[256] See this argument strongly maintained in Giese (Ueber den Æolischen Dialekt, sect. 14. p. 160, seqq.). He notices several other particulars in the Homeric language,—the plenitude and variety of interchangeable grammatical forms,—the numerous metrical licenses, set right by appropriate oral intonations,—which indicate a language as yet not constrained by the fixity of written authority.
The same line of argument is taken by O. Müller (History of the Literature of Ancient Greece, ch. iv. s. 5).
Giese has shown also, in the same chapter, that all the manuscripts of Homer mentioned in the Scholia, were written in the Ionic alphabet (with Η and Ω as marks for the long vowels, and no special mark for the rough breathing), in so far as the special citations out of them enable us to verify.
[257] Nitzsch and Welcker argue, that because the Homeric poems were heard with great delight and interest, therefore the first rudiments of the art of writing, even while beset by a thousand mechanical difficulties, would be employed to record them. I cannot adopt this opinion, which appears to me to derive all its plausibility from our present familiarity with reading and writing. The first step from the recited to the written poem is really one of great violence, as well as useless for any want then actually felt. I much more agree with Wolf when he says: “Diu enim illorum hominum vita et simplicitas nihil admodum habuit, quod scripturâ dignum videretur: in aliis omnibus occupati agunt illi, quæ posteri scribunt, vel (ut de quibusdam populis accepimus) etiam monstratam operam hanc spernunt tanquam indecori otii: carmina autem quæ pangunt, longo usu sic ore fundere et excipere consueverunt, ut cantu et recitatione cum maxime vigentia deducere ad mutas notas, ex illius ætatis sensu nihil aliud esset, quam perimere ea et vitali vi ac spiritu privare.” (Prolegom. s. xv. p. 59.)
Some good remarks on this subject are to be found in William Humboldt’s Introduction to his elaborate treatise Ueber die Kawi-Sprache, in reference to the oral tales current among the Basques. He, too, observes how great and repulsive a proceeding it is, to pass at first from verse sung, or recited, to verse written; implying that the words are conceived detached from the Vortrag, the accompanying music, and the surrounding and sympathizing assembly. The Basque tales have no charm for the people themselves, when put in Spanish words and read (Introduction, sect. xx. p. 258-259).
Unwritten prose tales, preserved in the memory, and said to be repeated nearly in the same words from age to age, are mentioned by Mariner, in the Tonga Islands (Mariner’s Account, vol. ii. p. 377).
The Druidical poems were kept unwritten by design, after writing was in established use for other purposes (Cæsar, B. G. vi. 13).
[258] Mr. Fynes Clinton (Fasti Hellenici, vol. i. pp. 368-373) treats it as a matter of certainty that Archilochus and Alkman wrote their poems. I am not aware of any evidence for announcing this as positively known,—except, indeed, an admission of Wolf, which is, doubtless, good as an argumentum ad hominem, but is not to be received as proof (Wolf, Proleg. p. 50). The evidences mentioned by Mr. Clinton (p. 368) certainly cannot be regarded as proving anything to the point.
Giese (Ueber den Æolischen Dialekt, p. 172) places the first writing of the separate rhapsodies composing the Iliad in the seventh century B. C.
[259] The songs of the Icelandic Skalds were preserved orally for a period longer than two centuries,—P. A. Müller thinks very much longer,—before they were collected, or embodied in written story by Snorro and Sæmund (Lange, Untersuchungen über die Gesch. der Nördischen Heldensage. p. 98; also, Introduct. pp. xx-xxviii). He confounds, however, often, the preservation of the songs from old time,—with the question, whether they have or have not an historical basis.