32 



ON THE MATHEMATICAL PROBABILITY 



There would still remain an odds of five to one against that particular arrangement, which 

 might be readily and most satisfactorily overcome, by supposing that the root in question 

 had been handed down to each language from some older and extinct dialect. That this 

 is the true solution of the problem, is rendered still more likely by the following conside- 

 rations. 



1. Although the Chinese is entirely destitute of grammatical inflections, and it is, there- 

 fore, impossible to subject it to any grammatical comparisons, except such as are merely 

 syntactical, its syntax, as Chev. l'mnsen has demonstrated, is clearly of the Aryan type. 



2. The Mandarin dialect has no final gutturals, but the terminal 4 g is often replaced by a 

 "uttural in the other dialects. Thus, the mandarin word pa ( g, to bind, to tie, becomes pak in 

 Hok-Keen, and is thus naturally associated with Sanscrit pag, paf, Greek, ™r~ xax—mfrv—, 

 Latin, pang, pac, pax, Gcr., Eng., and Sw., pack, in the same way that mag is connected 

 with a , M x-,i,.e r - ma;/—, magn—, &c* IIok-Keen being in the southeast quarter of the: 

 Chinese empire, closer resemblance to the Aryan forms would naturally be; expected, and the 

 discovery of that resemblance greatly strengthens the probability of a common origin. 



3. The broad vowels air often degraded in Chinese, as well as in the Aryan languages. 

 E. g., pa 4 g, ma,g, become respectively pi,g, mi ( g, without losing their primitive meaning or 

 producing any change except a greater specialty of apYdication. So in Latin, pang-o, vm- 

 ping-o; English, rnong-rel, ming-le. 



4. The Mandarin root ma, to add to, to increase, which may either be the primitive 

 from which ma g was derived, or an abbreviated form of the latter word, is found in v /,l„.. 

 The double sense of greatness and seniority, is common both to Latin major, and Chinese 



mag 



From the cumulative evidence that I have thus briefly presented, I am unable to draw 

 any other conclusion than that the Chinese and Aryan words here given, have a, common 

 origin. And finding, as 1 do, that comparisons equally striking, can be made with nearly 

 every Chinese root, I can readily believe, with Bunsen and others, that the Chinese is the 

 oldest language of which we have any record, " the monument of antediluvian speech." 



I have purposely based my calculations upon assumptions the most favorable that I could 

 imagine, for the production of abnormal resemblances. Every deviation in any of the 

 hypothetical postulates, such as new evidence of pre-historical national intercourse, an 

 increase in the number of similar roots which are traceable in different families of languages, 

 well-sustained philosophical generalisations which lead to just methods of philological 

 study, the discovery of a new law of verbal derivation or transformation, while it may 

 have some tendency to complicate the problem, has a, still greater tendency to render its 



* The Hok-Keen word mi.g, an incantation, similarly serves to connect the root ma,g, with (idyr* 



avnv, magus 



&c. 



