Anthropology

47. Linguistic Relationship: The Speech Family

The question that the historian and anthropologist are likely to ask most frequently of the philologist, is whether this and that language are or are not related. Relationship in such connection means descent from a common source, as two brothers are descended from the same father, or two cousins from a common grandfather. If languages can be demonstrated to possess such common source, it is clear that the peoples who spoke them must at one time have been in close contact, or perhaps have constituted a single people. If, on the other hand, the languages of two peoples prove wholly dissimilar, though their racial types and cultures be virtually identical, as indeed is sometimes found to be the case—witness the Hungarians and their neighbors—it is evident that an element of discontinuous development must somewhere be reckoned with. Perhaps one part of an originally single racial group gradually modified its speech beyond recognition, or under the shock of conquest, migration, or other historical accident entirely discarded it in favor of a new and foreign tongue. Or the opposite may be true: the two groups were originally distinct in all respects, but, being brought in contact, their cultures interpenetrated, intermarriage followed, and the two physical types became assimilated into one while the languages remained dissimilar. In short, if one wishes full understanding of a people, one must take its language into consideration. This means that it must be classified. If a historical classification is to be more than barrenly logical, it must have reference to relationship, development, origin. In a word, it must be a genetic classification.

The term used to indicate that two or more languages have a common source but are unrelated to all others, or seem so in the present state of knowledge, is “linguistic family.” “Linguistic stock” is frequently used as a synonym. This is the fundamental concept in the classification of languages. Without a clear idea of its meaning one involves himself in confusion on attempting to use philology as an aid to other branches of human history.

There is no abstract reason against referring to a group of unrelated languages as a “family” because they are all spoken in one area, nor against denominating as “families,” as has sometimes been done, the major subdivisions of a group of languages admittedly of common origin. Again, languages that show certain similarities of type or structure, such as inflection, might conceivably be put into one “family.” But there is this objection to all such usages: they do not commit themselves on the point of genetic relationship, or they contradict it, or only partially exhaust it. Yet commonness of origin is so important in many connections that it is indispensable to have one term which denotes its ascertainable presence. And for this quality there happens to be no generally understood designation other than “linguistic family,” or its synonym, “linguistic stock.” This phrase will therefore be used here strictly in the sense of the whole of a group of languages sprung from a single source, and only in that sense. Other groupings will be indicated by phrases like “languages of such and such an area,” “subfamily,” “division of a family,” or “languages of similar type.”

48. Criteria of Relationship

The question that first arises in regard to linguistic families is how the relationship of their constituent idioms is determined. In brief, the method is one of comparison. If a considerable proportion of the words and grammatical forms of two languages are reasonably similar, similar enough to indicate that the resemblances cannot be due to mere accident, these similar words and forms must go back to a common source, and if this source is not borrowing, the two tongues are related. If comparison fails to bring out any such degree of resemblance, the languages are classed in distinct families.

Of course it is possible that the reason two languages seem unrelated is not that they are really so, but that they have in the lapse of ages become so much differentiated that one cannot any longer find resemblance between their forms. In that event true relationship would be obscured by its remoteness. Theoretically there is high probability that many families of languages, customarily regarded as totally distinct, do go back in the far past to a common origin, and that ignorance of their history, or inability to analyze them deeply, prevents recognition of their relationship. From time to time it happens that groups of languages which at first seemed unrelated are shown by more intensive study to possess elements enough in common to compel the recognition of their original unity. In that case what were supposed to be several “families” become merged in one. The scope of a particular family may be thus enlarged; but the scope of the generic concept of “family” is not altered.

Whether there is any hope that comparative philology may ultimately be prosecuted with sufficient success to lead all the varied forms of human speech back to a single origin, is an interesting speculation. A fair statement is that such a possibility cannot be denied, but that the science is still far from such a realization, and that progress toward it is necessarily slow. Of more immediate concern is an ordering and summarizing of the knowledge in hand with a view to such positive inferences as can be drawn.

In an estimate of the similarity of languages, words that count as evidence must meet two requirements: they must be alike or traceably similar in sound; and they must be alike or similar in meaning. This double requirement holds, whether full words or separable parts of words, roots or grammatical forms, are compared. The English word eel and the French île, meaning island, are pronounced almost exactly alike, yet their meaning is so different that no sane person would regard them as sprung from the same origin. As a matter of fact île is derived from Latin insula, whereas eel has a cognate in German aal. These prototypes insula and aal being as different in sound as they are in meaning, any possibility that eel and île might be related is easily disposed of. Yet if the Latin and German equivalents were lost, if nothing were known of the history of the English and French languages, and if île meant not island but, say, fish or watersnake, then it might be reasonable to think of a connection.

Such doubtful cases, of which a certain proportion are likely to be adjudged wrongly, are bound to come up in regard to the less investigated languages, particularly those of nations without writing, the earlier stages of whose speech have perished without trace. In proportion as more is known of a language, or as careful analysis can reconstruct more of its past stages, the number of such borderline cases obviously becomes fewer.

Before genetic connection between two languages can be thought of, the number of their words similar in sound and sense must be reasonably large. An isolated handful of resemblances obviously are either importations—loan words—or the result of coincidence. Thus in the native Californian language known as Yuki, ko means go, and kom means come. Yet examination of Yuki reveals no further instances of the same kind. It would therefore be absurd to dream of a connection: one swallow does not make a summer. This lone pair of resemblances means nothing except that the mathematical law of probability has operated. Among the thousands of words in one language, a number are likely to be similar in sound to words of another language; and of this number again a small fraction, perhaps one or two or five in all, will happen to bear some resemblance in meaning also. In short, the similarities upon which a verdict of genetic relationship is based must be sufficiently numerous to fall well beyond possibility of mere coincidence; and it must also be possible to prove with reasonable certainty that they are not the result of one language borrowing words from another, as, for instance, English borrowed from French and Latin.

At the same time it is not necessary that the similarities extend to the point of identity. In fact, too close a resemblance between part of the stock of two languages immediately raises a presumption of borrowing. For every language is continually changing, and once a mother tongue has split into several branches, each of these goes on modifying its sounds, and gradually shifting the meaning of its words, generation after generation. In short, where connection is real, it must be veiled by a certain degree of distortion.

Take the English word foot and the Latin word of the same meaning, pes. To offhand inspection the sounds or forms of the two words do not seem similar. The resemblance becomes more definite in other forms of pes, for instance the genitive case ped-is or the accusative ped-em. Obviously the stem or elementary portion of the Latin word is not pes but ped-; and the d is closer to the English t of foot than is the s of pes. The probability of relationship is increased by the Greek word for foot, pous, whose stem proves to be pod-, with vowel closer to that of English. Meanwhile, it would be recognized that there are English words beginning with ped-, such as pedal, pedestrian, pedestal, all of which have a clear association with the idea of foot. All these words however possess almost exact equivalents in Latin. One would therefore be justified in concluding from these facts what indeed the history of the languages proves, namely, that pedal, pedestrian, and pedestal are Latin words taken over into English; whereas foot and pes and pous, and for that matter German fuss, are derivatives from a common form which once existed in the now extinct mother tongue from which Greek and Latin and English and German are derived.

49. Sound Equivalences and Phonetic Laws

The question next arises whether it is possible to account for the distortions which have modified the original word into foot, ped-, etc. What has caused the initial sound of this ancient word to become p in Latin and f in English, and its last consonant to be d in Latin and Greek, t in English, and ss in German? To answer this seemingly innocent question with accuracy for this one word alone would involve a treatise on the whole group of languages in question, and even then the causes, as causes, could scarcely be set down with certainty. But it has proved possible to assemble a large number of instances of parallel distortion in which Latin p corresponds to English f, or d to t. Evidently philology has got hold of a generalized phenomenon here. Since father corresponds to pater, full to pl-enus, for to pro, fish to piscis, and so on in case after case, we are evidently face to face with a happening that has occurred with regularity and to which the name “law” is therefore applicable.

The f of foot and p of pes are both lip sounds. They differ preëminently in that f can be prolonged indefinitely, whereas p is a momentary sound. It is produced by closure of the lips for a fraction of a second during which there is an interruption of sound production, followed by a somewhat explosive release of the breath which has been impounded in the mouth cavity. This explosion is of necessity instantaneous. Since it is preceded by occlusion, or stoppage of the breath, it is customary to speak of sounds produced by a process like p as “stops.” F, on the other hand, is a “continuant,” or more specifically a “fricative.”

The English word three begins with a sound which, although conventionally represented by the two letters th, is a simple sound and in a class with f in being fricative. Th is formed by putting the tongue lightly across the teeth, just as f is made by placing the lower lip against the edge of the upper teeth. In both cases the breath is expelled with friction through a narrow passage. Now if the fricative f is represented in Latin by the stop p, then, if regularity holds good, the English fricative th ought to be represented in Latin by the stop sound in the corresponding dental position, namely t. The Latin word for three is in fact tres; for thin, ten-uis; for mother, mater; for thou, tu, and so on. The regularity therefore extends beyond the limits of the single labial class of sounds, and applies with equal force to the dentals; and, it may be added, to the palatals or gutturals as well.

As one passes from English and Latin to German, one finds the initial sound of the word meaning three, drei, to be somewhat different from th and t but still clearly allied, since it also is made by the tongue against the teeth. D is a stop like t, but the vocal cords vibrate while it is being pronounced, whereas in t the vocal cords are silent. D is “voiced” or “sonant,” t “unvoiced” or “surd.” Hence the formulation: Latin, surd stop; German, sonant stop; English, fricative. This triple equivalence can be substantiated in other words. For instance, ten-uis, dünn, thin; tu, du, thou.

If it is the English word that contains a surd stop, what will be the equivalent in Latin and German? Compare ten, Latin decem, German zehn. Again the three classes of sounds run parallel; but the place of their appearance in the three languages has shifted.

The third possible placing of the three sounds in the three languages is when English has the sonant stop, d. By exclusion it might be predicted that Latin should then show the fricative th and German the surd stop t. The word daughter confirms. The German is tochter. Latin in this case fails us, the original corresponding stem having gone out of use and been replaced by the word filia. But Greek, whose sounds align with those of Latin as opposed to English and German, provides the th as expected: thygater. Compare death, tod, thanatos.

Let us bring together these results so that the eye may grasp them:

These relations apply not only to the dentals d, t, th (z), which have been chosen for illustration, but also to the labials, p, b, f, and to the palatals k, g, h (gh, ch).

It is evident that most of the sounds occur in all three groups of languages, but not in the same words. The sound t is common to English, Latin, and German, but when it appears in a particular word in one of these languages it is replaced by d and th in the two others. This replacing is known as a “sound shift.” The sound shifts just enumerated constitute the famous Grimm’s Law. This was the first discovered important phonetic law or system of sound substitutions. Yet it is only one of a number of shifts that have been worked out for the Indo-European group of languages to which English, German, and Latin belong. So far, only stopped and fricative consonants have been reviewed here, and no vowels have been considered. Other groups of languages also show shifts, but often different ones, as between l and n, or s and k, or p and k.

The significance of a shift lies in the fact that its regularity cannot be explained on any other ground than that the words in which the law is operative must originally have been the same. That is, Latin duo, German zwei, English two are all only variants of a word which meant “two” in the mother tongue from which these three languages are descended. This example alone is of course insufficient evidence for the existence of such a common mother tongue. But that each of the shifts discussed is substantiated by hundreds or thousands of words in which it holds true, puts the shift beyond the possibility of mere accident. The explanation of coincidence is ruled out. The resemblances therefore are both genuine and genetic. The conclusion becomes inevitable that the languages thus linked are later modifications of a former single speech.

It is in this way that linguistic relationship is determined. Where an ancient sound shift, a law of phonetic change, can be established by a sufficient number of cases, argument ceases. It is true that when most of a language has perished, or when an unwritten language has been but fragmentarily recorded or its analysis not carried far, a strong presumption of genetic unity may crowd in on the investigator who is not yet in a position to present the evidence of laws. The indications may be strong enough to warrant a tentative assumption of relationship. But the final test is always the establishment of laws of sound equivalence that hold good with predominating regularity.

50. The Principal Speech Families

The number of linguistic families is not a matter of much theoretical import. From what has already been said it appears that the number can perhaps never be determined with absolute accuracy. As knowledge accumulates and dissection is carried to greater refinements, new phonetic laws will uncover and serve to unite what now seem to be separate stocks. Yet for the practical purpose of classification and tracing relationship the linguistic family will remain a valuable tool. A rapid survey of the principal families is therefore worth while.

In Asia and Europe, which must be considered a unit in this connection, the number of stocks, according to conservative reckoning, does not exceed twenty-five. The most important of these, in point of number of speakers, is the Indo-European or Indo-Germanic or Aryan family, whose territory for several thousand years has comprised southwestern Asia and the greater part, but by no means all, of Europe. The most populous branches of the Indo-European family are the Indic, Slavic, Germanic, and Romance or Latin. Others are Persian or Iranic, Armenian, Greek, Albanian, Baltic or Lithuanian, and Keltic. From Europe various Indo-European languages, such as English, Spanish, French, Russian, have in recent centuries been carried to other continents, until in some, such as the Americas and Australia, the greater area is now inhabited by peoples speaking Indo-European. As the accompanying maps are intended to depict the historic or native distribution of languages they do not show this diffusion. It will be noted that the distribution of Indo-European has the form of a long belt stretching from western Europe to northeastern India, with an interruption only in Asia Minor (Fig. 12). Turkish peoples displaced Indo-Europeans there about a thousand years ago, thus breaking the territorial continuity. It is probable that another link between the western and eastern Indo-Europeans once stretched around the north of the Caspian sea. Here also there are Turks now.

Almost equaling Indo-European in the number of its speakers is Sinitic, which is generally held to include Chinese proper with its dialects; the Tibeto-Burman branch; the T’ai or Shan-Siamese branch; and probably some minor divisions like Lolo.

In extent of territory occupied the Altaic stock rivals the Indo-European. Its three main divisions, Turkish, Mongolian, and Tungus-Manchu, cover most of northern and central Asia and some tracts in Europe. The Turks, as just noted, are the only stock that within the period of history has gained appreciable territory at Indo-European expense. The Uralic or Finno-Ugric family has eastern Europe and northwestern Asia as its home, with the Finns and Hungarian Magyars as its most civilized and best known representatives. This is a scattered stock. Most scholars unite the three Altaic divisions, Finno-Ugric, and Samoyed into a vast Ural-Altaic family.

Of the Semitic family, Arabic is the chief living representative, with Abyssinian in Africa as a little known half-sister. Arabic is one of the most widely diffused of all languages, and as the orthodox vehicle of Mohammedanism has served an important function as a culture carrier. Several great nations of ancient times also spoke Semitic tongues: the Babylonians, Assyrians, Phœnicians, Carthaginians, and Hebrews.

Southern India is Dravidian. While people of this family enter little into our customary thoughts, they number over fifty millions. Japanese and Korean also merit mention as important stock tongues. Anamese, by some regarded as an offshoot from Chinese, may constitute a separate stock. Several minor families will be found on the Asiatic map, most of them consisting of uncivilized peoples or limited in their territory or the number of their speakers. Yet, so far as can be judged from present knowledge, they form units of the same order of independence as the great Indo-European, Semitic, and Ural-Altaic stocks.

Language distributions in Africa are in the main simple (Fig. 13). The whole of northern Africa beyond latitude 10°, and parts of east Africa almost to the equator, were at one time Hamitic. This is the family to which the language of ancient Egypt belonged. Hamitic and Semitic, named after sons of Noah, probably derive from a common source, although the separation of the common mother tongue into the African Hamitic and the Asiatic Semitic divisions must have occurred very anciently. In the past thousand years Hamitic has yielded ground before Semitic, due to the spread of Arabic in Mohammedan Africa.

Africa south of the equator is the home of the great Bantu family, except in the extreme southwest of the continent. There a tract of considerable area, though of small populational density, was in the possession of the backward Bushmen and Hottentots, distinctive in their physical type as well as languages.

Between the equator and latitude ten north, in the belt known as the Sudan, there is much greater speech diversity than elsewhere in Africa. The languages of the Sudan fall into several families, perhaps into a fairly large number. Opinion conflicts or is unsettled as to their classification. They are, at least in the main, non-Hamitic and non-Bantu; but this negative fact does not preclude their having had either a single or a dozen origins. It has usually been easier to throw them all into a vague group designated as non-Hamitic and non-Bantu than to compare them in detail.

In Oceania conditions are similar to those of Africa, in that there are a few great, widely branching stocks and one rather small area, New Guinea, of astounding speech diversity. Indeed, superficially this variety is the outstanding linguistic feature of New Guinea. The hundreds of Papuan dialects of the island look as if they might require twenty or more families to accommodate them. However, it is inconceivable that so small a population should time and again have evolved totally new forms of speech. It is much more likely that something in the mode of life or habits of mind of the Papuans has favored the breaking up of their speech into local dialects and an unusually rapid modification of these into markedly differentiated languages. What the circumstances were that favored this tendency to segregation and change can be only conjectured. At any rate, New Guinea ranks with the Sudan, western North America, and the Amazonian region of South America, as one of the areas of greatest linguistic multiplicity.

All the remainder of Oceania is either Australian or Malayo-Polynesian in speech. The Australian idioms have been imperfectly recorded. They were numerous and locally much varied, but seem to derive from a single mother tongue.

All the East Indies, including part of the Malay Peninsula, and all of the island world of the Pacific—Polynesia, Micronesia, and Melanesia—always excepting interior New Guinea—are the habitat of the closely-knit Malayo-Polynesian family, whose unity was quickly recognized by philologists. From Madagascar to Easter Island this speech stretches more than half-way around our planet. Some authorities believe that the Mon-Khmer languages of southern Indo-China and the Kolarian or Munda-Kol tongues of India are related in origin to Malayo-Polynesian, and denominate the larger whole, the Austronesian family.

North and South America, according to the usual reckoning, contain more native language families than all the remainder of the world. The orthodox classification allots about seventy-five families to North America (some fifty of them represented within the borders of the United States) and another seventy-five to South America. They varied greatly in size at the time of discovery, some being confined to a few hundred souls, whereas others stretched through tribe after tribe over enormous areas. Their distribution is so irregular and their areas so disproportionate as to be impossible of vivid representation except on a large-scale map in colors. The most important in extent of territory, number of speakers, or the cultural importance of the nations adhering to them, are, in North America, Eskimo, Athabascan, Algonkin, Iroquoian, Muskogean, Siouan, Uto-Aztecan, Maya; and in South America, Chibcha, Quechua, Aymara, Araucanian, Arawak, Carib, Tupi, Tapuya. It will be seen on the maps (Figs. 14, 15) that these sixteen groups held the greater part of the area of the double continent, the remaining smaller areas being crowded with about ten times as many stocks. Obviously, as in New Guinea, there cannot well have been such an original multiplicity; in fact, recent studies are tending to consolidate the hundred and fifty New World families into considerably fewer groups. But the evidence for such reductions is necessarily difficult to bring and much of it is still incomplete. The stocks mentioned above have been long determined and generally accepted.

About a third of humanity to-day speaks some form of Indo-European. A quarter talks some dialect of Sinitic stock. Semitic, Dravidian, Ural-Altaic, Japanese, Malayo-Polynesian, Bantu have each from about fifty to a hundred million speakers. The languages included in these eight families form the speech of approximately ninety per cent of living human beings.

51. Classification of Languages by Types

A classification is widely prevalent which puts languages according to their structure into three types: inflective, agglutinating, and isolating. To this some add a fourth type, the polysynthetic or incorporating. While the classification is largely misrepresentative, it enters so abundantly into current thought about human speech that it is worth presenting, analyzing, and, so far as it is invalid, refuting.

An inflecting language expresses relations or grammatical form by adding prefixes or suffixes which cannot stand alone, or if they stood alone would mean nothing; or that operates by internal modifications of the stem, which also can have no independent existence. The -ing of killing is such an inflection; so are the vowel changes and the ending -en in the conjugation write, wrote, written.

An isolating language expresses such relations or forms by separate words or isolated particles. English heart of man is isolating, where the Latin equivalent cor hominis is inflective, the per se meaningless suffix -is rendering the genitive or possessive force of the English word of.

An agglutinative language glues together into solid words elements for which a definite meaning of their own can be traced. English does not use this mechanism for purposes that are ordinarily reckoned as strictly grammatical, but does employ it for closely related purposes. Under-take, rest-less, are examples; and in a form like light-ly, which goes back to light-like, the force of the suffix which converts the adjective into the adverb is of a kind that in descriptions of most languages would be considered grammatical or formal.

Polysynthetic languages are agglutinative ones carried to a high pitch, or those that can compound words into equivalents of fair sized sentences. Steam-boat-propeller-blade might be called a polysynthetic form if we spoke or wrote it in one word as modern German and ancient Greek would.

Incorporating languages embody the object noun, or the pronoun representing it, into the word that contains the verb stem. This construction is totally foreign to English.[7]

Each of these classes evidently defines one or more distinctive linguistic processes. There are different mechanisms at work in kill-ing, of man, light-ly. The distinction is therefore both valid and valuable. Its abuse lies in trying to slap the label of one type on a whole language. The instances given show that English employs most of the several distinct processes. Obviously it would be arbitrary to classify English as outright of one type. This is also the situation for most other languages. There are a few languages that tend prevailingly in one direction or the other: Sanskrit and Latin and Hebrew toward the inflective structure, Turkish toward the agglutinative, Chinese toward the isolating. But they form a small minority, and most of them contain certain processes of types other than their predominating ones. Sanskrit, for instance, has polysynthetic traits, Hebrew incorporating ones. Therefore, so long as these concepts are used to picture a language in detail, with balanced recognition of the different processes employed by it, they are valuable tools to philological description. When on the other hand the concepts are degraded into catchwords designating three or four compartments into one of which every language is somehow to be stuffed, they grossly misrepresent most of the facts. The concepts, in short, apply usefully to types of linguistic processes, inadequately to types of languages.

Why then has the classification of human languages into inflecting, agglutinating, isolating, and polysynthetic or incorporating ones been repeated so often? First of all, because languages vary almost infinitely, and a true or natural classification, other than the genetic one into families, is intricate. The mind craves simplicity and the three or four supposedly all-embracing types are a temptation.

A second reason lies deeper. As philology grew up into a systematic body of knowledge, it centered its first interests on Latin and Greek, then on Sanskrit and the other older Indo-European languages. These happened to have inflective processes unusually well developed. They also happened to be the languages from which the native speech of the philologists was derived. What is our own seems good to us; consequently Indo-European was elevated into the highest or inflective class of languages. As a sort of after-thought, Semitic, which includes Hebrew, the language of part of our Scriptures, was included. Then Chinese, which follows an unusually simple plan of structure that is the opposite in many ways of the complex structure of old Indo-European, and which was the speech of a civilized people, was set apart as a class of the second rank. This left the majority of human languages to be dumped into a third class, or a third and fourth class, with the pleasing implication that they were less capable of abstraction, more materialistic, cruder, and generally inferior. Philologists are customarily regarded as extreme examples of passionless, dry, objective human beings. The history of this philological classification indicates that they too are influenced by emotional and self-complacent impulses.

52. Permanence of Language and Race

It is sometimes thought because a new language is readily learned, especially in youth, that language is a relatively unstable factor in human history, less permanent than race. It is necessary to guard against two fallacies in this connection. The first is to argue from individuals to societies; the second, that because change is possible, it takes place.

As a matter of fact, languages often preserve their existence, and even their territory, with surprising tenacity in the face of conquest, new religions and culture, and the economic disadvantages of unintelligibility. To-day, Breton, a Keltic dialect, maintains itself in France as the every-day language of the people in the isolated province of Brittany—a sort of philological fossil. It has withstood the influence of two thousand years of contact, first with Latin, then with Frankish German, at last with French. Its Welsh sister-tongue flourishes in spite of the Anglo-Saxon speech of the remainder of Great Britain. The original inhabitants of Spain were mostly of non-Aryan stock. Keltic, Roman, and Gothic invasions have successively swept over them and finally left the language of the country Romance, but the original speech also survives the vicissitudes of thousands of years and is still spoken in the western Pyrenees as Basque. Ancient Egypt was conquered by the Hyksos, the Assyrian, the Persian, the Macedonian, and the Roman, but whatever the official speech of the ruling class, the people continued to speak Egyptian. Finally, the Arab came and brought with him a new religion, which entailed use of the Arabic language. Egypt has at last become Arabic-speaking, but until a century or two ago the Coptic language, the daughter of the ancient Egyptian tongue of five thousand years ago, was kept alive by the native Christians along the Nile, and even to-day it survives in ritual. The boundary between French on the one side and German, Dutch, and Flemish on the other, has been accurately known for over six hundred years. With all the wars and conquests back and forth across the speech line, endless political changes and cultural influences, this line has scarcely anywhere shifted more than a few dozen miles, and in places has not moved by a comfortable afternoon’s stroll.

While populations can learn and unlearn languages, they tend to do so with reluctance and infinite slowness, especially while they remain in their inherited territories. Speech tends to be one of the most persistent ethnic characters.

In general, where two populations mingle, the speech of the more numerous prevails, even if it be the subject nationality. A wide gap in culture may overcome the influence of the majority, yet the speech of a culturally more active and advanced population ordinarily wrests permanent territory to itself slowly except where there is an actual crowding out or numerical swamping of the natives. This explains the numerous survivals and “islands” of speech: Keltic, Albanian, Basque, Caucasian, in Europe; Dravidian and Kolarian in India; Nahuatl and Maya and many others in modern Mexico; Quechua in Peru; Aymara in Bolivia; Tupi in Brazil. There are cases to the contrary, like the rapid spread of Latin in most of Gaul after Cæsar’s conquest, but they seem exceptional.

As to the relative permanence of race and speech, everything depends on the side from which the question is approached. From the point of view of hereditary strains, race must be the more conservative, because it can change rapidly only through admixture with another race, whereas a language may be completely exchanged in a short time. From the point of view of history, however, which regards human actions within given territories, speech is often more stable. Wars or trade or migration may bring one racial element after another into an area until the type has become altered or diluted, and yet the original language, or one directly descended from it, remains. The introduction of the negro from Africa to America illustrates this distinction. From the point of view of biology, the negro has at least partially preserved his type, although he has taken on a wholly new language. As a matter of history, the reverse is true: English continues to be the speech of the southern United States, whereas the population now consists of two races instead of one, and the negro element has been altered by the infusion of white blood. It is a fallacy to think, because one can learn French or become a Christian and yet is powerless to change his eye color or head shape, that language and culture are altogether less stable than race. Speech and culture have an existence of their own, whose integrity does not depend on hereditary integrity. The two may move together or separately.

53. The Biological and Historical Nature of Language

It is a truism, but one important never to forget in the study of man, that the faculty of speech is innate, but every language wholly acquired. Moreover, the environment of which languages are the product is not a natural one, that is, geographic or climatic, but social. All words and speech forms that are learned—and they constitute almost the complete mass of language—are imitated directly from other human beings. Those new forms that from time to time come into use rest on existing speech material, are shaped according to tendencies already operative although perhaps more or less hidden, cannot generally be attributed, as regards origin, or at least entire origin, to single individuals; in short, present a history similar to that of inventions and new institutions. Language thus is a superorganic product; which of course does not contradict—in fact implies—that it rests on an organic basis.

The “speech” of the animals other than man has something in common with human languages. It consists of sounds produced by the body, accompanied by certain mental activities or conditions, and capable of arousing certain definite responses in other individuals of the species. It differs from human speech in several fundamental particulars. First of all, the cries and calls and murmurs of the brutes appear to be wholly instinctive. A fowl raised alone in an incubator will peep and crow or cluck as it will scratch and peck. A dog reared by a foster cat will bark, or growl, or whine, or yelp, when it has attained the requisite age, and on application of the proper stimulus, as he will wag or crouch or hunt or dig, and no differently from the dog brought up in association with other dogs. By contrast, the Japanese infant turned over to American foster parents never utters or knows a single Japanese word, learns only English, and learns that as well as do his Caucasian step-brothers. Evidently then, animal speech is to all intents wholly organic and not at all “social” in the sense of being superorganic. If this summary is not absolutely exact, it departs from the truth only infinitesimally.

Further, animal speech has no “meaning,” does not serve as a vehicle of “communication.” The opposite is often assumed popularly, because we anthropomorphize. If it is said that a dog’s growl “means” anger, and that his bark “communicates” suspicion or excitement to his fellows, the words are used in a sense different from their significance when we say that the term red “means” the color at one end of the spectrum, or that a message of departure “communicates” information. The animal sounds convey knowledge only of subjective states. They “impart” the fact that the utterer feels anger, excitement, fear, pain, contentment, or some other affect. They are immediate reflex responses to a feeling. They may be “understood” in the sense that a sympathetic feeling is evoked or at any rate mobilized; and thereby they may lead or tend to lead to action by the hearers. In the same way, any man instinctively “understands” the moan of a fellow human being. But the moan does not tell whether the pain is of a second’s or a week’s duration, due to a blow or to gas in the bowel, to an ulcerated tooth or to mental anguish. There is no communication of anything objective, of ideas as distinct from feelings, as when we say red or break or up or water. Not one of these simple concepts can be communicated as such by any brute speech.

One consequence is the “arbitrariness” of human speech. Why should the sound-cluster red denote that particular color rather than green? Why does the same word often designate quite distinct ideas in different languages—the approximate sound group lay meaning “milk” in French; lass “a girl” in English, “tired” in French, “allow” in German? Such facts are physiologically arbitrary; just as it is physiologically arbitrary and organically meaningless that Americans live in a republic and Britons under a monarchy, or that they turn respectively to the right and left on the road. Phenomena like these have other social, cultural, or superorganic phenomena as their immediate causes or antecedents. In the light of such antecedents, viewed on the level of history, these phenomena are intelligible: we know why the United States is a republic, we can trace the development of words like lay and lass. It is only from the biological plane that such facts seem insignificant or arbitrary.

54. Problems of the Relation of Language and Culture

This association of language and civilization, or let us say the linguistic and non-linguistic constituents of culture, brings up the problem whether it would be possible for one to exist without the other. Actually, of course, no such case is known. Speculatively, different conclusions might be reached. It is difficult to imagine any generalized thinking taking place without words or symbols derived from words. Religious beliefs and certain phases of social organization also seem dependent on speech: caste ranking, marriage regulations, kinship recognition, law, and the like. On the other hand, it is conceivable that a considerable series of inventions might be made, and the applied arts might be developed in a fair measure by imitation, among a speechless people. Finally there seems no reason why certain elements of culture, such as music, should not flourish as successfully in a society without as with language.

For the converse, a cultureless species of animal might conceivably develop and use a form of true speech. Such communications as “The river is rising,” “Bite it off,” “What do you find inside?” would be within the range of thought of such a species. Why then have even the most intelligent of the brutes failed to develop a language? Possibly because such a language would lack a definite survival value for the species, in the absence of accompanying culture.

On the whole, however, it would seem that language and culture rest, in a way which is not yet fully understood, on the same set of faculties, and that these, for some reason that is still more obscure, developed in the ancestors of man, while remaining in abeyance in other species. Even the anthropoid apes seem virtually devoid of the impulse to communicate, in spite of freely expressing their affective states of mind by voice, facial gesture, and bodily movement. The most responsive to man of all species, the dog, learns to accept a considerable stock of culture in the sense of fitting himself to it: he develops conscience and manners, for example. Yet, however highly bred, he does not hand on his accomplishments to his progeny, who again depend on their human masters for what they acquire. A group of the best reared dogs left to themselves for a few years would lose all their politeness and revert to the predomestic habits of their species. In short, the culture impulse is lacking in the dog except so far as it is instilled by man; and in most animals it can notoriously be instilled only to a very limited degree. In the same way, the impulse toward communication can be said to be wanting. A dog may understand a hundred words of command and express in his behavior fifty shades of emotion; only rarely does he seem to try to communicate information of objective fact. Very likely we are attributing to him even in these rare cases the impulse which we should feel. In the event of a member of the family being injured or lost, it is certain that a good dog expresses his agitation, uneasiness, disturbed attachment; but much less certain is it that he intends to summon help, as we spontaneously incline to believe because such summoning would be our own reaction to the situation.

The history and causes of the development in incipient man of the group of traits that may be called the faculties for speech and civilization remain one of the darkest areas in the field of knowledge. It is plain that these faculties lie essentially in the sphere of what is ordinarily called the mind, rather than in the body, since men and the apes are far more similar in their general physiques than they are in the degree of their ability to use their physiques for non-physiological purposes. Or, if this antithesis of physical and mental seem unfortunate, it might be said that the growth of the faculties for speech and culture was connected more with special developments of the central nervous system than with those of the remainder of the body.

55. Period of the Origin of Language

Is, then, human language as old as culture? It is difficult to be positive, because words perish like beliefs and institutions, whereas stone tools endure as direct evidence. On the whole, however, it would appear that the first rudiments of what deserves to be called language are about as ancient as the first culture manifestations, not only because of the theoretically close association of the two phases, but in the light of circumstantial testimony, namely the skull interiors of fossil men. In Piltdown as well as Neandertal man, those brain cortex areas in which the nervous activities connected with auditory and motor speech are most centralized in modern man, are fairly well developed, as shown by casts of the skull interiors, which conform closely to the brain surface. The general frontal region, the largest area of the cortex believed to be devoted to associative functions—in loose parlance, to thought—is also greater than in any known ape. More than one authority has therefore felt justified in attributing speech to the ancestors of man that lived well back in the Pleistocene. The lower jaws of Piltdown and in a measure of Heidelberg man, it is true, are narrow and chinless, thus leaving somewhat less free play to the tongue than living human races enjoy. But this factor is probably of less importance than the one of mental facultative development. The parrot is lipless and yet can reproduce the sounds of human speech. What he lacks is language faculty; and this, it seems, fossil man already had in some measure.

56. Culture, Speech, and Nationality

This point of view raises the question whether one ought to speak of language and culture or rather of language as a part of culture. So far as the process of their transmission is concerned, and the type of mechanism of their development, it is clear that language and culture are one. For practical purposes it is generally convenient to keep them distinct. There is no doubt that two peoples can share in what is substantially the same culture and yet speak fundamentally different idioms; for instance, the Finno-Ugric Magyars or Hungarians among the adjacent Slavs, Germans, and Latins of central Europe, who are all Indo-Europeans. The other way around, the northern Hindus and west Europeans are certainly different culturally, yet their languages go back to a common origin. In fact it has become a commonplace that the arguing of connection between the three factors of race, language, and culture (or nationality), the making of inferences from one to the other, is logically unsound (§ 33). One can no more think correctly in terms of Aryan heads or a Semitic race, for instance, than of blond linguistic types, Catholic physiques, or inflecting social institutions.

At the same time, speech and culture tend to form something of a unit as opposed to race. It is possible for a population to substitute a wholly new language and type of civilization for the old ones, as the American negro has done, and yet to remain relatively unmodified racially, or at least to carry on its former physical type unchanged in a large proportion of its members. On the other hand, a change of speech without some change of culture seems impossible. Certainly wherever Greek, Latin, Spanish, English, Arabic, Pali, Chinese have penetrated, there have been established new phases of civilization. In a lower degree, the same principle probably holds true of every gain of one language at the expense of another, even when the spreading idiom is not associated with a great or active culture.

The linkage of speech and culture is further perceptible in the degree to which they both contribute, in most cases, to the idea of nationality. What chiefly marks off the French nation from the Italian, the Dutch from the German, the Swedish from the Norwegian—their respective customs and ideals, or the language gap? It would be difficult to say. The cultural differences tend to crystallize around language differences, and then in turn are reinforced by language, so that the two factors interact complexly. Nationality, especially in its modern developments, includes another factor, that of social or political segregation, which may in some degree run counter to both speech and culture. Switzerland with its German, French, and Italian speaking population, or Belgium, almost equally divided between Flemings and Walloons, are striking examples. Yet however successfully Switzerland and Belgium maintain their national unity, it is clear that this is a composite of subnational elements, each of which possesses a certain cultural as well as linguistic distinctness. Thus the Walloon speaks a French dialect, the Fleming a Dutch one; and the point of view, temperament, historic antecedents, and minor customs of the two groups are perceptibly different. Similarly, both the history and the outlook and therefore the culture of the French and German cantons of Switzerland are definitely distinguishable.

57. Relative Worth of Languages

One respect in which languages differ from cultures is that they cannot, like the latter, be rated as higher and lower. Of course, even as regards culture, such rating is often a dubious procedure, meaning little more than that the person making the comparison assumes his own culture to be the highest and estimates other cultures low in proportion as they vary. Although this is a subjective and uncritical procedure, nevertheless certain objective comparisons are possible. Some cultures surpass others in their quantitative content: they possess more different arts, abilities, and items of knowledge. Also, some culture traits may be considered intrinsically superior to others: metal tools against stone ones, for instance, since metal is adopted by all stone culture peoples who can secure it, whereas the reverse is not true. Further, in most cases a new addition does not wholly obliterate an older element, this retaining a subsidiary place, or perhaps serving some more special function than before. In this way the culture becomes more differentiated. The old art may even attain a higher degree of perfection than it had previously; as the finest polish was given to stone implements in northern Europe after bronze was known. In general, accretion is the process typical of culture growth. Older elements come to function in a more limited sphere as new ones are added, but are not extirpated by them. Oars and sails remain as constituent parts of the stock of civilization after it has added steam and motor boats. In the senses then that a culture has a larger content of elements, that these elements are more differentiated, and that a greater proportion of these elements are of the kind that inherently tend to supersede related elements, the culture may be considered superior.

As regards languages, there are also quantitative differences. Some contain several times as many words as others. But vocabulary is largely a cultural matter. A people that uses more materials, manufactures more objects, possesses knowledge of a larger array of facts, and makes finer discriminations in thought, must inevitably have more words. Yet even notable increases in size of speech content appear not to be accompanied by appreciable changes in form. A larger vocabulary does not mean a different type of structure. Grammar seems to be little influenced by culture status. No clear correspondence has yet been traceable between type or degree of civilization and type of language. Neither the presence nor the absence of particular features of tense, number, case, reduplication, or the like seems ever to have been of demonstrable advantage toward the attainment of higher culture. The speech of the former and modern nations most active in the propagation of culture has been of quite diverse type. The languages of the Egyptians (Hamitic); Sumerians; Babylonians and Arabs (Semitic); Hindus and Greeks (ancient Indo-European); Anglo-Saxons (modern Indo-European); Chinese; and Mayas, are about as different as exist. The Sumerian type of civilization was taken over bodily and successfully by the Semitic Babylonians. The bulk of Japanese culture is Chinese; yet Japanese speech is built on wholly different principles.

Then, it is impossible to rate one speech trait or type as inherently or objectively superior to another on any basis like that which justifies the placing of a metal culture above a stone culture. If wealth of grammatical apparatus is a criterion of superiority, Latin is a higher language than French, and Anglo-Saxon than English. But if lack of declensions and conjugations is a virtue, then Chinese surpasses English almost as much as English surpasses Latin. There is no reason favoring one of these possible judgments rather than its opposite. Amabo is no better or worse than I shall love as a means of expressing the same idea. The one is more compact, the other more plastic. There are times when compactness is a virtue, occasions when plasticity has advantages. By the Latin or synthetic standard, the English expression is loose jointed, lacking in structure; by the English or analytic standard, the Latin form is over-condensed, adhering unnecessarily to form. One cannot similarly balance the merits of a steel and a flint knife, of a medical and a shamanistic phase of society. The one cuts or cures better than the other.

So, from the point of view of civilization, language does not matter. Language will always keep up with whatever pace culture sets it. If a new object is invented or a new distinction of thought made, a word is coined or imported or modified in meaning to express the new concept. If a thousand or ten thousand new words are required, they are developed. When it desires to express abstractions like futurity or plurality, any language is capable of doing so, even if it does not habitually express them. If a language is unprovided with formal means for the purpose, such as a grammatical suffix, it falls back on content and uses a word or circumlocution. If the life of a people changes and comes to be conducted along lines that render it frequently important to express an idea like futurity to which previously little attention has been paid, the appropriate circumlocution soon becomes standardized, conveniently brief, and unambiguous. In general, every language is capable of indefinite modification and expansion and thereby is enabled to meet cultural demands almost at once. This is shown by the fact that virtually anything spoken or written can be translated into almost every other language without serious impairment of substance. The æsthetic charm of the original may be lost in the translation; the new forms coined in the receiving language are likely at first to seem awkward; but the meaning, the business of speech, gets expressed.

58. Size of Vocabulary

The tendency is so instinctive in us to presuppose and therefore to find qualities of inferiority, poverty, or incompleteness in the speech of populations of more backward culture than our own, that a widespread, though unfounded, belief has grown up that the languages of savages and barbarians are extremely limited quantitatively—in the range of their vocabulary. Similar misconceptions are current as to the number of words actually used by single individuals of civilized communities. It is true that no one, not even the most learned and prolific writer, uses all the words of the English language as they are found in an unabridged dictionary. All of us understand many words which we habitually encounter in reading and may even hear frequently spoken, but of which our utterance faculties for some reason have not made us master. In short, a language, being the property and product of a community, possesses more words than can ever be used by a single individual, the sum total of whose ideas is necessarily less than that of his group. Added to this are a certain mental sluggishness, which restricts most of us to a greater or less degree, and the force of habit. Having spoken a certain word a number of times, our brain becomes accustomed to it and we are likely to employ it to the exclusion of its synonyms or in place of words of related but distinguishable meaning.

The degree to which all this affects the speech of the normal man has, however, been greatly exaggerated. Because there are, all told, including technical terms, a hundred thousand or more words in our dictionaries, and because Shakespeare in his writings used 24,000 different words, Milton in his poems 17,000, and the English Bible contains 7,200, it has been concluded that the average man, whose range of thought and power of expression are so much less, must use an enormously smaller vocabulary. It has been stated that many a peasant goes through life without using more than 300 or 400 words, that the vocabulary of Italian grand opera is about 600, and that he is a person above the average who employs more than 3,000 to 4,000 words. If such were the case it would be natural that the uncivilized man, whose life is simpler, and whose knowledge more confined, should be content with an exceedingly small vocabulary.

But it is certain that the figures just cited are erroneous. If any one who considers himself an average person will take the trouble to make a list of his speaking vocabulary, he will quickly discover that he knows, and on occasion uses, the names of at least one to two thousand different things. That is, his vocabulary contains so many concrete nouns. To these must be added the abstract nouns, the verbs, adjectives, pronouns, and the other parts of speech, the short and familiar words that are indispensable to communication in any language. It may thus be safely estimated that it is an exceptionally ignorant and stupid person in a civilized country that has not at his command a vocabulary of several thousand words.

Test counts based on dictionaries show, for people of bookish tastes, a knowledge of about 30,000 to 35,000 words. Most of these would perhaps never be spoken by the individuals tested, would not be at their actual command, but it seems that at least 10,000 would be so controlled. The carefully counted vocabulary of a five and a half year old American boy comprised 1,528 understandingly used words, besides participles and other inflected forms. Two boys between two and three years used 642 and 677 different words.

It is therefore likely that statements as to the paucity of the speech of unlettered peoples are equally exaggerated. He who professes to declare on the strength of his observation that a native language consists of only a few hundred terms, displays chiefly his ignorance. He has either not taken the trouble to exhaust the vocabulary or has not known how to do so. It is true that the traveler or settler can usually converse with natives to the satisfaction of his own needs with two or three hundred words. Even the missionary can do a great deal with this stock, if it is properly chosen. But it does not follow that because a civilized person has not learned more of a language, that there is no more. On this point the testimony of the student is the evidence to be considered.

Dictionaries compiled by missionaries or philologists of languages previously unwritten run to surprising figures. Thus, the number of words recorded in Klamath, the speech of a culturally rude American Indian tribe, is 7,000; in Navaho, 11,000; in Zulu, 17,000; in Dakota, 19,000; in Maya, 20,000; in Nahuatl, 27,000. It may safely be said that every existing language, no matter how backward its speakers are in their general civilization, possesses a vocabulary of at least 5,000 words.

59. Quality of Speech Sounds

Another mistaken assumption that is frequently made is that the speech of non-literary peoples is harsh, its pronunciation more difficult than ours. This belief is purely subjective. When one has heard and uttered a language all his life, its sounds come to one’s mouth with a minimum of effort; but unfamiliar vowels and consonants are formed awkwardly and inaccurately. No adult reared in an Anglo-Saxon community finds th difficult. Nor does a French or German child, whose speech habits are still plastic, find long difficulty in mastering the particular tongue control necessary to the production of the th sound. But the adult Frenchman or German, whose muscular habits have settled in other lines, tries and tries and falls back on s or t. A Spaniard, however, would agree with the Anglo-Saxon as to the ease and “naturalness” of th. Conversely, the “rough” ch flows spontaneously out of the mouth of a German or Scotchman, whereas English, French, and Italians have to struggle long to master it, and are tempted to substitute k. German ö and French u trouble us, our “short” u is equally resistant to Continental tongues.

Even a novel position can make a familiar sound strange and forbidding. Most Anglo-Saxons fail on the first try to say ngis; many give up and declare it beyond their capacity to learn. Yet it is only sing pronounced backward. English uses ng finally and medially in words, not initially. Any English speaker can quickly acquire its use in the new position if, to keep from being disconcerted, he follows some such sequence as sing, singing, stinging, ringing, inging, nging, ngis.

So with surd l—Welsh ll—which is ordinary l minus the accompaniment of vocal cord vibrations. A little practice makes possible the throwing on or off of these vibrations, the “voicing” of speech, for any sound, with as much ease as one would turn a faucet on or off. Surd l thereupon flows with the same readiness as sonant l. As a matter of fact we often pronounce it unconsciously at the end of words like little. When it comes at the beginning, however, as in the tribal name usually written Tlingit, Americans tend to substitute something more habitual, such as kl, which is familiar from clip, clean, clear, close, clam, and many other words. The simple surd l has even been repeatedly described quite inappropriately as a “click”; which is about as far from picturing it with correctness as calling it a thump or a sigh; all because it comes in an unaccustomed position.

Combinations of sounds, especially of consonants, are indeed of variable difficulty for anatomical reasons. Some, like nd and ts and pf, have their components telescope or join naturally through being formed in the same part of the mouth. Others, like kw (qu), have the two elements articulated widely apart, but for that reason the elements can easily be formed simultaneously. Still others, like kt and ths, are intrinsically difficult, because the elements differ in place of production but are alike in method, and therefore come under the operation of the generic rule that similar sounds require more effort to join and yet discriminate than dissimilar ones; for much the same reason that it is on the whole easier to acquire the pronunciation of a wholly new type of sound than of one which differs subtly from one already known. Yet in these matters too, habit rather than anatomical functioning determines the reaction. German pf comes hard to adult Anglo-Saxons, English kw and ths to Germans. So far as degree of accumulation of consonants is concerned, English is one of the extremest of all languages. Monosyllables like tract, stripped (stripd), sixths (siksths), must seem irremediably hard to most speakers of other idioms.

Children’s speech in all languages shows that certain sounds are, as a rule, learned earlier than others, and are therefore presumably somewhat easier physiologically. Sounds like p and t which are formed with the mobile lips and front of the tongue normally precede back tongue sounds like k. B, d, g, which are voiced like vowels, tend to precede voiceless p, t, k. Stops or momentary sounds, such as b, d, g, p, t, k, generally come earlier than the fricative continuants f, v, th, s, z, which require a delicate adjustment of lip or tongue—close proximity without firm contact—whereas the stops involve only a making and breaking of jerky contact. But so slight are the differences of effort or skill in all these cases, that as a rule only a few months separate the learning of the easier from that of the more difficult sounds; and adults no longer feel the differences. The only sound or class of sounds seriously harder than others seems to be that denoted by the letter r. Not only do children usually acquire r late, but among all races there appears to be a certain percentage of individuals who never learn to form the sound right, but substitute one approaching g or w or j or l. The reason is that r stands alone among speech sounds. It is the only one produced by blowing the tongue into a few gross vibrations; which means that this organ must be held in a special condition of laxness and yet elevated so that the flow of breath may bear on it. However, even this inherent difficulty has been insufficient to prevent many languages from changing easier sounds into r.