68 A GLOSSARY OF LATER 



the presumption is that the Greeks spelled their Words as they pronounced them."^ 

 Consequently, for every change in pronunciation they made a corresponding change in 

 spelling. Had it not been for this natural mode of representing sounds, the very 

 existence of euphonic and of dialectic changes, as also of barbarisms, would now be a 

 mere matter of speculation. 



It must be observed, however, that before the introduction of the vowels H and 12, 

 the characters E and O were either long or short. Further, E and O represented also 

 the diphthongs EI and OT, respectively, when the I and T were neither radical nor 

 characteristic letters.^^^ And as the sound NG (in han(/) had no appropriate character, 

 the Greeks expressed it by N or P; as dvdvKrj or dvdyKr], dvye\o<; or dyje\o<;, ev^o'^ or 



In the seventh century the pronunciation of the vowels and diphthongs had departed 

 considerably from the ancient standard. Thus, all the vowels were isochronous. Fur- 

 ther, H was pronounced like I. As to the diphthongs, AI (short A) was equivalent to 

 E ; AI (long A), to A ; EI, to I ; HI, to I ; 01, to r ; and ni, to il. The mediajval 

 Greeks, however, had too much veneration for the classical language to lay hands upon 

 its orthography. They spelled their words, not as they pronounced them, but as 

 they found them written in the earlier manuscripts. 



Now when from any cause the same sound has more symbols than one, uneducated 

 persons are apt to employ those symbols indiscriminately. And this is precisely what 

 the mediaeval Greek copyists did., Thus, they would write eXaio<i for ekeo'^, epa> for a'lpa>, 

 ■^^prjo-eit for p^p/cret?, 'iSiov for ^Biov, Sva^x^epyj for Bva')(^epel, em? for ?jTt?, Ka9eiaei, for KaOicrei, 

 aai for mere/, dvC^L for avoi^ei. 



Inaccuracies of this sort must be carefully distinguished from those occurring in 

 inscriptions written when words were spelled as they were pronounced ; for the fonner 

 are orthographical mistakes ; the latter are barbarisms, that is, violations of the rules of 

 orthoepy. Thus, when, in a mediaeval manuscript, eXaw? is found for eXeo?, we infer 

 that the transcriber followed his ear rather than his copy ; that is, his pronunciation 

 was good (for that time), but his spelling was bad. When, however, inscriptions of 



^" Compare Quintil. 1, 7, 30 Ego, nisi quod consuetude obtinuerit, sic sciibendum quidque judico, quo- 

 modo sonat. Hie enim usus literarum, ut custodiant voces, et velut depositum reddant legentibus. Itaque id 

 exprimere debeut, quod dicturi sumus. 



Quintilian here is to be considered as expressing opinions common in his age, that is, in the latter part of 

 the first century, and the beginning of the second. 



^*^ See History of the Greek Alphabet, §§ 5. 6. For the pseudo-diphthongs El and OY, see Ibid. 

 § 20, 5, c, and 4, f. 



