[5 go] 
In the Latin anu Englifh languages we have no diverfity of 
tenfes, or of fingle words, to exprefs whether a paft ation has been 
done lately, or a long time fince. It is only from the tone of the 
fpeaker, from the circumftances of the event, or from the context, 
that we can find out the difference. I fave done it, fect, does 
not tell us whether it was done this inftant, or in the com- 
mencement of the fpeaker’s life, fifty years ago. But if the 
agent enters in hafte and perturbation, and fays, I have done 
it; from his looks and accents, and the circumftances of the 
time, we colle@ that he has done it the inftant before: it is the 
prefent perfec. If, on the other hand, he fays, J have done fuch 
things in my youth, we know that there is a confiderable interval 
between the doing of them and the prefent zra. In the paflive, 
the difference is manifefted in words, zt zs done, it has been done, 
but not fo in the active. To fupply this defe&t, which the Englifh 
and Latin languages labour under, in their active voice, in not diftin- 
guifhing, by different founds or words, the difference between what 
was lately perfected and what fome time ago, and in the paffive, 
in not making this diftin@ion but by the help of the auxiliary 
verbs, the Greeks feem to me to have invented their firft aorift, 
and to have intended by it to indicate the latter, as the preter- 
perfe& did the former. 
Tue opinion of Theodorus Gaza, as quoted by Lord Monboddo, 
with refpe& to the meaning of the preterperfect, in fome meafure 
coincides with this theory, which is alfo ftrengthened by the 
obfervation made in the effay upon the origin of languages, that 
the preterperfet was called wapaxeievoc, as being a time near to 
the prefent. But without relying upon this argument, I fhall 
proceed 
