400 Proceedings of the Royal Society 
by the most obvious and easy interflow of their vocal elements, can, 
according to the experience of all languages, issue in no reliable re- 
sult. Pronunciation is a matter of usage, not of argument. 
VI. What is called the modern G-reek pronunciation, or that 
used by the G-reeks speaking their own language at the present 
day, is not modern in our sense of that word — it is Byzantine and 
Alexandrian ; and in its most characteristic elements as old as the 
oldest G-reek manuscripts now existing — as old, we may say more 
correctly, as the general body of the ante-Nicene theology. That 
it is a corruption from the oldest classical pronunciation is self- 
evident, from the fact of its giving the slender sound of l to half a 
dozen different vowels and diphthongs. But this is only what may 
be said with equal truth of French, English, and other languages, 
which have passed through various stages of culture during succes- 
sive centuries. Their present pronunciation is, in many important 
points, a corruption of that which was originally the rule. 
VIT. In these circumstances, the practical question is not without 
difficulty how far the Byzantine Gfreek pronunciation should be 
acknowledged by those nations who, like the Scotch, have the 
happiness to use a pronunciation of the vowels and diphthongs, on 
the whole exhibiting a pretty fair approximation to what can be 
certainly known about the true pronunciation of G-reek in the 
palmy days of Attic eloquence. It is certain, for instance — or at 
least extremely likely, for the modern G-reeks show sturdy fight on 
this point — that the Scottish pronunciation of t], as a prolonged e, 
like the long English a in mate^ is classical, while the English and 
Byzantine pronunciation of that vowel as a long i is fundamentally 
false. Nevertheless, I lean to the opinion recently announced by 
the French Academy, that the Byzantine pronunciation, notwith- 
standing some obvious defects, should be accepted as a general 
basis for the pronunciation of the vowels and diphthongs by all 
European scholars ; and that for these four reasons : — (1.) It is not 
a modern innovation, but a historical fact of nearly two thousand 
years’ duration in its main points, and must be known to the student 
of early manuscripts, as the only true key to a whole class of blun- 
ders made by the early transcribers. (2.) Though a corruption in 
some points, it is a characteristic corruption, and, in fact, only the 
development of a marked national tendency — a tendency well 
