Language: Its Nature, Development and Origin

X.—§ 1. Confusion of Words.

Some of the most typical childish sound-substitutions can hardly be supposed to leave any traces in language as permanently spoken, because they are always thoroughly corrected by the children themselves at an early age; among these I reckon the almost universal pronunciation of t instead of k. When, therefore, we do find that in some words a t has taken the place of an earlier k, we must look for some more specific cause of the change: but this may, in some cases at any rate, be found in a tendency of children’s speech which is totally independent of the inability to pronounce the sound of k at an early age, and is, indeed, in no way to be reckoned among phonetic tendencies, namely, the confusion resulting from an association of two words of similar sound (cf. above, p. 122). This, I take it, is the explanation of the word mate in the sense ‘husband or wife,’ which has replaced the earlier make: a confusion was here natural, because the word mate, ‘companion,’ was similar not only in sound, but also in signification. The older name for the ‘soft roe’ of fishes was milk (as Dan. mælk, G. milch), but from the fifteenth century milt has been substituted for it, as if it were the same organ as the milt, ‘the spleen.’ Children will associate words of similar sound even in cases where there is no connecting link in their significations; thus we have bat for earlier bak, bakke (the animal, vespertilio), though the other word bat, ‘a stick,’ is far removed in sense.

I think we must explain the following cases of isolated sound-substitution as due to the same confusion with unconnected words in the minds of children hearing the new words for the first time: trunk in the sense of ‘proboscis of an elephant,’ formerly trump, from Fr. trompe, confused with trunk, ‘stem of a tree’; stark-naked, formerly start-naked, from start, ‘tail,’ confused with stark, ‘stiff’; vent, ‘air-hole,’ from Fr. fente, confused with vent, ‘breath’ (for this v cannot be due to the Southern dialectal transition from f, as in vat from fat, for that transition does not, as a rule, take place in French loans); cocoa for cacao, confused with coconut; match, from Fr. mèche, by confusion with the other match; chine, ‘rim of cask,’ from chime, cf. G. kimme, ‘border,’ confused with chine, ‘backbone.’ I give some of these examples with a little diffidence, though I have no doubt of the general principle of childish confusion of unrelated words as one of the sources of irregularities in the development of sounds.

These substitutions cannot of course be separated from instances of ‘popular etymology,’ as when the phrase to curry favour was substituted for the former to curry favel, where favel means ‘a fallow horse,’ as the type of fraud or duplicity (cf. G. den fahlen hengst reiten, ‘to act deceitfully,’ einen auf einem fahlen pferde ertappen, ‘to catch someone lying’).

X.—§ 2. Metanalysis.

We now come to the phenomenon for which I have ventured to coin the term ‘metanalysis,’ by which I mean that words or word-groups are by a new generation analyzed differently from the analysis of a former age. Each child has to find out for himself, in hearing the connected speech of other people, where one word ends and the next one begins, or what belongs to the kernel and what to the ending of a word, etc. (VII § 6). In most cases he will arrive at the same analysis as the former generation, but now and then he will put the boundaries in another place than formerly, and the new analysis may become general. A naddre (the ME. form for OE. an nædre) thus became an adder, a napron became an apron, an nauger: an auger, a numpire: an umpire; and in psychologically the same way an ewte (older form evete, OE. efete) became a newt: metanalysis accordingly sometimes shortens and sometimes lengthens a word. Riding as a name of one of the three districts of Yorkshire is due to a metanalysis of North Thriding (ON. þriðjungr, ‘third part’), as well as of East Thriding, West Thriding, after the sound of th had been assimilated to the preceding t.

One of the most frequent forms of metanalysis consists in the subtraction of an s, which originally belonged to the kernel of a word, but is mistaken for the plural ending; in this way we have pea instead of the earlier peas, pease, cherry for ME. cherris, Fr. cerise, asset from assets, Fr. assez, etc. Cf. also the vulgar Chinee, Portuguee, etc.[38]

The influence of a new generation is also seen in those cases in which formerly separate words coalesce into one, as when he breakfasts, he breakfasted, is said instead of he breaks fast, he broke fast; cf. vouchsafe, don (third person, vouchsafes, dons), instead of vouch safe, do on (third person, vouches safe, does on). Here, too, it is not probable that a person who has once learnt the real form of a word, and thus knows where it begins and where it ends, should have subsequently changed it: it is much more likely that all such changes originate with children who have once made a wrong analysis of what they have heard and then go on repeating the new forms all their lives.

X.—§ 3. Shiftings of Meanings.

Changes in the meaning of words are often so gradual that one cannot detect the different steps of the process, and changes of this sort, like the corresponding changes in the sounds of words, are to be ascribed quite as much to people already acquainted with the language as to the new generation. As examples we may mention the laxity that has changed the meaning of soon, which in OE. meant ‘at once,’ and in the same way of presently, originally ‘at present, now,’ and of the old anon. Dinner comes from OF. disner, which is the infinitive of the verb which in other forms was desjeun, whence modern French déjeune (Lat. *desjejunare); it thus meant ‘breakfast,’ but the hour of the meal thus termed was gradually shifted in the course of centuries, so that now we may have dinner twelve hours after breakfast. When picture, which originally meant ‘painting,’ came to be applied to drawings, photographs and other images; when hard came to be used as an epithet not only of nuts and stones, etc., but of words and labour; when fair, besides the old sense of ‘beautiful,’ acquired those of ‘blond’ and ‘morally just’; when meat, from meaning all kinds of food (as in sweetmeats, meat and drink), came to be restricted practically to one kind of food (butcher’s meat); when the verb grow, which at first was used only of plants, came to be used of animals, hairs, nails, feelings, etc., and, instead of implying always increase, might even be combined with such a predicative as smaller and smaller; when pretty, from the meaning ‘skilful, ingenious,’ came to be a general epithet of approval (cf. the modern American, a cunning child = ‘sweet’), and, besides meaning good-looking, became an adverb of degree, as in pretty bad: neither these nor countless similar shiftings need be ascribed to any influence on the part of the learners of English; they can easily be accounted for as the product of innumerable small extensions and restrictions on the part of the users of the language after they have once acquired it.

But along with changes of this sort we have others that have come about with a leap, and in which it is impossible to find intermediate stages between two seemingly heterogeneous meanings, as when bead, from meaning a ‘prayer,’ comes to mean ‘a perforated ball of glass or amber.’ In these cases the change is occasioned by certain connexions, where the whole sense can only be taken in one way, but the syntactical construction admits of various interpretations, so that an ambiguity at one point gives occasion for a new conception of the meaning of the word. The phrase to count your beads originally meant ‘to count your prayers,’ but because the prayers were reckoned by little balls, the word beads came to be transferred to these objects, and lost its original sense.[39] It seems clear that this misapprehension could not take place in the brains of those who had already associated the word with the original signification, while it was quite natural on the part of children who heard and understood the phrase as a whole, but unconsciously analyzed it differently from the previous generation.

There is another word which also meant ‘prayer’ originally, but has lost that meaning, viz. boon; through such phrases as ‘ask a boon’ and ‘grant a boon’ it came to be taken as meaning ‘a favour’ or ‘a good thing received.’

Orient was frequently used in such connexions as ‘orient pearl’ and ‘orient gem,’ and as these were lustrous, orient became an adjective meaning ‘shining,’ without any connexion with the geographical orient, as in Shakespeare, Venus 981, “an orient drop” (a tear), and Milton, PL i. 546, “Ten thousand banners rise into the air, With orient colours waving.”

There are no connecting links between the meanings of ‘glad’ and ‘obliged,’ ‘forced,’ but when fain came to be chiefly used in combinations like ‘he was fain to leave the country,’ it was natural for the younger generation to interpret the whole phrase as implying necessity instead of gladness.

We have similar phenomena in certain syntactical changes. When me thinks and me likes gave place to I think and I like, the chief cause of the change was that the child heard combinations like Mother thinks or Father likes, where mother and father can be either nominative or accusative-dative, and the construction is thus syntactically ambiguous. This leads to a ‘shunting’ of the meaning as well as of the construction of the verbs, which must have come about in a new brain which was not originally acquainted with the old construction.

As one of the factors bringing about changes in meaning many scholars mention forgetfulness; but it is important to keep in view that what happens is not real forgetting, that is, snapping of threads of thought that had already existed within the same consciousness, but the fact that the new individual never develops the threads of thought which in the elder generation bound one word to another. Sometimes there is no connexion of ideas in the child’s brain: a word is viewed quite singly as a whole and isolated, till later perhaps it is seen in its etymological relation. A little girl of six asked when she was born. “You were born on the 2nd of October.” “Why, then, I was born on my birthday!” she cried, her eyes beaming with joy at this wonderfully happy coincidence. Originally Fare well was only said to some one going away. If now the departing guest says Farewell to his friend who is staying at home, it can only be because the word Farewell has been conceived as a fixed formula, without any consciousness of the meaning of its parts.

Sometimes, on the other hand, new connexions of thought arise, as when we associate the word bound with bind in the phrase ‘he is bound for America.’ Our ancestors meant ‘he is ready to go’ (ON. búinn, ‘ready’), not ‘he is under an obligation to go.’ The establishment of new associations of this kind seems naturally to take place at the moment when the young mind makes acquaintance with the word: the phenomenon is, of course, closely related to “popular etymology” (see Ch. VI § 6).

X.—§ 4. Differentiations.

Linguistic ‘splittings’ or differentiations, whereby one word becomes two, may also be largely due to the transmission of the language to a new generation. The child may hear two pronunciations of the same word from different people, and then associate these with different ideas. Thus Paul Passy learnt the word meule in the sense of ‘grindstone’ from his father, and in the sense of ‘haycock’ from his mother; now the former in both senses pronounced [mœl], and the latter in both [mø·l], and the child thus came to distinguish [mœl] ‘grindstone’ and [mø·l] ‘haycock’ (Ch 23).

Or the child may have learnt the word at two different periods of its life, associated with different spheres. This, I take it, may be the reason why some speakers make a distinction between two pronunciations of the word medicine, in two and in three syllables: they take [medsin], but study [medisin].

Finally, the child can itself split words. A friend writes: “I remember that when a schoolboy said that it was a good thing that the new Headmaster was Dr. Wood, because he would then know when boys were ‘shamming,’ a schoolfellow remarked, ‘Wasn’t it funny? He did not know the difference between Doctor and Docter.’” In Danish the Japanese are indiscriminately called either Japanerne or Japaneserne; now, I once overheard my boy (6.10) lecturing his playfellows: “Japaneserne, that is the soldiers of Japan, but Japanerne, that is students and children and such-like.” It is, of course, possible that he may have heard one form originally when shown some pictures of Japanese soldiers, and the other on another occasion, and that this may have been the reason for his distinction. However this may be, I do not doubt that a number of differentiations of words are to be ascribed to the transmission of the language to a new generation. Others may have arisen in the speech of adults, such as the distinction between off and of (at first the stressed and unstressed form of the same preposition), or between thorough and through (the former is still used as a preposition in Shakespeare: “thorough bush, thorough brier”). But complete differentiation is not established till some individuals from the very first conceive the forms as two independent words.

X.—§ 5. Summary.

Instead of saying, as previous writers on these questions have done, either that children have no influence or that they have the chief influence on the development of language, it will be seen that I have divided the question into many, going through various fields of linguistic change and asking in each what may have been the influence of the child. The result of this investigation has been that there are certain fields in which it is both impossible and really also irrelevant to separate the share of the child and of the adult, because both will be apt to introduce changes of that kind; such are assimilations of neighbouring sounds and droppings of consonants in groups. Also, with regard to those very gradual shiftings either of sound or of meaning in which it is natural to assume many intermediate stages through which the sound or signification must have passed before arriving at the final result, children and adults must share the responsibility for the change. Clippings of words occur in the speech of both classes, but as a rule adults will keep the beginning of a word, while very small children will perceive or remember only the end of a word and use that for the whole. But finally there are some kinds of changes which must wholly or chiefly be charged to the account of children: such are those leaps in sound or signification in which intermediate stages are out of the question, as well as confusions of similar words and misdivisions of words, and the most violent differentiations of words.

I wish, however, here to insist on one point which has, I think, become more and more clear in the course of our disquisition, namely, that we ought not really to put the question like this: Are linguistic changes due to children or to grown-up people? The important distinction is not really one of age, which is evidently one of degree only, but that between the first learners of the sound or word in question and those who use it after having once learnt it. In the latter case we have mainly to do with infinitesimal glidings, the results of which, when summed up in the course of long periods of time, may be very considerable indeed, but in which it will always be possible to detect intermediate links connecting the extreme points. In contrast to these changes occurring after the correct (or original) form has been acquired by the individual, we have changes occurring simultaneously with the first acquisition of the word or form in question, and thus due to the fact of its transmission to a new generation, or, to speak more generally, and, indeed, more correctly, to new individuals. The exact age of the learner here is of little avail, as will be seen if we take some examples of metanalysis. It is highly probable that the first users of forms like a pea or a cherry, instead of a pease and a cherries, were little children; but a Chinee and a Portuguee are not necessarily, or not pre-eminently, children’s words: on the other hand, it is to me indubitable that these forms do not spring into existence in the mind of someone who has previously used the forms Chinese and Portuguese in the singular number, but must be due to the fact that the forms the Chinese and the Portuguese (used as plurals) have been at once apprehended as made up of Chinee, Portuguee + the plural ending -s by a person hearing them for the first time; similarly in all the other cases. We shall see in a later chapter that the adoption (on the part of children and adults alike) of sounds and words from a foreign tongue presents certain interesting points of resemblance with these instances of change: in both cases the innovation begins when some individual is first made acquainted with linguistic elements that are new to him.

X.—§ 6. Indirect Influence.

We have hitherto considered what elements of the language may be referred to a child’s first acquisition of language. But we have not yet done with the part which children play in linguistic development. There are two things which must be sharply distinguished from the phenomena discussed in the preceding chapter—the first, that grown-up people in many cases catch up the words and forms used by children and thereby give them a power of survival which they would not have otherwise; the second, that grown-up people alter their own language so as to meet children half-way.

As for the first point, we have already seen examples in which mothers and nurses have found the baby’s forms so pretty that they have adopted them themselves. Generally these forms are confined to the family circle, but they may under favourable circumstances be propagated further. A special case of the highest interest has been fully discussed in the section about words of the mamma-class.

As for the second point, grown-up people often adapt their speech to the more or less imaginary needs of their children by pronouncing words as they do, saying dood and tum for ‘good’ and ‘come,’ etc. This notion clearly depends on a misunderstanding, and can only retard the acquisition of the right pronunciation; the child understands good and come at least as well, if not better, and the consequence may be that when he is able himself to pronounce [g] and [k] he may consider it immaterial, because one can just as well say [d] and [t] as [g] and [k], or may be bewildered as to which words have the one sound and which the other. It can only be a benefit to the child if all who come in contact with it speak from the first as correctly, elegantly and clearly as possible—not, of course, in long, stilted sentences and with many learned book-words, but naturally and easily. When the child makes a mistake, the most effectual way of correcting it is certainly the indirect one of seeing that the child, soon after it has made the mistake, hears the correct form. If he says ‘A waps stinged me’: answer, ‘It stung you: did it hurt much when the wasp stung you?’ etc. No special emphasis even is needed; next time he will probably use the correct form.

But many parents are not so wise; they will say stinged themselves when once they have heard the child say so. And nurses and others have even developed a kind of artificial nursery language which they imagine makes matters easier for the little ones, but which is in many respects due to erroneous ideas of how children ought to talk rather than to real observation of the way children do talk. Many forms are handed over traditionally from one nurse to another, such as totties, tootems or tootsies for ‘feet’ (from trotters?), toothy-peg for ‘tooth,’ tummy or tumtum for ‘stomach,’ tootleums for ‘babies,’ shooshoo for ‘a fly.’ I give a connected specimen of this nursery language (from Egerton, Keynotes, 85): “Didsum was denn? Oo did! Was ums de prettiest itta sweetums denn? Oo was. An’ did um put ’em in a nasty shawl an’ joggle ’em in an ole puff-puff, um did, was a shame! Hitchy cum, hitchy cum, hitchy cum hi, Chinaman no likey me.” This reminds one of pidgin-English, and in a later chapter we shall see that that and similar bastard languages are partly due to the same mistaken notion that it is necessary to corrupt one’s language to be easily understood by children and inferior races.

Very frequently mothers and nurses talk to children in diminutives. When many of these have become established in ordinary speech, losing their force as diminutives and displacing the proper words, this is another result of nursery language. The phenomenon is widely seen in Romance languages, where auricula, Fr. oreille, It. orecchio, displaces auris, and avicellus, Fr. oiseau, It. uccello, displaces avis; we may remember that classical Latin had already oculus, for ‘eye.’[40] It is the same in Modern Greek. An example of the same tendency, though not of the same formal means of a diminutive ending, is seen in the English bird (originally = ‘young bird’) and rabbit (originally = ‘young rabbit’), which have displaced fowl and coney.

A very remarkable case of the influence of nursery language on normal speech is seen in many countries, viz. in the displacing of the old word for ‘right’ (as opposed to left). The distinction of right and left is not easy for small children: some children in the upper classes at school only know which is which by looking at some wart, or something of the sort, on one of their hands, and have to think every time. Meanwhile mothers and nurses will frequently insist on the use of the right (dextera) hand, and when they are not understood, will think they make it easier for the child by saying ‘No, the right hand,’ and so it comes about that in many languages the word that originally means ‘correct’ is used with the meaning ‘dexter.’ So we have in English right, in German recht, which displaces zeso, Fr. droit, which displaces destre; in Spanish also la derecha has begun to be used instead of la diestra; similarly, in Swedish den vackra handen instead of högra, and in Jutlandish dialects den kjön hånd instead of höjre.

X.—§ 7. New Languages.

In a subsequent chapter (XIV § 5) we shall consider the theory that epochs in which the changes of some language proceed at a more rapid pace than at others are due to the fact that in times of fierce, widely extended wars many men leave home and remain abroad, either as settlers or as corpses, while the women left behind have to do the field-work, etc., and neglect their homes, the consequence being that the children are left more to themselves, and therefore do not get their mistakes in speech corrected as much as usual.

A somewhat related idea is at the bottom of a theory advanced as early as 1886 by the American ethnologist Horatio Hale (see “The Origin of Languages,” in the American Association for the Advancement of Science, XXXV, 1886, and “The Development of Language,” the Canadian Institute, Toronto, 1888). As these papers seem to have been entirely unnoticed by leading philologists, I shall give a short abstract of them, leaving out what appears to me to be erroneous in the light of recent linguistic thought and research, namely, his application of the theory to explain the supposed three stages of linguistic development, the monosyllabic, the agglutinative and the flexional.

Hale was struck with the fact that in Oregon, in a region not much larger than France, we find at least thirty different families of languages living together. It is impossible to believe that thirty separate communities of speechless precursors of man should have begun to talk independently of one another in thirty distinct languages in this district. Hale therefore concludes that the origin of linguistic stocks is to be found in the language-making instinct of very young children. When two children who are just beginning to speak are thrown much together, they sometimes invent a complete language, sufficient for all purposes of mutual intercourse, and yet totally unintelligible to their parents. In an ordinary household, the conditions under which such a language would be formed are most likely to occur in the case of twins, and Hale now proceeds to mention those instances—five in all—that he has come across of languages framed in this manner by young children. He concludes: “It becomes evident that, to ensure the creation of a speech which shall be a parent of a new language stock, all that is needed is that two or more young children should be placed by themselves in a condition where they will be entirely, or in a large degree, free from the presence and influence of their elders. They must, of course, continue in this condition long enough to grow up, to form a household, and to have descendants to whom they can communicate their new speech.”

These conditions he finds among the hunting tribes of America, in which it is common for single families to wander off from the main band. “In modern times, when the whole country is occupied, their flight would merely carry them into the territory of another tribe, among whom, if well received, they would quickly be absorbed. But in the primitive period, when a vast uninhabited region stretched before them, it would be easy for them to find some sheltered nook or fruitful valley.... If under such circumstances disease or the casualties of a hunter’s life should carry off the parents, the survival of the children would, it is evident, depend mainly upon the nature of the climate and the ease with which food could be procured at all seasons of the year. In ancient Europe, after the present climatal conditions were established, it is doubtful if a family of children under ten years of age could have lived through a single winter. We are not, therefore, surprised to find that no more than four or five language stocks are represented in Europe.... Of Northern America, east of the Rocky Mountains and north of the tropics, the same may be said.... But there is one region where Nature seems to offer herself as the willing nurse and bountiful stepmother of the feeble and unprotected ... California. Its wonderful climate (follows a long description).... Need we wonder that, in such a mild and fruitful region, a great number of separate tribes were found, speaking languages which a careful investigation has classed in nineteen distinct linguistic stocks?” In Oregon, and in the interior of Brazil, Hale finds similar climatic conditions with the same result, a great number of totally dissimilar languages, while in Australia, whose climate is as mild as that of any of these regions, we find hundreds, perhaps thousands, of petty tribes, as completely isolated as those of South America, but all speaking languages of the same stock—because “the other conditions are such as would make it impossible for an isolated group of young children to survive. The whole of Australia is subject to severe droughts, and is so scantily provided with edible products that the aborigines are often reduced to the greatest straits.”

This, then, is Hale’s theory. Let us now look a little closer into the proofs adduced. They are, as it will be seen, of a twofold order. He invokes the language-creating tendencies of young children on the one hand, and on the other the geographical distribution of linguistic stocks or genera.

As to the first, it is true that so competent a psychologist as Wundt denies the possibility in very strong terms.[41] But facts certainly do not justify this foregone conclusion. I must first refer the reader to Hale’s own report of the five instances known to him. Unfortunately, the linguistic material collected by him is so scanty that we can form only a very imperfect idea of the languages which he says children have developed and of the relation between them and the language of the parents. But otherwise his report is very instructive, and I shall call special attention to the fact that in most cases the children seem to have been ‘spoilt’ by their parents; this is also the case with regard to one of the families, though it does not appear from Hale’s own extracts from the book in which he found his facts (G. Watson, Universe of Language, N.Y., 1878).

The only word recorded in this case is nī-si-boo-a for ‘carriage’; how that came into existence, I dare not conjecture; but when it is said that the syllables of it were sometimes so repeated that they made a much longer word, this agrees very well with what I have myself observed with regard to ordinary children’s playful word-coinages. In the next case, described by E. R. Hun, M.D., of Albany, more words are given. Some of these bear a strong resemblance to French, although neither the parents nor servants spoke that language; and Hale thinks that some person may have “amused herself, innocently enough, by teaching the child a few words of that tongue.” This, however, does not seem necessary to explain the words recorded. Feu, pronounced, we are told, like the French word, signified ‘fire, light, cigar, sun’: it may be either E. fire or else an imitation of the sound fff without a vowel, or [fə·] used in blowing out a candle or a match or in smoking, so as to amuse the child, exactly as in the case of one of my little Danish friends, who used fff as the name for ‘smoke, steam,’ and later for ‘funnel, chimney,’ and finally anything standing upright against the sky, for instance, a flagstaff. Petee-petee, the name which the Albany girl gave to her brother, and which Dr. Hun derived from F. petit, may be just as well from E. pet or petty; and to explain her word for ‘I,’ ma, we need not go to F. moi, as E. me or my may obviously be thus distorted by any child. Her word for ‘not’ is said to have been ne-pas, though the exact pronunciation is not given. This cannot have been taken from the French, at any rate not from real French, as ne and pas are here separated, and ne is more often than not pronounced without the vowel or omitted altogether; the girl’s word, if pronounced something like ['nepa·] may be nothing else than an imperfect childish pronunciation of never, cf. the negroes’ form nebber. Too, ‘all, everything,’ of course resembles Fr. tout, but how should anyone have been able to teach this girl, who did not speak any intelligible language, a French word of this abstract character? Some of the other words admit of a natural explanation from English: go-go, ‘delicacy, as sugar, candy or dessert,’ is probably goody-goody, or a reduplicated form of good; deer, ‘money,’ may be from dear, ‘expensive’; odo, ‘to send for, to go out, to take away,’ is evidently out, as in ma odo, ‘I want to go out’; gaän, ‘God,’ must be the English word, in spite of the difference in pronunciation, for the child would never think of inventing this idea on its own accord; pa-ma, ‘to go to sleep, pillow, bed,’ is from by-bye or an independent word of the mamma-class; mea, ‘cat, fur,’ of course is imitative of the sound of the cat. For the rest of the words I have no conjectures to offer. Some of the derived meanings are curious, though perhaps not more startling than many found in the speech of ordinary children; papa and mamma separately had their usual signification, but papa-mamma meant ‘church, prayer-book, cross, priest’: the parents were punctual in church observances; gar odo, ‘horse out, to send for the horse,’ came to mean ‘pencil and paper,’ as the father used, when the carriage was wanted, to write an order and send it to the stable. In the remaining three cases of ‘invented’ languages no specimens are given, except shindikik, ‘cat.’ In all cases the children seem to have talked together fluently when by themselves in their own gibberish.

But there exists on record a case better elucidated than Hale’s five cases, namely that of the Icelandic girl Sæunn. (See Jonasson and Eschricht in Dansk Maanedsskrift, Copenhagen, 1858.) She was born in the beginning of the last century on a farm in Húnavatns-syssel in the northern part of Iceland, and began early to converse with her twin brother in a language that was entirely unintelligible to their surroundings. Her parents were disquieted, and therefore resolved to send away the brother, who died soon afterwards. They now tried to teach the girl Icelandic, but soon (too soon, evidently!) came to the conclusion that she could not learn it, and then they were foolish enough to learn her language, as did also her brothers and sisters and even some of their friends. In order that she might be confirmed, her elder brother translated the catechism and acted as interpreter between the parson and the girl. She is described as intelligent—she even composed poetry in her own language—but shy and distrustful. Jonasson gives a few specimens of her language, some of which Eschricht succeeds in interpreting as based on Icelandic words, though strangely disfigured. The language to Jonasson, who had heard it, seemed totally dissimilar to Icelandic in sounds and construction; it had no flexions, and lacked pronouns. The vocabulary was so limited that she very often had to supplement a phrase by means of nods or gestures; and it was difficult to carry on a conversation with her in the dark. The ingenuity of some of the compounds and metaphors is greatly admired by Jonasson, though to the more sober mind of Eschricht they appear rather childish or primitive, as when a ‘wether’ is called mepok-ill from me (imitation of the sound) + pok, ‘a little bag’ (Icel. poki) + ill, ‘to cut.’ The only complete sentence recorded is ‘Dirfa offo nonona uhuh,’ which means: ‘Sigurdur gets up extremely late.’ In his analysis of the whole case Eschricht succeeds in stripping it of the mystical glamour in which it evidently appeared to Jonasson as well as to the girl’s relatives; he is undoubtedly right in maintaining that if the parents had persisted in only talking Icelandic to her, she would soon have forgotten her own language; he compares her words with some strange disfigurements of Danish which he had observed among children in his own family and acquaintanceship.

I read this report a good many years ago, and afterwards I tried on two occasions to obtain precise information about similar cases I had seen mentioned, one in Halland (Sweden) and the other in Finland, but without success. But in 1903, when I was lecturing on the language of children in the University of Copenhagen, I had the good fortune to hear of a case not far from Copenhagen of two children speaking a language of their own. I investigated the case as well as I could, by seeing and hearing them several times and thus checking the words and sentences which their teacher, who was constantly with them, kindly took down in accordance with my directions. I am thus enabled to give a fairly full account of their language, though unfortunately my investigation was interrupted by a long voyage in 1904.

The boys were twins, about five and a half years old when I saw them, and so alike that even the people who were about them every day had difficulty in distinguishing them from each other. Their mother (a single woman) neglected them shamefully when they were quite small, and they were left very much to shift for themselves. For a long time, while their mother was ill in a hospital, they lived in an out-of-the-way place with an old woman, who is said to have been very deaf, and who at any rate troubled herself very little about them. When they were four years old, the parish authorities discovered how sadly neglected they were and that they spoke quite unintelligibly, and therefore sent them to a ‘children’s home’ in Seeland, where they were properly taken care of. At first they were extremely shy and reticent, and it was a long time before they felt at home with the other children. When I first saw them, they had in so far learnt the ordinary language that they were able to understand many everyday sentences spoken to them, and could do what they were told (e.g. ‘Take the footstool and put it in my room near the stove’), but they could not speak Danish and said very little in the presence of anybody else. When they were by themselves they conversed pretty freely and in a completely unintelligible gibberish, as I had the opportunity to convince myself when standing behind a door one day when they thought they were not observed. Afterwards I got to be in a way good friends with them—they called me py-ma, py being their word for ‘smoke, smoking, pipe, cigar,’ so that I got my name from the chocolate cigars which I used to ingratiate myself with them—and then I got them to repeat words and phrases which their teacher had written out for me, and thus was enabled to write down everything phonetically.

An analysis of the sounds occurring in their words showed me that their vocal organs were perfectly normal. Most of the words were evidently Danish words, however much distorted and shortened; a voiceless l, which does not occur in Danish, and which I write here lh, was a very frequent sound. This, combined with an inclination to make many words end in -p, was enough to disguise words very effectually, as when sort (black) was made lhop. I shall give the children’s pronunciations of the names of some of their new playfellows, adding in brackets the Danish substratum: lhep (Svend), lhip (Vilhelm), lip (Elisabeth), lop (Charlotte), bap (Mandse); similarly the doctor was called dop. In many cases there was phonetic assimilation at a distance, as when milk (mælk) was called bep, flower (blomst) bop, light (lys) lhylh, sugar (sukker) lholh, cold (kulde) lhulh, sometimes also ulh, bed (seng) sæjs, fish (fisk) se-is.

I subjoin a few complete sentences: nina enaj una enaj hæna mad enaj, ‘we shall not fetch food for the young rabbits’: nina rabbit (kanin), enaj negation (nej, no), repeated several times in each negative sentence, as in Old English and in Bantu languages, una young (unge). Bap ep dop, ‘Mandse has broken the hobby-horse,’ literally ‘Mandse horse piece.’ Hos ia bov lhalh, ‘brother’s trousers are wet, Maria,’ literally ‘trousers Maria brother water.’ The words are put together without any flexions, and the word order is totally different from that of Danish.

Only in one case was I unable to identify words that I understood either as ‘little language’ forms of Danish words or else as sound-imitations; but then it must be remembered that they spoke a good deal that neither I nor any of the people about them could make anything of. And then, unfortunately, when I began to study it, their language was already to a great extent ‘humanized’ in comparison to what it was when they first came to the children’s home. In fact, I noticed a constant progress during the short time I observed the boys, and in some of the last sentences I have noted, I even find the genitive case employed.

The idiom of these twins cannot, of course, be called an independent, still less a complete or fully developed language; but if they were able to produce something so different from the language spoken around them at the beginning of the twentieth century and in a civilized country, there can to my mind be no doubt that Hale is right in his contention that children left to themselves even more than these were, in an uninhabited region where they were still not liable to die from hunger or cold, would be able to develop a language for their mutual understanding that might become so different from that of their parents as really to constitute a new stock of language. So that we can now pass to the other—geographical—side of what Hale advances in favour of his theory.

So far as I can see, the facts here tally very well with the theory. Take, on the one hand, the Eskimo languages, spoken with astonishingly little variation from the east coast of Greenland to Alaska, an immense stretch of territory in which small children if left to themselves would be sure to die very soon indeed. Or take the Finnish-Ugrian languages in the other hemisphere, exhibiting a similar close relationship, though spread over wide areas. And then, on the other hand, the American languages already adduced by Hale. I do not pretend to any deeper knowledge of these languages; but from the most recent works of very able specialists I gather an impression of the utmost variety in phonetics, in grammatical structure and in vocabulary; see especially Roland B. Dixon and Alfred L. Kroeber, “The Native Languages of California,” in the American Anthropologist, 1903. Even where recent research seems to establish some kind of kinship between families hitherto considered as distinguished stocks (as in Dixon’s interesting paper, “Linguistic Relationships within the Shasta-Achomawi Stock,” XV Congrès des Américanistes, 1906) the similarities are still so incomplete, so capricious and generally so remote that they seem to support Hale’s explanation rather than a gradual splitting of the usual kind.

As for Brazil, I shall quote some interesting remarks from C. F. P. v. Martius, Beiträge zur Ethnographie u. Sprachenkunde Amerika’s, 1867, i. p. 46: “In Brazil we see a scant and unevenly distributed native population, uniform in bodily structure, temperament, customs and manner of living generally, but presenting a really astonishing diversity in language. A language is often confined to a few mutually related individuals; it is in truth a family heirloom and isolates its speakers from all other people so as to render any attempt at understanding impossible. On the vessel in which we travelled up the rivers in the interior of Brazil, we often, among twenty Indian rowers, could count only three or four that were at all able to speak together ... they sat there side by side dumb and stupid.”

Hale’s theory is worthy, then, of consideration, and now, at the close of our voyage round the world of children’s language, we have gained a post of vantage from which we can overlook the whole globe and see that the peculiar word-forms which children use in their ‘little language’ period can actually throw light on the distribution of languages and groups of languages over the great continents. Yes,