September 25, 1891.] 



SCIENCE. 



175 



geal, causes a copious secretion of mucus. In the last case there 

 is no secretion of mucus when the vagi are {livided. Perhaps the 

 result is due to a reflex secretion, the reflex centre bein? in the 

 bulb, while the vagus is the efferent channel for the impulses 

 affecting the secretory glands. In birds also it would appear that 

 the vagus influences the secretion of gastric juice. Oxenfeld 

 finds that in birds fpigeons) stimulation of the peripheral end of 

 :the vagus is followed by a copious secretion of acid gastric juice. 

 At the same time the stomach is forcibly contracted, and it might 

 be assumed that the increased quantity of gastric juice was simply 

 forced out of the glands by the concentration of the musculature 

 of the stomach. Oxenfeld, however, is of opinion that this is not 

 the true explanation, and he assumes that the vagus contains se- 

 cretory fibres for the gastric glands. 



LETTERS TO THE EDITOR. 



^*t Correspondents are requested to be as brief as possible. The writer^s name 

 is in all cases required as proof of good faith. 



On request in advance, one hundred copies of the number containing his 

 <ionimunication will be furnished free to any correspondent . 



The editor will be glad to publish any queries consonant mith the character 

 of the journal. 



HovT- Children Learn to Talk.— A Study in the Development 



of Language. — Children's Vocabularies. 



Philologists and others interested in the origin of language and 

 the development of intellect find very striking analogies between 

 the development of speech and intelligence in the race and in the 

 child, and have obtained some very valuable hints as to the laws 

 ■determining the growth of language. Scientific psychologists 

 and educators have also gained many important truths from the 

 study of children. A much more extensive and detailed study, 

 however, is now necessary to further progress in either line. 



The first thing to be done in every scientific investigation is to 

 collect a large number of reliable facts, from which generaliza- 

 tions may be made and theories found that will guide the investi- 

 gator in further researches, and lead to the discovery and unifica- 

 tion of a general law of nature that can usually be turned to 

 practical account by the inventor, educator, or legislator. Facts 

 of every kind in regard to development of intelligence in children 

 and their progress in language are important, and in the earlier 

 stages of the investigation the common and ordinary facts rather 

 than the unusual and extraordinary are the most valuable. For 

 the purpose of securing such facts and arousing interest in the 

 study of children I wrote an article some months ago entitled 

 ■" Children as Teachers," and published it in a number of papers. 

 The records sent me in response to the request in that article are 

 very interesting and suggestive. Many interested in the subject, 

 however, doubtless overestimated the difficulty of securing valua- 

 ble records, and therefore I have not yet received a sufficient num- 

 ber of records to justify me in making a full report, as I had 

 promised, at present. The records so far examined serve to bring 

 out the great individual differences in children rather than to 

 show what is common to all, yet they are common chai-acteristics 

 suggestive of general laws sufficient to confirm me in the belief 

 that a comparison of a number of such records will give very val- 

 uable results. 



A convenient method for those who cannot keep a daily record 

 of a child's progress in language was adopted by some reporting 

 to me. During a certain period special attention was paid to the 

 child's language, and all words the child was known to use un- 

 derstandingly were noted down in alphabetical order (the child's 

 pronunciatioQ of the words being indicated as nearly as possible), 

 and this was taken as the child's vocabulary at that age. A few 

 months later the process was repeated, and the progress that had 

 been made could then readily be seen by comparing the two 

 records. 



The number of words used by children two years old differs 

 considerably, but is usually larger than parents supposed. The 

 number varies from a very few words for the child who is back- 

 ■ward in learning to talk, though perhaps not less intelligent other- 

 wise, up to a thousand words for children more precocious in that 



particular. Judging from the records in my possession, from two 

 to four hundred words is the more common number. 



The rate at which new words are acquired varies greatly for 

 different children and at different ages. After they are once 

 fairly started in learning language, it is usually quite rapid, es- 

 pecially with those who are late in beginning to talk. For children 

 just past two years of age, from sixty to one hundred words per 

 month seems to be a common number. If new words should con- 

 tinue to be acquired at this rate until maturity, as they probably 

 are by those who study and read much, an adult would have a 

 vocabulary of from 15,000 to 35,000 words (see "Size of an 

 Ordinary Vocabulary," Science, Aug. 31, 1891). The additional 

 words used by a child do not represent all of his progress in lan- 

 guage. He may have learned the meaning of many words he has 

 had no occasion to use; he may have learned something about 

 forming plurals and the different parts of verbs, and considerable 

 about how to put words together in sentences. The progress in 

 the latter respect may be shown by keeping a record of his charac- 

 teristic attempts at sentence making, being careful to omit sen- 

 tences that are evidently repeated from memory. 



The part of speech most used by children seems to be the noun. 

 About 60 per cent of the words in the English language are 

 nouns, 33 per cent adjectives, 11 per cent verbs, and 5+ per cent 

 adverbs, while conjunctions, prepositions, and pronouns form but 

 an insignificant portion of the whole. In an ordinary vocabulary, 

 taking " Robinson Crusoe " as the standard, the proportion of 

 nouns is smaller, and in a still smaller vocabulary there seems to 

 be occasion for the use of a greater variety of verbs than nouns, 

 and a necessity for the use of a number of prepositions, pronouns, 

 and conjunctions. On a page of " Robinson Crusoe " containing 

 315 different words, but 24 per cent were nouns. Hence the fact 

 that in a child's vocabulary of a few hundred words from 55 to 85 

 per cent of them are nouns, while but few of the prepositions, 

 pronouns, and conjunctions that it hears repeated so frequently 

 are used, is quite significant. Nouns, however, are not always 

 learned easier and earlier than other parts of speech, for such a 

 verb as "come," or adjective as "hot," may be among the first 

 words learned. Any word which can be associated with a dis- 

 tinct, sensible experience can readily be learned, but abstract 

 terms are not found in children's vocabularies. 



General terms are used by children, at quite an early age, with 

 some degree of correctness, though of course all that is connotated 

 or included under a general term is not understood by any one 

 until its scientific meaning is known. A general term is applied 

 to all individuals having certain characteristics, though they may 

 differ in other respects. The accuracy with which a child uses 

 general terms depends upon the distinctness of his ideas of the 

 special characteristics to which the term is applied, and his power 

 of noting and discriminating those special qualities among a vari- 

 ety of others. His attainments in these two respects are limited 

 by his previous experience. A child who calls a goat a ''dog" 

 may lack in clearness of conception of the characteristics of dogs, 

 or in his powers of discrimination, or only in experience. In the 

 latter case he classifies it with the group of animals it resembles 

 more closely than any others with which he is acquainted. A 

 child of twenty-six months who found a small crab in her oyster 

 soup classified it at once with the group of animals it seemed to 

 her to most resemble, and called it a "bug," then performing a 

 considerable act of inference, she gave it the more definite name 

 "oyster-bug." A little girl of less than eighteen months, who 

 learned the word "cut" in connection with the use of a knife, not 

 only called all knives " cutie," but applied the same term to shears 

 when she saw the same operation performed with them, and later 

 to a sickle with which grass was being cut. Nothing is more in- 

 teresting or important in the study of children than the way in 

 which they generalize, classify, and infer, and instances of such 

 childish judgments and inferences, so odd to us, yet really so 

 natural and logical from their point of view, should be carefully 

 noted and recorded. 



Children sometimes form a language of their own entirely dif- 

 ferent from that of their parents. This is more likely to occur 

 with children of the same age, especially if they are alone together 

 much. Instances are known of children forming apparently quite 



