OF ACCIDENTAL LINGUISTIC RESEMBLANCES. 



27 



4. The objection to fanciful or accidental resemblances, while it is the one most frequently 

 urged, is also the most plausible, and I have thought it worthy of a, brief examination, 

 which has led to conclusions surprising even to myself, well prepared as I was, by twenty- 

 five years' study of radical philology, to admit the truths of winch those results are the 

 natural indications. 1 had supposed that an argument on which so much stress is laid, 

 oik- which is urged so often and so triumphantly, must have some claims to respect, but I 

 find that a, purely and strictly accidental coincidence* in sound and sense, is nearly, if not 

 altogether, impossible. 



For, in the first place, we must reject from the category of casual resemblance, all words 

 that are determined by fixed and known laws; such, for example, as indicate nothing more 

 Ihan a uniformity of idea, or of vocalization, that is dependent on a uniformity of human 

 nature in its mental and physical constitution, and all words which are clearly or even 

 probably derived from an onomatopeeia. Words of the latter class are probably not so 

 numerous as is often supposed, and. even those which may possibly have had an onomato- 

 poeic source (as e. g. Eng., beef; Fr., boeuf; Lat, bov ; Gr., ,%<_,; Chin., moo), are often 

 so transformed that their origin can only be made evident by a series of connected deriva- 

 tions, which serve to strengthen a, conviction in the original family union of nations that 

 are now most widely separated. 



Fetus them suppose that in two languages Which we wish to make the subjects of 

 comparison, all words of a, merely imitative character, and all other analogues that could 

 be reasonably referred to uniformity of organization, have been discarded. Suppose, 

 moreover, that for other quasi-accidental and unknown reasons, there still remains a degree 

 of resemblance so incredible that there is a precise correspondence between all tin- ideas 

 represented in the two languages, and a, correspondence equally precise in the sound of 

 two-thirds of the words that they employ. Still, tin; word thai; denotes hoi/ in one language, 

 wdl be more likely to denote horse, or anything else rather than boy, in the other, and a 

 single instance of merely fortuitous coincidence will be improbable. If the resemblance 

 is still more striking, so that together with the correspondence of ideas, there is a, precise 

 agreement in the sound of all the syllables and words, then; will probably be a, single 

 coincidence, and no more. 



Some of the resnlts of the following discussion have been anticipated by Dr. Thomas 

 Young (Phil. Trans., 1819, p. 7!), sqq.), but the importance of the subject seemed suffi- 

 cient to warrant a, more general investigation. 



ii the present paper, I use the term coincidence, merelj to denote words in two different languages which agree 

 both in sound and meaning. 



