SOURCES OF MELANESIAN MATERIAL. 5 
Since we have mentioned the numerical sum of one of the diction- 
aries of Melanesia, we may properly give passing consideration to the 
size of the vocabulary of these savages. From time to time it has 
interested popular fancy to compute the average vocabulary of the 
lower classes in our civilization. "There seems to have been in this 
amusement no particular precision in the establishment of what shall 
be regarded as a vocabulary unit to be counted. In general it has 
been proposed that the English hind—for in the United States a more 
rigid compulsion in elementary education is operative against these 
speculations—has a working vocabulary of not much beyond a thou- 
sand vocables, and by increments of 500 at each step the higgler, the 
small tradesman, and the tinker, who is always an outland man and 
increases culture by voyaging along the broad highway, may attain 
to the high sum of 2,500 words with which to traffic in the affairs of 
life as it presents itself to him. Be this as it may, and it is scarcely 
worth our while to regard the speculation as other than curious, the 
condition among the Melanesian savages is radically different. He 
knows no social gradation of education; in his society there is no dis- 
tinction between the learned and the uncouth; in effect, that which 
anyone knows is known by all; the only difference in vocabulary is 
that which must exist between the inexperience of youth and the 
stores of the aged and which reaches its highest point in a few very 
ancient men who retain names of former customs which have passed 
out of use in the advance of progress. We have seen that the Mota 
dictionary, avowedly incomplete, contains 5,000 vocables. In the 
course of recent study directed upon one of the least-known languages 
of the Philippines, I had occasion to compute the extent of the vocabu- 
lary of a circumjacent and to a certain extent allied folk, the Visayas. 
In the Visayan dictionary compiled by Fr. Juan Félix de la Encarna- 
cién we find 12,000 vocables, and the Visayas are scarcely more 
advanced upon the road of civilization than the men of Mota or other 
of the islands of Melanesia. Accordingly we feel justified in appraising 
the best of our working vocabularies of Melanesia as representative 
of barely 50 per cent of the languages with which they deal, and from 
that highest level we deal with material which dwindles rapidly down 
to the level of mere word lists. 
Thus we see how imperfect is the equipment with which we may 
give value to the third stage in which our Melanesian material has 
been presented to use, that which has assumed the form and method 
of comparative philology. Some of the material in the second class 
is prepared in such a manner as to fit it for consideration in this third 
class; some even of the primal class of discovery record comes to us 
equipped with the machinery of comparison. The dictionary of 
Efaté affords an excellent instance of the mingling of the second and 
third classes. The prime object of the author appears to have been to 
