Phaedr. 228 D. Σωκρ. Δείξας γε πρῶτον, ὦ φιλότης, τί ἄρα ἐν τῇ ἀριστερᾷ ἔχεις ὑπὸ τῷ ἱματίῳ. τοπάζω γάρ σε ἔχειν τὸν λόγον αὐτόν. (And so he had a copy of Lysias' speech, which he presently reads.)

230 D. . . σὺ ἐμοὶ λόγους οὕτω προτείνων ἐν βιβλίοις τήν τε ̓Αττικὴν φαίνει περιάξειν ἅπασαν καὶ ὅποι ἂν ἄλλοσε βούλῃ.

273 A. τόν γε Τισίαν αὐτὸν πεπάτηκας ἀκριβῶς. (This same phrase, πεπατηκέναι τινά, to be familiar with an author, occurs in the Birds of Aristophanes (471) οὐδ' Αίσωπον πεπάτηκας. It seems to imply almost necessarily the use of a copy of the author's works. The Birds came out in 415 Β. C. Mr. Paley speaks of this phrase as new in the time of Plato's literary activity.)

276 C. (The passage speaking of pen and ink, already quoted.)

Theaet. 152 A. Σωκ. φησί γάρ που πάντων χρημάτων μέτρον ἄνθρωπον εἶναι. . ἀνέγνωκας γάρ που; Θεαίτ. Ανέγνωκα καὶ

πολλάκις.

.

162 Α. εἰ ἀληθὴς ἡ ἀλήθεια Πρωταγόρου, ἀλλὰ μὴ παίζουσα ἐκ τοῦ ἀδύτου τῆς βίβλου ἐφθέγξατο.

166 C. οὐ μόνον αὐτὸς ὑηνεῖς, ἀλλὰ καὶ τοὺς ἀκούοντας τοῦτο δρᾶν εἰς τὰ συγγράμματά μου ἀναπείθεις.

Soph. 232 D. Ξέν. Τά γε μὴν περὶ πασῶν τε καὶ κατὰ μίαν ἑκάστην τέχνην, ἃ δεῖ πρὸς ἕκαστον αὐτὸν τὸν δημιουργὸν ἀντειπεῖν, δεδημοσιωμένα που καταβέβληται γεγραμμένα τῷ βουλομένῳ μαθεῖν. Θεαίτ. Τὰ Πρωταγόρειά μοι φαίνει περί τε πάλης καὶ τῶν ἄλλων τεχνῶν εἰρηκέναι. Ξένο Καὶ πολλῶν γε, ὦ μακάριε, ἑτέρων.

Polit. 293 Α. τοὺς ἰατροὺς δὲ οὐχ ἥκιστα νενομίκαμεν, ἐάν τε ἑκόντας ἐάν τε ἄκοντας ἡμᾶς ἱῶνται, καὶ ἐὰν κατὰ γράμματα ἢ χωρὶς γραμμάτων, . . πάντως οὐδὲν ἧττον ιατρούς φαμεν κτλ.

.

Parmen. 128 D. διὰ τοιαύτην τὴ φιλονεικίαν ὑπὸ νέου ὄντος ἐμοῦ ἐγράφη, καί τις αὐτὸ ἔκλεψε γραφέν, ὥστε οὐδὲ βουλεύσασθαι ἐξεγένετο, εἶτ ̓ ἐξοιστέον αὐτὸ εἰς τὸ φῶς εἴτε μή.

In these passages we see that books were so common in Plato's time that not to know the contents of a certain one would prove a man deficient in education,-that they were put before schoolboys to learn lessons out of,-that particular ones were read again and again by the same person, that there were books on rhetoric, on the uses of salt, on cookery,

on medicine, on wrestling, and, in a word, on all arts,that once a book was stolen and circulated while the author was still deliberating about publishing it,-that a man overheard another reading from a book and immediately got hold of the book to read it for himself. If now the use of books was so general in all circles of life in Plato's time, the first thirty or forty years after 400 B. C., and if, as we have previously seen, mention of reading and writing, of tablets, papyrus, and parchments goes back to about 450 B. C., and the mention of books and of book-writers (copyists) and book-selling comes along between 420 and 405 B. C., can it be supposed that so quick-witted a people as the Athenians, so interested especially in every stimulus to mental activity, failed to see the capabilities of this contrivance and to make use of it in that earlier period?

I may be permitted in conclusion briefly to restate the evidence as to that earlier period. We have in Pindar before 450 B. C. a metaphor drawn from the arts of writing and reading. We have in Aeschylos, before 460 B. C., repeatedly the metaphor from writing, and once a mention of tablets and of papyrus. We have in Herodotos, before 425 B. C., frequent reference to writing on papyros, and once a recognition of that as the usual material for writing, occasionally supplemented. by parchment. We have abundant fragments of Hekataios (540-480 B. C.) and other early historians, in a style of composition that forbids the idea of oral transmission. We have from the comic poets Kratinos (before 420 B. C.), Eupolis (before 412 B. C.), and Plato (probably before 405 B. C.), fragments containing mention of book-writing, paper, and book-selling. We have from Aristophanes (in plays down to 405 B. C.) reference to books as used by authors and readers, and consulted by his own audience. We have in Thukydides (probably before 405 B. C.) reference to the works of his predecessors implying knowledge of their contents on his part, and a suggestion that other historical inquirers would consult his own work as he had theirs. Finally we have in Xenophon (in reference to a time before 400 B. C.) mention of books as read among a company of friends, as bought by a collector of

a library, and as exported to the shores of the Euxine sea. Now in view of this evidence, recognizing the fragmentary character of the remains we have of the literature of the fifth century before Christ, are we not justified in holding that the use of writing on papyrus for the purpose of preserving and multiplying copies of works of literature began as early as the middle of that century and rapidly grew to be a familiar matter of common life before its end?

It will be observed that I have confined myself to the production of the evidence attainable on my subject with only the necessary explanation of it. My purpose has been simply to bring together all the passages which I could find containing real evidence, in the hope that the collection, not elsewhere made so far as I know, might be of service to any one wishing to ascertain the facts.

IV.—The Declension of the Definite Article in the Cypriote Inscriptions.*

BY ISAAC H. HALL,

PHILADELPHIA, PA.

The riddle of the Cypriote Inscriptions, up to this time, cannot be said to have been completely solved. Not to mention particular knots that occur in sundry inscriptions whose purport is well known, or those places where no final test has given the last word to the discussion, there remain a few inscriptions whose general purport is still a puzzle, some whose characters have not been made out, and some which, though every character is known, refuse to yield any intelligible combinations. Chief of these are (1.) the longest inscription in the Cesnola collection, in part quite plain, but presenting some unique difficulties, though there is scarcely a doubt as to the reading

* In this article the names 1. of the cases are abbreviated by their initials in capitals (N. G. D. A.); designations 2. of number (s. p.), and 3. of gender (m. f. n.), by their initials in small letters.

of any character; (2.) The Naked (or Bearded) Archer inscription of the British Museum, which, in spite of Dr. W. Deecke's attempt published in the Athenaeum of May 22, 1880, and thought by him a complete success, is yet a problem to be solved, since his reading depends upon the arbitrary forcing of several characters to read what they demonstrably cannot ; (3.) one other important inscription in the British Museum, which is still the subject of discussion, though parts of it are plain; and (4.) a long inscription in the new Cesnola collection, still unread, though it will probably yield to perseverance, and gives promise of solving finally one riddle; with (5.) some minor ones in nearly the same state as the last.

Thus any attempt to construct a grammar or vocabulary of the language, or writing, must still be only tentative, and subject to correction.

The Cypriote syllabary consists of characters representing open syllables only. Of these, five represent the simple vowels, a, e, i, o, u, with no distinction between long and short. The others represent a single consonant followed by a vowel; with no distinction between smooth, middle, and rough in the consonants, and no distinction between long and short in the vowels. Thus the same character stands for Kε, γε, χε, κη, γη, χη. As far as thus discovered, the syllabary is complete for the consonants k, t, p, l, r, m, 8; that is, it has a character for each of these consonants in combination with each of the five vowels. The n series lacks only nu; the z series lacks zi, zo, zu, and the syllable ze rests on only one example, and therefore only on probable conjecture. The y or i-syllables (for the writing has a yod mobile, or consonant) lack the io and iu, perhaps as not required by the language. The digamma or w-syllables, lack wi and wu. Thus, in theory, the syllabary is nearly complete. There is one other character, read by Deecke, Siegismund, and myself as xe, because, though thus an anomaly in the system of writing, it appears to have no other Greek equivalent. For what we may call the same reason, Dr. Ahrens reads it as equivalent to a vowel with the Greek sumpi, or the Hebrew shin. According to either view the character is an anomaly in the structure of

the syllabary; and so, for that matter are the z-syllables, unless we abandon the idea that the was in any sense a double consonant. With the exception of this character for re or she, to speak provisionally, the Greek double consonants are represented in the Cypriote writing by two characters.

Since the language, so far as deciphered, is Greek, the analogy of ordinary Greek cannot be disregarded in transliterating Cypriote. Yet in treating Cypriote as a system by itself a procedure to which it seems fairly entitled—there is obviously much liberty or play allowable in the choice between a long or a short vowel, as well as in the choice between a smooth, a middle, or a rough mute. How far dialectic analogy may or must curb this liberty is a thing obvious enough in many cases, but in other cases quite as obscure. It does not yet appear at what point it is proper to make the choice in the case of the vowels, especially. The transliterator feels that the strictest truth both allows and compels him to adopt the Roman vowel, whose quantity has no visible ear-marks, even at the risk of being thought ignorant of Greek dialects. Moreover, the Cypriote writing has contributed enough in other respects to the knowledge of very ancient Greek peculiarities to justify us in keeping this door open for further light in that particular direction.

As this is not a treatise on the Cypriote epigraphy, it is not necessary to speak at length of the principles of the combination of two characters (or of three, as sometimes happens) into one syllable. These are generally plain enough, and to be understood as a matter of course. It may be mentioned, however, that a word, especially an elided proclitic, has sometimes its end in the middle of a character, so to speak. Also, the n, or Greek, is often not written, and has to be understood. Whether it was pronounced or not in the spoken language, or whether it had passed into an unwritten nasal, or whatever else had become of it, it is not to our purpose here to inquire. The writing never doubles a consonant. Iota adscript, answering to iota subscript in Greek, is sometimes written and sometimes not. The e vowel is often replaced by the i vowel. When a word ends in a consonant,