E-mail Archives : Plato's middle dialogues (about stylometry) October 25, 1995

This page is part of the "e-mail archives" section of a site, Plato and his dialogues, dedicated to developing a new interpretation of Plato's dialogues. The "e-mail archives" section includes HTML edited versions of posts that I submitted on various e-mail discussion lists about Plato and ancient philosophy.

This post was part of a long discussion on the dating of Plato's dialogues and stylometry triggered by the following post to the <plato> list from Anthony F. Beavers, Ph.D., Assistant Professor of Philosophy and Religion, The University of Evansville, Evansville, Indiana, <tb2@evansville.edu>, dated Fri, 13 Oct 95 21:25:39 -0500, Subject : plato's middle dialogues

Fellow Platonists:

I am needing to examine the current situation of the debate concerning the dating of Plato's middle dialogues, principally the Symposium, Phaedo, Republic and Phaedrus. I would like to brush up on the latest research respecting the order in which these texts were written and anything on the relations between them. Can someone please advise me on where I might begin my research? Any help will be greatly appreciated.

To: plato <plato@freelance.com>
Date : October 25, 1995, 18:20:55
Subject : re: Plato's middle dialogues (pt. 1)

Seems to me the whole discussion on stylometrics is messy because you are confusing "stylometry" and "stylistics". Let me try and explain what I mean:

- For me, "stylometry", at least the way I understand it as applied to Platonic studies, is a ... let's say technè, which uses the count of supposely "unconscious" features of style (eg.: frequency of enclitic particles, of hiatus, etc.) in various dialogues to try and set a relative chronology of these dialogues (in the case of Plato, taking the Laws, supposedly "known" to be his last work, as a reference or "starting point" of the study). It must be noted that, staying at this raw level, the method relies on several assumption it does not and cannot prove, lest it run into a circle:

1) It assumes the particularities it studies are actually unconscious, and not the result of deliberate stylistic design (in that sense, stylometry is the opposite of stylistics);

2) It must assume each dialogue was written as a whole during a relatively short period of time, and not reworked over time; otherwise, the "unconscious" features it studies might have changed, and the results would be biased. Or it must find a way, other that stylometry, to isolate parts of dialogues supposed to be from different periods, and large enough for statistical data to still be meaningfull. It is no chance that stylometry was "successfull" (by that I don't mean I agree with its results, but that it found wide acceptance) with Plato, and not with Aristotle. But the fact that each dialogue clearly "shows" a certain unity of style doesn't prove it was written at once in its current form; it might as well be the result of recrafting over the years, as some traditions that came down to us have it.

Besides, stylometry proves nothing with regard to absolute chronology, only that such dialogue might be "prior to" or "later than" another, not saying by how many years.

At that point, we may question the "intrinsic" value of this technè: as far as I am concerned, I tend to think that, within the limits stated above, it may have some limited value as a help to determine the order in which the dialogues were written.

But it is a completely different question to decide whether this result has any value in understanding the dialogues. Sure, this technè may give us an order of reading that might be as good as any other to read the dialogues. But is it the right order to understand them, is it the order, assuming there was one, in which Plato intended them to be read, is stylometry the right technique to try and recover this order, that obviously Plato didn't leave us? In order to answer "yes" to these questions, we must make another assumptions which has nothing to do with stylometry, and which is the one that usually critics of the results of stylometry challenge (at least it is my case):

3) Plato wrote his dialogues during most of his life, at times where he was holding different views, and thus the chronology of the dialogues is required to reconstruct Plato's "evolution", itself required to understand his dialogues.

This "evolution" hypothesis has nothing to do with stylometrics, except that it gives it a use (which is different from giving it a value). Obviously, if, as I hold it, Plato wrote all his dialogues as a set, probably over a relatively short period of time late in his life, with a scheme in mind, for them to be read in a certain order, it doesn't matter any more in what order he wrote them, and whether he wrote them in a different order. He may even have written several of them at the same time, moving back and forth from one to another, reworking early pieces as he proceeded, and who knows what, in which case, stylometry, though it might still be "scientifically" "true", would be of no use at all in understanding Plato.

- "Stylistics", on the other hand, is a whole other business: to me, it is based on the idea that form and matter are inseparable (quite a Platonic idea, is in't it), it assumes that such a writer as Plato meant business in writing dialogues, had something to say in doing so, and had a design in mind to get his "message" across (it doesn't matter that the "message" was not an "answer", a dogmatic "theory", but only a set of "hints" along the road to philosophia, quite the contrary!). It purports to study style as one aspect of the "form" that may help the reader understand the purpose of the author. And it is all the more important in a case like Plato's dialogues, precisely because they don't give ready-made answers. In such works, everything may be a "hint" contributing to the end result.

In that perspective, I hold Brandwood's "Word index to Plato" a very usefull tool in studying the dialogues, I would even say an indispensible one if you mean business. The deliberate (not unconscious) use of certain words or their absence in certain dialogues may be quite meaningfull and helpfull in trying to understand them. Sure, there is no more way to ascertain such feature is deliberate rather than unconscious, but it is no reason not to try and search hints in that area. Well! If that too is "stylometry" (which it is not for me), let it be so. But I'd suggest then that, for the clarity of discussion, we distinguish between what I described as "stylometry" in the strict sense (to which I stated my objections) and what I would prefer to call "stylistics", which is a needed tool among many in studying any work of art.

And what greater work of art than Plato's dialogues?!...