886 REPORT— 1887. 



jilso a record of the ideas and beliefs, the emotions and the hopes of the past gene- 

 rations of the world. In spoken language, accordingly, we may discover the 

 fossilised records of early humanity, as well as the reflection of the thoughts that 

 move the society of to-day. What fossils are to the geologist words are to the 

 comparative philologist. 



But we must be careful not to press the testimony of language beyond its 

 legitimate limits. Language is essentially a social product, the creation of a com- 

 munity of men living together and moved by the same wants and desires. It is one 

 of the chief bonds that bind a community together, and its existence and develop- 

 ment depend upon the community to which it belongs. If the community is 

 changed by conquest or intermarriage or any other cause the language of the com- 

 munity changes too. The individual who quits one community for another has 

 at the same time to shift his language. The Frenchman who naturalises himself in 

 England must acquire Engli.sh ; the negro who is born in the United States must 

 adopt the language that is spoken there. 



Language is thus a characteristic of a community, and not of an individual. 

 The neglect of this foct has introduced untold mischief not only into philology, but 

 into ethnology as well. Eace and language have been confused together, and the 

 fact that a man speaks a particular language has too often been assumed, in spite of 

 daily experience, to prove that he belongs to a particular race. When scholars had 

 discovered that the Sanskrit of India belonged to the same linguistic family as the 

 European languages, they jumped to the conclusion that the dark-skinned Hindu 

 and the light-haired Scandinavian must also belong to one and the same race. 

 Time after time have I taken up books which sought to determine the racial 

 affinities of savage or barbarous tribes by means of their language. Language and 

 race, in short, have been used as synonymous terms. 



The fallrcy is still so common, still so frequently peeps out where we should 

 least expect it, that I think it is hardly superfluous, even now, to draw attention 

 to it. And yet we have only to look around us to see how contrary it is to all the 

 facts of experience. AVe Englishmen are boimd together by a common language, 

 but the historian and the craniologist will alike tell us that the blood that runs 

 in our veins is derived from a very various ancestiy. Kelt and Teuton, Scan- 

 dinavian and Roman have struggled together for the mastery in our island since it 

 first came within the horizon of history, and in the remoter days of which history 

 and traditioi\ are silent archceology assures us that there were yet other races who 

 fought aud iiingled together. The Jews bave wandered through the world adopt- 

 ing the languages of the peoples among whom they have settled, and in Transyl- 

 vania they even look upon an old form of Spanish as their sacred tongue. The 

 Cornishman now speaks English ; is he on that account less of a Kelt than the 

 Welshman or the Breton ? 



Language, however, is not wholly without value to the ethnologist. Though a 

 commou language is not a test of race, it is a test of social contact. And social 

 coutact may mean — indeed very generally does mean — a certain amount of inter- 

 marriage as well. The penal laws passed against the Welsh in the fifteenth century 

 were not sufficient to prevent marriages now and then between the Welsh and 

 the English, and in spite of the social ostracism of the negro in the Northern States 

 of America intermarriages have taken place there between the black and the white 

 popidation. But in the case of such intermarrying the racial traits of one member 

 only of the union are, as a general rule, preserved. The physical and moral type of 

 the stronger parent prevails in the end, though it is often not easy to tell before- 

 hand on which side the strength will lie. Sometimes, indeed, the physical and 

 moral characters are not inherited together, the child following one of his parents 

 in physical type while he inherits his moral and intellectual qualities from the other. 

 But even in such cases the types preserve a wonderful fixity, and testify to the 

 difficulty of changing what we call ttie characteristics of race. 



Herein lies one of the most obvious differences between race and language, a 

 difference which is of itself sufficient to show how impossible it must be to argue 

 from the one to the other. While the characteristics of race seem almost indelible, 

 language is as fluctuating and variable as the waves of the sea. It is perpetually 



