STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. 490 Book Reviews prove the point. There, after opposing as ivavria the contraries imvakiv avavevav, etc., Plato further illustrates the two opposed categories by imOv- peiv, iO&tiv, f}ov\jto~dai, on the one hand, and afiovXeiv, pi) WeXav, /«/ iiridvuuv, on the other. This, Apelt says, is in itself enough to prove that Plato supposed contrary opposition to include contradictory. I reply: (1) the technical distinction is not in question in this psychological passage and Plato was as free to speak of opposites or incompatibles, in the broad sense, as we are. Plato, as his treatment of Prodicus shows, made distinctions only when he needed them. He does not need to distinguish contrary and contradictory propositions here. Neither contraries nor contradictories can be true "of the same thing at the same time in the same relation," etc. (2) pi) ide\uv is only in grammatical form the contradictory of i$i\av. It means "to be unwilling," and if we take it so the purely technical inaccuracy disappears. Plato probably coined afiovXcw, perhaps to avoid fii) f3ov\eo-0 drew after it fii) lirifivimv. In any case, the opposition of propositions is not involved at all. Plato is merely piling up synonyms to illustrate a real psychological opposition, and the scrutiny that finds in his words a confusion of contrary and contradictory could find it equally in any modern writer. Thus once more is verified the presumption (I have never said that it is a certainty) that the "mistakes" of Plato are usually the misapprehensions of his interpreters. Paul Shoret Spracfdiche Forschungen zur Chronologie der platonischen Dicdoge. Von Hans v. Arnim. Vienna: Holder, 1912. Pp. 235. Professor von Arnim believes that the distrust of Sprachstatistik still expressed in certain quarters is partly justified by the inadequacy of past presentations of the method, including his own. It is not enough to establish stylistic coincidences between a group of dialogues suspected to belong to the same period. We must know negatively that there do not exist other coin- cidences that might associate the group with an altogether different set of dialogues. The practitioner of the method cannot shift this burden upon his opponents. He must himself compare every dialogue with every other dialogue — r) iroXXairXacriov to tpyov. To undertake this for every trait of style would demand more than "die entsagungsvolle Arbeit vieler Menschenleben." Meanwhile, it is possible to select certain specially significant linguistic phenomena, the exhaustive study of which will practically suffice. For this purpose, Professor von Arnim has chosen the forms or formulae of assent in the dialogues. He has made com- plete collections of these and some allied expressions, but confines himself in Book Reviews 491 the present study to the publication of the results yielded by the expressions of practically unqualified positive assent. In his first table, he gives the number of occurrences in each dialogue of every such expression, from the simple vat to iravramun fiiev ovv. — koi toot' o|iov &iurxvp%tvd(u — <£iAe? yoiv ooto) ytyvto-Oai, etc. Each book of the Republic and the Laws is reckoned as a dialogue. The second table exhibits for each dialogue (a) the forms of assent that occur repeatedly both in it and in other dia- logues, with the numbers for each; (6) forms found once in it and repeatedly in other dialogues; (c) forms found once in it and once in other dialogues; (d) its atrai Xtyo/xtva. The third table shows the number of kinds of expressions of assent and the number of occurrences which each dialogue has in common with every other. In the fourth table, which to the uninitiated eye resembles a table of logarithms, these statistics are worked up into "quotients" or coefficients of affinity between each dialogue and every other. The method by which these quotients are calculated takes careful account both of the number of kinds of expressions, the number of occurrences, and other factors. But it is too complicated to be explained here. The provisional result is the establishment of the following order: Ion, Protagoras — Laches, Republic I, Lysis, Charmides, Euthyphron, Euthy- demus, Gorgias, Meno, Hippias II, Cratylus — Symposium, Hippias I, Phaedo (Crito), Republic II-X, Theaetetus, Parmenides, Phaedrus, Sophist, Politicus, Philebus, Laws I-XII. This is in broad agreement with the conclusions of other investigators. Especially notable are only the early date of the Protagoras, the long list of writings that intervene between Republic I and the remainder of the Republic, and the position of the Phaedrus between the Parmenides and the Sophist. Detailed criticism, both of the method and the results, must be post- poned until Professor von Arnim has worked up the rest of his enormous material. I may be permitted, however, to define my position with regard to his provisional conclusions. The position assigned to the Phaedrus is in accord with the argument in my Unity of Plato's Thought that it is a mature work and that its dialectical method is essentially that of the Sophist. If the Protagoras really is Plato's first or second work, it would help to confirm my view of the ' ' unity of Plato's thought' ' by showing that in this earliest writing his art at least is mature. The long interval between Republic I and II would seem to imply that Republic I was originally an independent dialogue. This view is of course possible and has been maintained. But the burden of proof rests on the maintainer. It is an admirable introduction as it stands. Differences in style are perhaps in part due to difference of theme. The forms of assent in a controversial dialogue may be expected to differ from those in a didactic discussion with a consenting respondent. Professor von Arnim does not, I believe, discuss this point. But he has carefully considered the general probability that stylistic differences may be due to the subject- matter, and he believes that he has eliminated that objection in respect of 492 Book Reviews the formulae included in his statistics. I am not quite convinced. The nature of the theme, if we extend the phrase to include dramatic setting, polemic or satiric intentions, and the character of the respondents, affects the style in more subtle ways than Sprachstatistik has yet taken cognizance of. I think, for example, that I could account for the frequency of 8c ye, observed by Ritter in Republic VIII. I have no a priori objection to statistics of style. On the contrary I believe that it is in theory possible to determine by this method the dates of the dialogues. My reserves relate to three points: (1) The further more precise dating of the dialogues will not appreciably affect the interpretation of Plato's thought. (2) Statisticians must pay much closer attention than they have hitherto given to variations due to the nature of the theme. (3) Though courtesy requires us to presume the correctness of statistics, I have never attempted verification without discovering errors, whether sufficient to invalidate the conclusions is of course in each case arguable. Professor von Arnim has repeatedly proved himself a careful worker, and I have no doubt that his present investigation contains as few as could be expected of the inadvertences quas aut incuria fudit aut humana parum cavit natura. Yet as I turn over his tables, my eye now and then falls on a statement that awakens my suspicion. I will give but one example, which I dare say is accidental and not typical. In Table I, under ot/uu, appears the entry: oleo-Oai ye XPV ^ r * ^* io- an d even olov ye