/38 INTERNATIONAL CONGRESS OF EDUCATION. 
are spelt with a ¢ (such as “corner,” “come,” ¢ crown,” and the like). 
And so the list of % appears small, and the conclusion is hastily 
irawn that this is a difficult sound for the child to make because he has 
but few words beginning with it as initial letter. Several other false con- 
clusions of a similar nature are reached. It is very desirable that those 
who classify child words henceforth should take the greatest pains to 
arrange them no¢ alphabetically, but phonetically. 
[n a careful study of some twenty-five child vocabularies, comprising 
aearly six thousand words, some facts have come to light different from 
vhat might ordinarily be expected, as a result of this phonetic arrange- 
ment. The % sound turns out to be not at all a difficult sound for the 
average child. As an initial sound it occupies third place in order of 
frequency, standing indeed above m, which is usually supposed to be one 
»f the easiest of sounds. 
[ have several times met with the remark that the young child learns to 
speak largely by watching the lips of those who speak in his presence, and 
imitating the movements which are there visible. Hence, on this theory, 
those sounds will be best and earliest learned the movements involved in 
which are most plainly visible, and therefore most easily imitated, such as 
she labials, for example. This theory is not at all borne out by my obser- 
vations. The % sound, whose movements are absolutely out of sight, is 
earned more readily than some other sounds whose movements are 
plainly visible, such as #%.in ¢ think.” 
So far as I am able to judge, the earliest vocal movements are not imi- 
tative at all, or only so to a very limited extent, and are what Preyer 
would call impulsive movements. The child utters sounds before he is of 
an age to be able to appreciate and imitate the sounds and movements of 
others. He combines these sounds afterward in imitation of others, but 
she faculty of imitation seems to play little or no part in the first begin- 
nings of infant articulation. 
The consideration of the relative frequency of the different parts of 
speech in the vocabulary of the child also yields some interesting and 
raluable results. 
Among those who have studied the language of children it has generally 
been supposed that substantives, names of things, are more readily learned 
shan predicative words, inasmuch as they are usually more numerous than 
:he latter in the speech of children. But it is obvious that we must take 
into account not only absolute but relative frequency ; that is, we must 
not only compare the number of nouns in the child’s speech with the 
aumber of verbs, but also compare the speech of the child with that of the 
adult, and prove that he has acquired a larger proportion of his future 
substantive vocabulary than of his future verb vocabulary. But according 
to the child words examined by me, the opposite is the case. In the 
vocabulary of the child of two years of age (taking the average of my