TWO SEMITIC PROBLEMS 93 



adjectives), the substantive modifications are expressed by prefixes 

 (as ma and to in nouns). Of the precise history of the development of 

 triliteral stems we know very little. It may be said to be probable 

 that these have arisen from biliterals: biliteral stems exist in Semitic, 

 and certain triliterals have obviously been formed from biliterals (as 

 where the second radical is doubled or a w or y is inserted). If this 

 be the case, the expansion has not stopped at triliterals. Quadri- 

 literal simple stems have been formed, and a large number of 

 derived stems, which express merely a modification of the meaning 

 of the simple stem. It is by means of prefixes that the Semitic lan- 

 guages have chosen to express the idea of reflex and causative action, 

 the former by na and to, the latter by a (ha) and sa. These prefixes 

 are variously combined together and variously attached to the stem. 

 Thus with na from katala we have inkatala (for nakatald), and niktal 

 (for naktal). The mode of combination is merely a matter of euphony. 

 The essential thing is that the signification of the simple stem is 

 modified in a substantive way by means of a prefix. This modifica- 

 tion attaches no new content to the signification of the stem, but only 

 indicates a certain direction of the action. 



When now we find the Imperfect made from the simple stem by 

 a prefix, the question arises whether it belongs in the same category 

 with the derived stems above referred to. If we compare these 

 derived stems with the Imperfect, we discover an analogy in the 

 two cases, though with a difference. Niktal differs from katal merely 

 in that it indicates that the act is directed by the subject toward 

 himself; aktala in like manner differs from katala only in that it 

 states that the action is not performed, but is caused to be per- 

 formed, by the subject. So yaktulu differs from katala in affirming 

 not simply that the subject performs the act, but that he moves 

 toward the performance of the act. How such distinctions may 

 inhere in these prefixes it is not necessary to inquire; we do not 

 know their origin or the history of their development. It is sufficient 

 that they have these significations. Thus the Imperfect is naturally 

 based on the Perfect, or rather on the stem which is the base of the 

 Perfect. Both in form and in meaning it comes after the Perfect. 

 It is a peculiar Semitic formation, ignoring the element of time, and 

 choosing only to distinguish between the conception of an act pure 

 and simple (Perfect) and an act toward which the subject moves. 

 This grammatical conception has its advantages and its disadvan- 

 tages logically and rhetorically. But it falls in with the Semitic 

 system of expressing certain modifications by means of prefixes. 

 The Imperfect may attach itself to any stem, simple or derived, 

 triliteral or quadriliteral. Originally, it would seem, the Imperfect 

 of the simple stem assumed two forms: which, if we leave out the 

 later vowel distinctions, may be written yakatala and yaktala. Of 



