Page 92

Page 92

stiohometry THE NEW SCHAFF-HERZOG 82

point, that such varieties of measurement, if they exist, are extremely rare. In regard to the actual reduction of a prose-passage to its equivalent verselength, there is an important case in Galen (v. 655, ed. Kiihn), where, having quoted a sentence from Hippocrates, he continues:

elg ltev ovros E UyoS evvla Kdt rptaKOVra avX7<aairv 8arep eari dvoiv Kai ~yfaerrs e7rw tfalterpcrv Kre.

If Galen, according to this, then reckons thirtynine syllables as being equivalent to two hexameters and a half, or, as he continues, eighty-two syllables to five hexameters, the hexameter can hardly be different from a sixteen-syllabled rhythm. The assumption is easy that stichometric measurement is made by preference in syllables of which sixteen go to the hexameter, or unit-verse. The number sixteen invites attention as being the number of syllables in the first line of the Iliad, and as being a square number, a peculiarity which always had a certain attractiveness for early calculators. That the term stichos deflects in the direction of hexameter verse as against any other line of poetry which might have been chosen for a proper unit of measurement, will appear from Montfaucoa (Bibl. Coislin., p. 597), where there is quoted from a tenth-century manuscript a catalogue of poets as writers by stichoi, and writers of iambics can only have resulted from a specialization of the meaning of the term stichos by constant use in a particular sense.

In the demonstration of the same point by actual measurement, the most important researches are those published by the late C. Graux (in Revue de Philologie, Apr., 1878), in which he demonstrated, by an actual estimation of the number

3. This of letters in certain works, that the

Measure- stichos represented not a clause, nor a. meat number of words, but a fixed quantity

Confirmed. of writing. The average number of letters to the verse he found to vary between narrow limits, generally thirty-four to thirty-eight letters; and an enumeration of the letters in fifty lines of the Iliad opened at random supplied him with an average of 37.7 letters to the verse. This very important identification of the stichos with the hexameter is the starting-point for a great many new critical investigations as to the integrity of transmitted texts, their early form, etc. Whether the unit of measurement is a certain num ber of syllables, or a certain number of letters, is not at first sight easy to decide. It is tolerably cer tain that the measured line is, as above stated, a space-line, and not a sense-line; but to discriminate between a letter-line and a syllable-line is a more delicate matter. If the former be adopted, the unit should probably be fixed at thirty-six letters, be cause this is the nearest symmetrical number to the average hexameter. There are very few instances, however, in which the actual letters of a line are found to be numbered; while the custom can readily be traced of limiting a line by the division of the syllables, in the earliest manuscripts. More over, there is the actual measurement in the passage quoted from Galen; and Pliny seems to allude to the custom of syllable-counting, when, in one of his epistles, he demands an equally long reply from

his correspondent, and threatens to count, not only the pages, but the verses on the page, and the syllables of each verse (Ego non paginas tarctum, sed versus edam syllabasque numerabo; Pliny, iv. 11). The preference must, therefore, be given to the syllable-line. It is comparatively easy to count the copmass of a book in sixteen-syllable rhythms, but a toilsome process to estimate with equal accuracy the number of thirty-six-letter lines.

It is interesting to compare the relative sizes of the two line-units. bI. Graux deduces 37.7 as the average hexameter in letters, and Diels (Hermes, vol, xvii.) makes the average of the first fifty lines in Homer to be 15.6 syllables. A verse

4. Partial of sixteen syllables is then equivalent Stichometry.to about 1.074 verses of thirty-six letters each. In precisely the same way as M. Graux determined the average number of letters to the verse from the total stichometry, in the manuscripts of Herodotus, Demosthenes, Eusebius, Gregory of Nazianzus, etc., one may examine the partial stichometry. This has been done for Tsoerates by Fuhr (Rheinisches Dluseurrt, xxxvii. 468); for the Plato manuscripts, by Schanz (Hermes, xvi. 309); and for the Demosthenes manuscripts, by W. v. Christ, in the able discussion entitled Die Atticusausgabe des Demosthenes (Miinehen, 1882). The partial stichometry is of the highest value for the study of texts; and in every case the data which it supplies are found to accord very closely with the fundamental statements above as to the paleographical meaning of the word stichos. There are traces of partial stichometry in the great Vatican manuscript of the Old and New Testaments (cf. E. Nestle, in Correspondent-Blatt fur die Gelehrten and Realschulere Wiirttembergs, 1883; and J. R. Harris, Stichometry, pp. 59-64, London, 1893). The foregoing investigations received striking and unexpected confirmation through the discovery, by Professor Mommsen in 1885 of a list of the canonical books of the Old and New Testaments and of the works of Cyprian in the Phillipps Library at Cheltenham. These lists were accompanied by . stichometric annotations, to which the scribe attached the information that the index of verses in the city of Rome is not clearly given, and elsewhere, through greed of gain, they do not preserve it in full; but that he went through the books in detail, counting sixteen syllables to the line; according to the standard line of Vergil, and appended the number of verses. The importance of this statement is evident. There was not only a stichometry of the Vulgate and of the works of Cyprian by which the purchaser of books in Carthage or elsewhere could be protected against the rapacity of the bookseller, but the hexameter standard was clearly defined as the unit of measurement.

Some degree of confusion is introduced by the existence, apparently, in early times, of an alternar tive iambic verse of twelve syllables, g. Cola and as well as by the introduction of wri-

Commata. ting by cola and commata. The latter of these points has been an especial ground of combat, in consequence of the counte nance which the custom seemed to lend to the theory of sense-lines in opposition to space-lines. The ex-