work on this subject, from which we are justified in expecting the 

 happiest results. 



Perhaps it is worth while to survey the field for a moment, to 

 specify what one would like to have if wishes were horses, to 

 sketch, however tentatively, a kind of programme. 



In the first place we desire to know thoroughly the Germanic 

 foundations. For this, of course, a complete Anglo-Saxon syntax 

 is necessary: not a mere list of verbs with the cases they govern, 

 or an array of the different ways in which the numerals may be 

 arranged, or a set of statistics comprehending the relative frequency 

 of the weak adjective and the strong. These things are well enough, 

 and we cannot get along without them. But what we must look 

 forward to is something far less mechanical, a system of Anglo-Saxon 

 constructions such as we already have for the classical tongues, 

 discriminated as finely as the nature of the idiom will admit, arranged 

 both logically and historically, complete both for the poetry and the 

 prose, and supported at every point by exhaustive material. This, 

 of course, is not the task of one. man, or perhaps of one generation; 

 but we are not talking about what can be accomplished to-morrow or 

 next day. We are trying to face the problem of English syntax as 

 it stands, probably the most stupendous problem of all those with 

 which it is oilr business to grapple. 



This ideal system of Anglo-Saxon syntax is needed, as we have 

 seen, as a foundation for our whole structure. We must know how 

 the syntax of our language stood when English was a purely Germanic 

 speech, before we can reason with certainty as to what took place 

 when our idiom was subjected to those extraordinarily complex 

 forces which have made it unique among the languages of articulate- 

 speaking men. 



Here, at the very threshold, we are confronted by a difficulty of no 

 small proportions. Since most of our Anglo-Saxon prose is translated 

 from the Latin, we cannot trust its syntactic evidence without 

 careful scrutiny. At every step, therefore, the possibility of foreign 

 literary influence must be borne in mind. We must compare the 

 constructions of poetry, and we must appeal constantly to the 

 testimony of the other Germanic languages. Nor must we forget, in 

 examining the poetical texts, the archaizing tendency of all expres- 

 sion in verse. Finally, we must bear in mind the probability of 

 syntactic differences coincident with differences of dialect, and we 

 must remember the special complications that have resulted from 

 the partial transference of a large body of Anglo-Saxon poetry from 

 one dialect, more or less completely, into another. 



If we can do all this, and we shall be forced to do it somehow and 

 sometime, we shall be in a position to study with intelligence the 

 bewildering syntax of the Middle English period. 



