"Anyway, my dear, those in the sanctuary of Zeus at Dodona used to say that the earliest oracles came from an oak tree. And for the men of that time, because they lacked the wisdom you, today' youth, have, they were simple-minded enough to be satisfied with listening to an oak tree or a stone, so long as it was speaking the truth. You, on the contrary, need to know who is speaking and where he comes from ; you are not satisfied with examining if things are actually this way or that way." (Phædrus, 275b-c)

If ancient scholars already had doubts about the authorship of some of the works attributed to Plato (there is for instance a very old tradition attributing the Epinomis to Philip of Opus, a disciple of Plato who is also said to have "edited" the Laws left unfinished at Plato's death), they had very little concern for the chronology of their writing, which is of so much concern to modern scholars, and were more concerned with the order of reading than with the order of composition. In other words, most of them, including later heads of the Academy, viewed Plato's corpus as a whole intended to help the student in philosophy progress toward truth and wisdom, with some dialogues easier for beginners, some more fit for advanced students.

So, why is it that the order of composition became so paramount lately ?.. We may trace this change of perspective back to the time, in the XVIIth and XVIIIth century, where learned people started to apply the first tools of modern criticism to all ancient literature. The result, with Plato's dialogues as with so many other texts, was to challenge Plato's authorship of most of the dialogues at some time or other. But, in many cases, the criteria used to do so had more to do with the understanding the author had of platonism as a doctrine, and what he deemed "fit" of Plato's assumed style and ideas than with "objective" features of the dialogues independent of the understanding one may have of them.

In other words, these studies led to stressing the apparent contradictions there might be between dialogues, and, after each author had chosen what he thought was true platonism, to his rejection of what would contradict it as inauthentic.

Then came "Darwinism" ! Not "zoological" Darwinism, but its counterpart in so many other fields of thought, "evolutionism" under one form or another. In our case, it led to the assumption that, in order to explain some of the supposed contradictions between dialogues, one only had to assume that Plato's thought had "evolved" during his lifetime of about eighty years. And here we are ! Now, it was of the utmost importance to know when each dialogue had been written to determine what period of Plato's thought it might belong to ! Unfortunately, this approach was no more reliable that the one it was supposed to supersede, because, due to our lack of evidence on the dates of composition of Plato's dialogues, scholars were in a vicious circle and had to put in their assumptions what they expected to prove at the end, namely what could be earlier, middle and late platonism. Thus, for instance, using Socrates' influence as a criteria (the bigger his supposed influence, the earlier the dialogue) was no better at that, because Socrates didn't write a single line and most of what we know about him comes precisely from Plato's dialogues !... Or else, they had to "psychologize" or play Sherlock Holmes with contemporary literature and evidence, which is no better. The point is, there is not a single quotation, or even explicit mention, of any of the dialogues before those we find in Aristotle's works, that is, after Plato's death. At best, scholars point at some passages in speeches of Isocrates, for instance, that might look like allusions to dialogues, but might as well be explained by oral tradition and discussion that most certainly were taking place in the Athens of the time between some of its most prominent thinkers...

Besides, even if there were such quotations or mentions of dialogues, it would remain to date the works in which they were taking place, which in most cases, is about as hard and as devoid of "objective" basis as dating Plato's dialogues ! In most cases, such scholarly works pile up hypotheses on top of assumptions, using earlier scholars' assumptions and hypotheses as facts, provided they have been quoted in enough other studies.

One ray of light, or what was deemed one, in such a mess came with what is called "stylometry", that is, the technique of "measuring" the stylistic features of writings such as frequency of use of certain words of little semantic meaning (those enclitic particles of which the Greek is so rich, for instance), of rhetorical devices and the like, provided it had more to do with the general way of speaking and writing than with the specific topic of this or that dialogue. By comparing measurements in various dialogues, it was thought, it might be possible to "class" them in some way as closer to or farther away from one another. And using the style of the Laws, generally assumed to be Plato's last work, as a reference, it might be possible to measure the "distance" away from the Laws of each dialogue. Unfortunately, if this method may have some merit to weed away some apocryphal writings when it shows use of words or thoughts of a later century, it is much harder to implement with the writings of a single author.

First, it is based on statistical data, which means that it may not be significant with shorter dialogues. Then, it must assume that each dialogue was written in one shot, during a homogeneous phase of the author's stylistic evolution, and not reworked in later stages (which is why, for instance, it would not work well with Aristotle's works, which are most likely made of strata from different times, stitched together at some point in time, by him or a pupil). And it must rely on supposedly "unconscious" features, and thus doesn't work well if the author is consciously trying to mimic somebody else's style, which Plato masterfully does more than once. And again, eventually, it can only give relative results, not absolute dating, which requires a supplementary step not warranted by the method. Yet, despite all these caveats and more, and encouraged by the help of computers in the counting business, scholars have devised a theory according to which Plato's dialogues fall in three main periods :

• early dialogues, written in his youth, when he was still under Socrates' influence (that's why these dialogues are often referred to as "Socratic" dialogues), starting probably with the Apology written soon after Socrates' death ;
• middle dialogues, including, in an order which is not constant among scholars, such great dialogues as the Republic, the Symposium and the Phædrus, and dating from the peak of Plato's career, while he was in his forties and fifties ;
• late dialogues, which become more metaphysical with Socrates progressively fading away, and where the so-called "theory of forms" gets challenges by Plato himself.

What I want to propose in the subsequent pages is that there may be other ways of looking at the dialogues, other assumptions regarding their purpose and composition, that are at least as credible as this one, and as compatible with the available data, once you forget about the wealth of literature scaffolding it. And I want to propose such a set of hypotheses, which look for strength in the dialogues themselves, not in doubtful works of reconstruction.

Plato and his dialogues : Home - Biography - Works and links to them - History of interpretation - New hypotheses - Map of dialogues : table version or non tabular version. Tools : Index of persons and locations - Detailed and synoptic chronologies - Maps of Ancient Greek World. Site information : About the author.

First published May 16, 1996 - Last updated November 21, 1998
© 1996 Bernard SUZANNE (click on name to send your comments via e-mail)
Quotations from theses pages are authorized provided they mention the author's name and source of quotation (including date of last update). Copies of these pages must not alter the text and must leave this copyright mention visible in full.