversation of the soul and perhaps also to the logos endiathetos of the Stoics. On this account and also because of hexin, I am inclined to attribute the interpolation to some adherent of the Stoic school, of which, in point of style, it is not unworthy.”
- Burnet (OCT) gives the following text:
   Esti d', hôs emoi dokei, ou peri onomatos amphisbètèsis, hois tosoutôn peri skepsis hosôn hèmin prokeitai.
   Ou gar oun, ephè.
   All' ho an monon dèloi pôs tèn hexin saphèneiai legein en psuchèi < arkesei;
   Nai. >
   Arkesei oun, èn d' egô, hôsper...
As can be seen, Burnet attributes the dubiuos words to Socrates, adding a verb at the end which is not in the manuscripts. To do this, he must also add a line attributed to Glaucon for reasons I will explain later (the presence of the words èn d' egô, “said I”, in the next sentence of Socrates). Besides, he modifies the verb at the beginning of this sentence of Socrates, replacing areskei (3rd person singular of present active indicative of the verb areskein (“it pleases, is satisfactory”) by arkesei, 3rd person singular of future active indicative of the verb arkein (“it will suffice”), which he assumes both at the end of the dubious line, attributed by him to Socrates, and at the beginning of his next line, after the added nai (“yes”) of Glaucon (which could explain, in his perspective, the omission of Glaucon’s nai, a copyist having skipped it between two identical words in the text he was copying), and which he finds better adapted to the dubious line as the main verb in a sentence which, as read in the manuscripts, has none. Lastly, he transforms the pros of the manuscripts into a pôs (“in a certain way”). The OCT edition doesn’t include a translation, but the meaning assumed by Burnet is probably something like this: “but it will suffice only to say clearly what the possession / condition / state / habit within the soul might in a certain way disclose”.
- Shorey (Loeb) gives the following text:
   Esti d', hôs emoi dokei, ou peri onomatos amphisbètèsis, hois tosoutôn peri skepsis hosôn hèmin prokeitai. Ou gar oun, ephèˑ [all' ho an monon dèloi pros tèn exô saphèneian, ha legei en psuchèi, arkesei.] Areskei goun, èn d' egô, hôsper...
The point above the line after ephè implies that he attributes the dubious words to Glaucon as part of his answer. He adds a note in the critical apparatus saying: “The text as printed is that of Hermann, brackets by Adam” and doesn’t translate the words between brackets, with a note on the translation saying: “The next sentence is hopelessly corrupt and is often considered an interpolation. The translation omits it.” His translation of the text surrounding them is:
“But I presume we shall not dispute about the name when things of such moment lie before us for consideration." "No, indeed," he said. * * *"Are you satisfied, then," said I…”
- Chambry (Budé) gives the following text:
   Esti d', hôs emoi dokei, ou peri onomatos amphisbètèsis, hois tosoutôn peri skepsis hosôn hèmin prokeitai.
   Ou gar oun, ephè †all' ho an monon dèloi pros tèn hexin saphèneiai legei en psuchèi†.
   Areskei oun, èn d' egô, hôsper...
He writes in the critical apparatus that he considers the section between the two †, which he assigns to Glaucon, as a corrupted interpolation and he gives the following translation in French:
   “Mais ce n'est pas, je pense, le moment de contester sur le nom, quand on a des questions aussi importantes à débattre que celles que nous nous sommes proposées.
   Non, en effet, dit-il [; il nous suffit d'un nom qui fasse voir clairement notre pensée].
   Je suis donc d'avis, repris-je, de faire comme... ”
A note on the text between brackets reads: “J'ai donné du texte mis entre deux croix la traduction que demande le passage et que semble indiquer les mots de cette phrase dont la construction est impossible” (I gave for the text printed between two crosses the translation required by this section and which the words of this sentence, whose construction is impossible, seem to point at.)
- Slings (OCT 2003) modifies Burnet’s text and gives the following text:
   Esti d', hôs emoi dokei, ou peri onomatos amphisbètèsis, hois tosoutôn peri skepsis hosôn hèmin prokeitai.
   Ou gar oun, ephè, all' ho an monon dèloi †pros tèn hexin saphèneiai legei en psuchèi†
   Areskei oun, èn d' egô, hôsper...
He returns to the attribution of these words to Glaucon, but puts part of them between two crosses, and calls it in the critical apparatus locus desperatus.

If we now look at the translations:
- Jowett translates: “But why should we dispute about names when we have realities of such importance to consider?
   Why indeed, he said, when any name will do which expresses the thought of the mind with clearness?
   At any rate, we are satisfied, as before…”
He considers the dubious words part of Glaucon’s line, but his loose translation doesn’t permit to determiner what exact Greek text he adopted. According to Adams in his commentary on these lines, in his edition of the Republic with Campbell, he gives the text of manuscript A, except that they insert ho before legei.
- Cornford translates, ignoring the dubious words with no explanatory note: “But in considering matters of such high importance we shall not quarrel about a name.
   Certainly not.
   We shall be satisfied, then,…”
- Bloom (Basic Books) translates: “But, in my opinion, there is no place for dispute about a name when a consideration is about things so great as those lying before us.
   No, there isn't, he said.
   Then it will be acceptable, I said, just as before...”
Adding in a note on “he said”: “in all but one of the manuscripts there follows a sentence of which there are several versions, none wholly intelligible. Hence I have left it out of the translation. Its point is apparently that if the clarity of the name mirrors the clarity of the soul in the particular faculty, Glaucon will be content.”
- Grube/Reeve (Hackett) translate, he too ignoring the dubious words with no explanatory note: “But I presume that we won't dispute about a name when we have so many more important matters to investigate.
   Of course not.
   It will therefore be enough to call...”
- Reeve (Hackett 2004)326 : “SOCRATES: …But don’t suppose we will dispute about names, with matters as important as those before us to investigate.
GLAUCON: Of course not, just as long as they express the state of clarity the soul possesses.
SOCRATES: It will be satisfactory, then...”
Contrary to what he did in his revision of Grube’s translation, here he keeps the dubious words and attributes them to Glaucon.

To complete this review, let me add a few words on what might have led Adams and others to consider these words as an interpolation of Stoic origin. Diogenes Laertius, in this Lives and Doctrines of Eminent Philosophers, includes at the end of his life of Zeno of Cithium, the founder of the Stoic school, a summary of the doctrines of this school. And the section of this summary dealing with dialectic includes a definition of epistèmè (“science, knowledge”) which reads: “tèn epistèmèn phasin […] hexin en phantasiôn prosdexei ametaptôton hupo logou (“they say epistèmè [to be][…] a state of mind in the acceptance of representations unchangeable under [the effect of] logos)” (DL VII, 47); the exact same definition appears at VII, 165, where it is attributed to Herillus of Carthage. Now, the words of Plato we are examining come immediately after Socrates has challenged the relevance of the word epistèmè to refer to what he previously talked about and uses the word hexis, which is key to the definition of epistèmè by the Stoics, according to Diogenes Laertius. That was probably enough for some scholars to make the connection between both texts and see the Stoics behind a sequence of words they couldn’t make sense of!
Yet, the word hexis is not rare in the dialogues, where it is used 63 times, two of these occurrences being close to our section, one at 509a5, in the parallel between good and sun, and the other at 511d4, in the analogy of the line. This later occurrence is particularly interesting since it appears in a reply of Glaucon reformulating what he has understood of what Socrates just said, at a point of his answer where he explains how he has understood the word… dianoia!... And he explains it as tèn tôn geômetrikôn te kai tèn tôn toioutôn hexin, “the habit of mind of those dealing with geometry and that of those dealing with similar [things]”. Now, it is precisely when Socrates reintroduces the word dianoia to substitute it to epistèmè that the word hexis shows up again. So, before claiming that hexis has here the « technical » meaning it had for the Stoics in the above quoted definition of epistèmè, it might be on order to first wonder if its use here is consistent with the meaning it has elsewhere in Plato’s dialogues, and particularly in the analogy of the line, to which the context of the words here examined refers. This implies that we answer two questions: who, between Socrates and Glaucon, is speaking the dubious words, and which reading should we adopt to arrive at a grammatically acceptable construction giving meaning to them?
The main grammatical problem posed by this member of phrase is the absence of a main verb if we adopt the reading legein of manuscripts F and M, and the difficulty of accepting a main verb in the 3rd person singular whose subject is not obvious, if we adopt the reading legei of A and D. To solve this problem, I suggest to accept the reading legein and to consider that this member of phrase is not a complete sentence by itself, but the continuation of the sentence initiated by Socrates with the words esti d’, and constitutes the second branch of an alternative, introduced by all’ (“but”) (which implies to retain the reading all’ ho of A and M rather than the reading allo of F and D, and thus for the whole group of words, the reading of M, the only manuscript giving both the reading all’ ho and the reading legein) answering the ou (“not”) of ou peri onomatos amphisbètèsis (“not [a disagreement about the name”) in the following structure:

This way of looking at it is, from a grammatical standpoint, consistent with the position of the ou in the first part of the sentence. The normal position of the negative ou(k) is in front of the word or words it concerns, so that ouk esti peri onomatos amphisbètèsis… is not in Greek the same thing as esti ou peri onomatos amphisbètèsis…: in the first case, the sentence is a negative sentence which says what something is not (“it is not a controversy about names”), while in the second case, it is a positive sentence which says what something is, but only after first saying, in a preliminary member of phrase, what it is not, so that, with the second form, an alla (“but”) is expected from the start (“it is, not a controversy about names, but…”). And to make sure that the difference would be clearly perceived, Plato inserted between esti and ou a parenthetical hôs emoi dokei (“as it seems to me”). For indeed, if the construction with the negative ou after esti might be acceptable owing to the great freedom in Greek regarding the order of words in a sentence, the negative version of the sentence, without an alla (“but”) part, would have been esti d' ou, hôs emoi dokei, peri onomatos amphisbètèsis and not the way Plato wrote it, where the negative is too disconnected from esti to be understood as applying to it. Yet, all translators quoted above translate (or “interpret”) as if Plato had written ouk esti d’, because they don’t find the alla called for by Plato’s construction in Socrates’ sentence before Glaucon interrupts him with his ou gar oun (“certainly not indeed”).
So constructed, the complete sentence opposes, from a grammatical standpoint, a negative part formed around a noun, amphisbètèsis (“controversy”) to a positive part formed as an infinitive proposition built around the verb legein, but nothing forbids such a grammatical asymmetry, even if the result may not be of the purest style, especially if, as we will see is the case, the two parts of the alternative are stated by two different persons, one, a bullish youngster, interrupting the other. But with this construction, amphisbètèsis (“controversy”) cannot be the subject of esti, so that we must assume an implied subject for it. But this subject is easy to deduce from the context: Socrates, in the previous sentence, is questioning the appropriateness of the word epistèmè (“knowledge, science”) for the “arts (technai)” they have been reviewing (arithmetic, geometry and so on), saying he would prefer another term, “connoting more clearness than “opinion” (doxa), but more obscurity than “science/knowledge” (epistèmè)” and reminds his interlocutors that earlier (in the analogy of the line) he suggested the use of dianoia. At this point comes the sentence we are analyzing. Its implied subject is obviously “what we are now doing, discussing about choices of words”. It is this type of discussion on the appropriateness of specific words regarding which he wants to say what it is after having said what it is not, namely, nitpicking on words in the manner of Prodicus (peri onomatos amphisbètèsis).
Regarding the second part of the alternative, the positive one, I understand monon (“only”) as relating to en psuchè (“in the soul”), highlighted by its position at the end of the sentence, locative adjunct to the verb an dèloi, subjunctive with an expressing a possibility (“may be plain”): whoever speaks these words talks about something which may be plain, but only in the soul, provided it has acquired the appropriate “state of mind” (pros tèn hèxin, where hèxin may refer to the acquisition of an habit as well as to the acquired habit as a “possession”, the resulting “state”), which may not be the case for all (possibility). And it is what can only be plain in the soul that we must try to express through a legein as “clearly” as possible. The verb dèloun is derived from the root dèlos and the adjective saphèneiai from the root saphès (“clear, plain, manifest, distinct”), whose meaning is very close to that of dèlos (“visible, clear, manifest, plain”), both referring to the idea of clarity, of evidence. Since the Greek uses words derived from two different roots, I did the same in English, translating saphèneia as “clarity” and dèloun as “be plain”.
Eventually, the opposition, whose grammatical asymmetry I mentioned earlier, is between words (onomata) which we might be tempted to confuse with what they simply point at and which would become objects of controversy in themselves, and an activity, the practice of logos (legein), which can only attempt to express as best as possible the result of a hexis which may or may not occur in the soul under the effect of the work of our mind. What is expressed with words is what the mind is able to understand of what it grasps directly ("intelligible") or through the senses (“visible”) through reflection and thought. But this expression cannot get enclosed in words each of which taken individually would describe exactly and in the same way for all what it refers to. It is in the activity of talking, in the dialegesthai understood in its most ordinary meaning, in the confrontation of viewpoints, that what we are trying to understand may become progressively clearer and an hexis, a “possession”, a “habit”, a “state of mind” may develop over time and make us able to understand and communicate more and more clearly. And this is probably the reason why Plato’s Socrates prefers the expression to dialegesthai, that is, a substantivized infinitive pointing at an activity, to the expression hè dialektikè, a substantivized adjective pointing at an intrinsic quality associated with this activity, hard to describe, and which, stopping, freezing this activity in the atemporality of a qualifier that deprives it of dynamics, voids it of all power.
These ideas are quite in line with what might be expected of Plato’s Socrates and their formulation at this point of the discussion has nothing to surprise us. But a problem remains. Indeed, all this would be quite satisfactory if we could admit that Socrates utters both parts of the sentence, only interrupted between its two parts by an exclamative ou gar oun (“certainly not indeed”) of Glaucon unable to wait till the end of the sentence to express his approbation. Unfortunately, the en d' egô (« said I ») which immediately follows areskei oun (“It is satisfactory then”) forces us to see areskei oun as the beginning of a line of Socrates, while the ephè (“he said”) forces us to see ou gar oun as the beginning of a line of Glaucon. So, unless we do what
Burnet did and insert an extra line attributed to Glaucon between psuchèi and areskei, having left no trace extant manuscripts, we must resolve to attribute the dubious words, that is, the second part of the alternative opened by Socrates, to Glaucon. Is this acceptable?
To answer this question, we may first notice that Glaucon’s answers are not limited to short ready-made formulas of assent or dissent (as is the case with Aristotle in the Parmenides), which means that it is not implausible for him to say more than a few words here. Thus, the problem is rather to figure out if it makes sense, dramatically and psychologically, that he complete a sentence initiated by Socrates expressing an idea that is not obvious and even poses problem to most scholars. So, to begin with, we should remember the long speech of Glaucon toward the end of the analogy of the line, at 511c3-d5, where he summarizes what he understood of Socrates explanations, and more specifically (511d2-5), of the end of it, already mentioned above, where he uses the word hexis to explain the word dianoia. In it, he also mentions hè tou dialegesthai epistèmè (“the science of to dialegesthai” 511c5), which makes what is observed (theôroumenon) appear saphesteron (“clearer”) than when using the so-called technai (“arts, crafts, techniques”), using the adjective saphès which is at the root of the word saphèneia used here. But it is precisely this discussion that Socrates is referring to here and the text we are considering serves as an introduction to a summary of it he gives in the sentence starting by areskei oun. So, it becomes easy to imagin a Glaucon, still proud of the way Socrates complimented him at the time (hikanôtata apedexô, “You have followed most sufficiently”, 511d6), seeing him return to the same topic and burning with impatience to show off in front of the audience that he is able to reformulate himself what the “teacher” is about to say to confirm that he has properly understood. And for Plato, from a dramatic, and even pedagogical, standpoint, having the end of the sentence, which constitutes an important contribution to the understanding of the role of language and dialegesthai, uttered by Glaucon rather than Socrates, with no formal approbation on his part other than his lack of commentary or criticism, is a way of highlighting this part of the sentence and of inviting the reader to wonder whether Socrates would have ended it the same way, had Glaucon let him finish it, and also a discreet illustration of the fact that, in a discussion properly conducted, there is not on the one hand a teacher who knows and on the other hand students who listen, even when the age difference is huge between the interlocutors, as is the case here, but people seeking together to better grasp a truth which transcends them and that all participants, even the oldest ones, must accept that truth may even come out of the mouth of babes.
And let us rembember that Plato didn’t have at his disposal ellipsis, or any punctuation signs for that matter, and couldn’t indicate in the written text the changes of speaker (hence the importance in the Republic of the “he said”, “I replied”, and so on, as means of indicating such changes which were not apparent in the typography).
A few words now on the interpolated clause that Socrates adds between the two parts of the alternative, which gives Glaucon time to take a breath before bursting in to finish Socrates’ sentence: hois tosoutôn peri skepsis hosôn hèmin prokeitai (which I have translated: “on [concepts] as broad as those about which an inquiry is set before us”, a translation I will now justify). These words are included in the first part of Socrates’ sentence and their meaning is biased by the way what precedes them is understood: if, as do all the translators quoted above, the sentence is analyzed as a negative sentence (ou negating esti) where Socrates only says what what he is doing regarding the words epistèmè and dianoia is not, the meaning is more or less something like “a controversy over names is not in order when we have such serious matters awaiting us” (Jowett:“But why should we dispute about names when we have realities of such importance to consider?”; Cornford: “But in considering matters of such high importance we shall not quarrel about a name”; Bloom: “But, in my opinion, there is no place for dispute about a name when a consideration is about things so great as those lying before us”; Grube/Reeve: “But I presume that we won't dispute about a name when we have so many more important matters to investigate”; Reeve 2004: “But don’t suppose we will dispute about names, with matters as important as those before us to investigate”). This reading amounts to saying that words don’t matter on subjects so important / great / serious /… (no explicit adjective is used in the Greek) as those Socrates and his companions are talking about! But this is nearly the contrary of what he is doing! Indeed, if we replace this sentence in its broader context, Socrates is commenting on the word epistèmè as opposed to technè, doxa (“opinion”) and dianoia, that he is about to substitute to noèsis in the list of pathèmata he introduced at the end of the analogy of the line which he repeats, with this modification, immediately after the section we are examining. In other words, he is trying to become more specific on the names of various concepts related to the general idea of “knowledge”. He may not be “disputing” about words if that means arguing with others on different names for the same thing, but he is not suggesting that names don’t matter either, since he is about to change names for things he has already talked about earlier. In fact, what he is showing, if not saying loud and clear, is that words do matter and that bad habits in their use (“the arts (technai)… which we often called “sciences” (epistèmai) out of habit”, 533d3-5) may become a hindrance to proper understanding and progress toward knowledge, and that, if we want to properly understand what “knowledge (epistèmè)” is (what the Theaetetus will not be able to do), we had better be careful in our use of this and related words. But for a discussion on the appropriateness of words to be fruitful, we must first be clear in our mind/psuchè on what we are talking about and more specifically on the relations between the concepts we are handling (the arts (technai) improperly called “sciences” (epistèmai) “would require another name, connoting more clearness than “opinion” (doxa), but more obscurity than “science/knowledge” (epistèmè)”, 533d5-6). This is what the second part of the sentence, read as a positive sentence only interrupted and continued by Glaucon, was intent on saying: when criticizing our earlier vocabulary, we are not arguing on names, but… But trying to say clearly what is (perhaps) clear in our mind but will never be perfectly clear when put in words, so as to be capable of better understanding one another. It is in this positive context that the interpolated words must be understood.
The difficulty in understanding them comes from the fact that most of these words are relative pronouns (hois, tosoutôn, hosôn) for which we must guess what they refer to and that some of these pronouns imply quantity (tosoutôn… hosôn) without specifying what they quantify and from which standpoint (what *** would be so what? so great? so numerous? so important? so broad?...) Each translator has to supply words which are not in the Greek text to make this clause understandable (the words in bold characters in what follows): “But why should we dispute about names when we have realities of such importance to consider?” (Jowett); “But in considering matters of such high importance we shall not quarrel about a name” (Cornford); “But, in my opinion, there is no place for dispute about a name when a consideration is about things so great as those lying before us” (Bloom); “But I presume that we won't dispute about a name when we have so many more important matters to investigate” (Grube/Reeve); “But don’t suppose we will dispute about names, with matters as important as those before us to investigate” (Reeve 2004). To properly understand it, we should start the grammatical analysis with what is most obvious and easily understandable, the words skepsis… hèmin prokeitai (“an inquiry… is set before us”) where skepsis (“inquiry”), nominative, is obviously the subject of the verb prokeitai (“is set before”), 3rd person singular indicative present passive. Obvious too is the fact that the preposition peri (“about”) introduces the genitive plural tosoutôn which immediately precedes it, its position after it being usual in Greek, itself calling for the hosôn which complements it (“so numerous / great / important / broad /… as…”) and refers to what is the object of the skepsis (“inquiry”). But if we take these words as object of skepsis (skepsis peri tosoutôn hosôn hèmin prokeitai…, “an inquiry about *** so *** as… is set before us”), even without deciding what the inquiry is about and under which quantitative viewpoint we evaluate it, number, depth, importance, broadness, or otherwise, there is nothing left in the clause to complement hosôn (“as”) which nonetheless requires a second member of comparison. And we don’t know what to do with the only remaining word, the relative hois, dative plural (“to which”), which may either be masculine, thus referring to hemin (“us”), that is, to the interlocutors of the discussion, or neutral, probably referring in this case to the object of the inquiry “set before us”, as do tosoutôn… hosôn… I suggest to understand this clause as if words were in the following English order: peri tosoutôn hosôn hois skepsis prokeitai hèmin (“on [concepts] as broad as those about which an inquiry is set before us”, and to see it as complementing the verb esti, understood positively (not affected by the negative ou), to make explicit what the alternative ou… alla… (“not… but…”) which follows is about. And I interpret the quantitative aspect of tosoutôn… hosôn… as regarding, not the importance of the subject under investigation, but the broadness of meanings of the words used, here epistèmè, dianoia, technè. In other words, the meaning of Socrates’ remark is not, as the quoted translators understand it, that arguing on names would be inappropriate in a discussion on such momentous matters as those they are talking about, when words are all we have at our disposal to try to understand one another and that it is on the most important matters that it would be most detrimental for us to stay in ambiguity as a result of the polysemy of the words being used, but that, on concepts as far-ranging as those under discussion, it would be useless to fight only about words, since no single word will make it possible for us to understand one another, and it is only through speech, through legein, which explains and clarifies them based on a “clearness” which must preexist in the soul, that we have a chance to verify whether we understand one another and thus have solid grounds to make progress toward a better understanding of the matter at hand. Thus, the quantitative import of tosoutôn… hosôn… refers, as I understand it, to the fact that the more abstract and general the concepts under investigation, the greater the danger of misunderstanding: there are no good reasons to fight about the exact meaning of the word “horse” unless we are naturalists looking for a precise delimitation of the animal species “horse”, or a paleontologist trying to determine precisely when in the past this specific species appeared, while it is almost certain that no two people put the same thing (or things) behind the word “science”, or in Greek epistèmè, not to mention words such as “good”, or “beautiful”, or “just”, or “right”, setting aside the fact that these words may assume different meanings in different contexts.
The requirement of speech, of a legein, grounded in a hexis of the soul rather than on a mere agreement on names, even relying on the kind of terse definitions Aristotle accustomed us to, is what explains the so called “aporetic” character of those of Plato’s dialogues called “Socratic”, supposed to fail because they don’t end up on a “definition” of the concept under investigation. The reason is that, for Plato, such “definitions” are deceptive and useless. What helps making progress in the understanding and contributes to “bound (horizein)” a concept, is not a “definition (horismos, from the same root horos, “boundary”, as horizein) as can be found in a dictionary, but rather the long conversation taking place in a “Socratic” dialogue, at the end of which, either a “definition” is no longer needed because we have understood what we were talking about, or, if we have not yet understood after exchanging hundreds or thousands of words, a definition given in a few words would be useless!...
Why then, because we have a hard time understanding a sentence, reject it as a Stoic interpolation, when, properly understood, it merely theorizes what Plato’s Socrates practiced at length in earlier dialogues? The suspicious words include epistèmè and hexis and it happens that the word hexis is used in a definition of epistèmè ascribed to Stoics by Diogenes Laertius. But these words are not trying to give a definition of epistèmè, but to describe an attitude toward words and discussion in common, to bring to light what makes it possible for us to understand one another through words and the priority of the grasp by the mind over the expression through words. In what sense could Plato use the word hexis in this context and how should we understand it? We should first note that hexis is a name of action derived from the verb echein, a verb whose primary meaning is “to hold, possess”, from which comes the meaning “to have”. Thus, the primary meaning of hexis is “possession” and, from there, “state (of body or of mind)”, that is, a set of tendencies, aptitudes or qualities that we “own/possess” as part of our personal identity, and eventually “habit” (as a matter of fact, “habit” comes from the latin habitus, derived from the verb habere,“to have”, which is the latin equivalent of echein, so that habitus is the latin equivalent of hexis). What Plato is trying to make us understand through the words of Glaucon ending a sentence started by Socrates is that steadfastness in speeches, that is, not changing all the time what we say on a given topic, can only result from prior clearness of “ideas” in our mind (noûs, where dianoia takes place), in our psuchè (“soul”, and more specifically in its logikon part, the part endowed with logos), a clearness which can only result from the “state” of our mind resulting from a “habit” in the way of looking at these “ideas”, of thinking and talking about them, to make them “ours” so they become a “possession” (the prime meaning of hexis) of our soul. Isn’t this what Socrates suggests when, in the Meno, at the end of the experiment with the slave, he tells Meno, talking about the boy who just found the answer to the geometry problem posed to him, that the opinions (doxai) which “in him, like a dream, have just been awakened”, “if he were interrogated many times on these same things in many different ways, he would in the end have a knowledge (episthèsetai) of them no less accurate than anybody else”(85c9)? And if we think that Plato’s Socrates expresses here his conviction that “habit” is what may move us from mere opinion (doxa) to “knowledge” (epistèmè), a conviction reformulated in the words we are analyzing, we may wonder whether there words, far from being a Stoic interpolation, might not on the contrary be one of the sources from which the Stoics got their definition of epistèmè!
But when these words are read the way I do, they may also, from another standpoint, reflect a conflict between Plato and Antisthenes and through him, the Stoics, who consider him as one of their forerunners. Indeed, Epictetus reports in his Discourses an opinion of Antisthenes, which he makes his, according to which archè paideuseôs hè tôn onomatôn episkepsis (“the beginning / foundation / principle of education [is] the inquiry about names” (Discourses, I, 17, 12), an opinion corroborated by the fact that the longest work of Antisthenes, according to the list given by Diogenes Laertius, was titled peri paideias è peri onomatôn (“on education or on names”, in five books; DL, VI, 17). And it seems indeed that Antisthenes, one of the followers of Socrates (Plato mentions him at Phaedo, 59b8 as one of those present in Socrates’ jail the day of his death), attached great importance to a clear definition of each sense of each word and to utmost rigor in their use. So, if in this section of the Republic, Plato’s Socrates is trying to suggest, as I think it is the case, that rigor in the use of words reaches its limits when dealing with abstract concepts and that it is more efficient to look for agreement through dialogue and accept a dose of arbitrariness in the choice of words, so long as an agreement is reached on the meaning they are given within the ongoing discussion, rather than assuming that we have made progress simply by defining once and for all the meaning(s) each word must have for all at all times, then, we may read this section as a way for Plato to distance himself from Antisthenes in their understanding of Socrates’ search for “definitions” of moral concepts. This difference is the one there is between, for instance, a definition of andreia (“courage”) in a dictionary or in Aristotle’s works (mesotès esti peri phobous kai tharrè (“it is the mean with regard to fear and boldness”, Nicomachean Ethics, III, 1115a7) and the Laches!... (<==)

(4) In Greek, alogos means “irrational” in the sense of “deprived of reason, unreasonable” for a person or a behavior as well as “irrational” for numers (e.g.: square root of two) and “incommensurable” for lines (e.g.: the side of a square and its diagonal), that is, having a ratio between them which is an irrational number (square root of two for the ratio of the diagonal of a square to its side).(<==)