2l6 MAN: PAST AND PRESENT. [CHAP. 



five planets with names of like meaning, and a year of 12 months 

 and 30 days, with the same cycle of intercalated days, while 

 several of the now obsolete names of the Chinese months answer 

 to those of the Babylonians. Even the name of the first 

 Chinese emperor who built an observatory, Nai-Kwang-ti, some- 

 what resembles that of the Elamite king, Kuder-na-hangti, who 

 conquered Chaldaea about 2280 B.C. 



All this can hardly be explained away as a mere series of 

 coincidences ; nevertheless neither Sinologues nor Akkadists are 

 quite convinced, and it is obvious that many of the resemblances 

 may be due to trade or intercourse both by the old overland caravan 

 routes, and by the seaborne traffic from Eridu at the head of 

 the Persian Gulf, which was a flourishing emporium 4000 or 5000 

 years ago. 



But, despite some verbal analogies, an almost insurmountable 

 difficulty is presented by the Akkadian and Chinese languages, 

 which no philological ingenuity can bring into such relation as is 

 required by the hypothesis. Mr T. G. Pinches has shown that 

 at a very early period, say some 5000 years ago, Akkadian already 

 consisted, "for the greater part, of words of one syllable," and was 

 "greatly affected by phonetic decay, the result being that an 

 enormous number of homophones were developed out of roots 

 originally quite distinct 1 ." This Akkadian scholar sends me a 

 number of instances, such as tu for tnra, to enter; ti for tila, 

 to live ; du for dumu, son ; du for dugu, good, as in Eridu, for 

 Gurudugu, "the good city," adding that "the list could be ex- 

 tended indefinitely 2 ." But de Lacouperie's Bak tribes, that is, 

 the first immigrants from south-west Asia, are not supposed to 

 have reached North China till about 2500 or 3000 B.C., at which 

 time the Chinese language was still in the untoned agglutinating 

 state, with but few monosyllabic homophones, and consequently 

 quite distinct from the Akkadian, as known to us from the 

 Assyrian syllabaries, bilingual lists, and earlier tablets from Nippur 

 or Lagash. 



Hence the linguistic argument seems to fail completely, while 



1 " Observations upon the Languages of the Early Inhabitants of Mesopo- 

 tamia," in Journ. R. As. Soc. xvi. Part 2. 



2 MS. note, May 7, 1896. 



