PREFACE. 



Philology has proved itself an admirable instrument for the classification of 

 nations into families upon the basis of linguistic affinities. A comparison of the 

 vocables and of the grammatical forms of certain languages has shown them to be 

 dialects of a common speech ; and these dialects, under a common name, have thus 

 been restored to their original imity as a family of languages. In this manner, and 

 by this instrumentalrty, the nations of the earth have been reduced, with more or 

 less of certainty, to a small number of independent families. 



Some of these families have been more definitely circumscribed than others. 

 The Aryan and Semitic languages have been successfully traced to their limits, and 

 the people by whom they are severally spoken are now recognized as famihes in 

 the strict and proper sense of the term. Of those remaining, the Turanian is 

 rather a great assemblage of nations, held together by slender affinities, than a 

 family in the Aryan or Semitic sense. With respect to the Malayan it approaches 

 nearer to the true standard, although its principal divisions are marked by 

 considerable differences. The Chinese and its cognates, as monosyllabic tongues, 

 are probably entitled upon linguistic grounds to the distinction of an independent 

 family of languages. On the other hand, the dialects and stock languages of the 

 American aborigines have not been explored, with sufficient thoroughness, to 

 determine the question whether they were derived from a common speech. So far 

 as the comparisons have been made they have been found to agree in general plan 

 and in grammatical structure. 



The remarkable results of comparative philology, and the efficiency of the 

 method upon which as a science it proceeds, yield encouraging assurance that it 

 will ultimately reduce all the nations of mankind to families as clearly circum- 

 scribed as the Aryan and Semitic. But it is probable that the number of these 

 families, as finally ascertained, will considerably exceed the number now recognized. 

 When this work of philology has been fully accomplished, the question will remain 

 whether the connection of any two or more of these families can be determined 

 from the materials of language. Such a result is not improbable, and yet, up to 

 the present time, no analysis of language, however searching and profound, has 



(V) 



