Sanscrit Writing and Language. 109 



brought clearly to perceive, how very slowly the art in which they originated, 

 must itself have advanced. The inquiry, indeed, is even in reference to dic- 

 tionaries alone, worthy of some attention ; and it will besides serve to point out 

 what very little right the Brahmans have to claim any share in the credit of 

 having originally and independently made out this invention, — an invention 

 which they could scarcely have even yet arrived at, if they had been left solely 

 to the resources of their own ingenuity. 



The most ancient works of which accounts have reached our days, having any 

 relation to the nature of dictionaries, were commentaries of Greek grammarians 

 on single authors, explaining the " words," used by each author, which had in 

 the course of time become obscure ; whence they were called yXaxraai. Such, 

 for instance, were the glosses of Homer, of Aristophanes, of Hippocrates, of 

 Plato. Some of these, as for example, Homeric glosses, are still extant ; but it 

 belongs to the very nature of such works in the course of successive ages to 

 receive continual accessions ; first additional " words" are inserted in the margin, 

 and, in the next transcription. Introduced into the text ; then again the margin 

 is filled, and again the text is swelled ; and so on. They cannot therefore, in 

 the state in which they are now found, be depended on as preserving any resem- 

 blance to their original form. Next came into use glosses for whole classes of 

 writers, as for instance, the poetic, the dramatic, the rhetoric, the philosophic, the 

 medical glosses. These also were called yXcoaa-ai ; afterwards, by a more 

 general denomination, Ae'^ety; and, when they came to be alphabetically ar- 

 ranged, Ae^ety Kara (TTOL^a.ov. The third great step in the approach to a 

 dictionary was made by Diogenlanus, a grammarian, placed by Suidas in the 

 second century, who is recorded by Hesychlus to have brought together in 

 alphabetic order all the words found in all the preceding collections, whether of 

 the first or second kind, and thus to have formed a compilation from the Homeric, 

 the Comic, the Tragic, the Lyric, the Rhetoric, the Medical, the Historic 

 glosses. Suidas, Indeed, states this work to have been an epitome of an older 

 one, the joint production of Pamphllus and Zopyrion ; but Hesychlus, who lived, 

 probably, in the fourth or fifth century,* and certainly much nearer than Suidas 

 to the age of Diogenlanus, is more to be relied on. 



* Albert, in the preface to his edition of Hesychius — after showing that nothing certain is known 

 as to the age in which this author lived, further than its being subsequent to the times of the several 



