LANGUAGE

AN INTRODUCTION TO THE STUDY OF SPEECH

BYEDWARD SAPIR

1939

1921

PREFACE

This little book aims to give a certain perspective on the subject oflanguage rather than to assemble facts about it. It has little to say ofthe ultimate psychological basis of speech and gives only enough of theactual descriptive or historical facts of particular languages toillustrate principles. Its main purpose is to show what I conceivelanguage to be, what is its variability in place and time, and what areits relations to other fundamental human interests--the problem ofthought, the nature of the historical process, race, culture, art. 

The perspective thus gained will be useful, I hope, both to linguisticstudents and to the outside public that is half inclined to dismisslinguistic notions as the private pedantries of essentially idle minds. Knowledge of the wider relations of their science is essential toprofessional students of language if they are to be saved from a sterileand purely technical attitude. Among contemporary writers of influenceon liberal thought Croce is one of the very few who have gained anunderstanding of the fundamental significance of language. He haspointed out its close relation to the problem of art. I am deeplyindebted to him for this insight. Quite aside from their intrinsicinterest, linguistic forms and historical processes have the greatestpossible diagnostic value for the understanding of some of the moredifficult and elusive problems in the psychology of thought and in thestrange, cumulative drift in the life of the human spirit that we callhistory or progress or evolution. This value depends chiefly on theunconscious and unrationalized nature of linguistic structure. 

I have avoided most of the technical terms and all of the technicalsymbols of the linguistic academy. There is not a single diacriticalmark in the book. Where possible, the discussion is based on Englishmaterial. It was necessary, however, for the scheme of the book, whichincludes a consideration of the protean forms in which human thought hasfound expression, to quote some exotic instances. For these no apologyseems necessary. Owing to limitations of space I have had to leave outmany ideas or principles that I should have liked to touch upon. Otherpoints have had to be barely hinted at in a sentence or flying phrase. Nevertheless, I trust that enough has here been brought together toserve as a stimulus for the more fundamental study of a neglected field. 

I desire to express my cordial appreciation of the friendly advice andhelpful suggestions of a number of friends who have read the work inmanuscript, notably Profs. A. L. Kroeber and R. H. Lowie of the Universityof California, Prof. W. D. Wallis of Reed College, and Prof. J. Zeitlinof the University of Illinois. 

EDWARD SAPIR. 

OTTAWA, ONT. , April 8, 1921. 

CONTENTS

PREFACE

CHAPTER

 I. INTRODUCTORY: LANGUAGE DEFINED

 Language a cultural, not a biologically inherited, function. Futility of interjectional and sound-imitative theories of the origin of speech. Definition of language. The psycho-physical basis of speech. Concepts and language. Is thought possible without language? Abbreviations and transfers of the speech process. The universality of language. 

 II. THE ELEMENTS OF SPEECH

 Sounds not properly elements of speech. Words and significant parts of words (radical elements, grammatical elements). Types of words. The word a formal, not a functional unit. The word has a real psychological existence. The sentence. The cognitive, volitional, and emotional aspects of speech. Feeling-tones of words. 

 III. THE SOUNDS OF LANGUAGE

 The vast number of possible sounds. The articulating organs and their share in the production of speech sounds: lungs, glottal cords, nose, mouth and its parts. Vowel articulations. How and where consonants are articulated. The phonetic habits of a language. The "values" of sounds. Phonetic patterns. 

 IV. FORM IN LANGUAGE: GRAMMATICAL PROCESSES

 Formal processes as distinct from grammatical functions. Intercrossing of the two points of view. Six main types of grammatical process. Word sequence as a method. Compounding of radical elements. Affixing: prefixes and suffixes; infixes. Internal vocalic change; consonantal change. Reduplication. Functional variations of stress; of pitch. 

 V. FORM IN LANGUAGE: GRAMMATICAL CONCEPTS

 Analysis of a typical English sentence. Types of concepts illustrated by it. Inconsistent expression of analogous concepts. How the same sentence may be expressed in other languages with striking differences in the selection and grouping of concepts. Essential and non-essential concepts. The mixing of essential relational concepts with secondary ones of more concrete order. Form for form's sake. Classification of linguistic concepts: basic or concrete, derivational, concrete relational, pure relational. Tendency for these types of concepts to flow into each other. Categories expressed in various grammatical systems. Order and stress as relating principles in the sentence. Concord. Parts of speech: no absolute classification possible; noun and verb. 

 VI. TYPES OF LINGUISTIC STRUCTURE

 The possibility of classifying languages. Difficulties. Classification into form-languages and formless languages not valid. Classification according to formal processes used not practicable. Classification according to degree of synthesis. "Inflective" and "agglutinative. " Fusion and symbolism as linguistic techniques. Agglutination. "Inflective" a confused term. Threefold classification suggested: what types of concepts are expressed? what is the prevailing technique? what is the degree of synthesis? Four fundamental conceptual types. Examples tabulated. Historical test of the validity of the suggested conceptual classification. 

 VII. LANGUAGE AS A HISTORICAL PRODUCT: DRIFT

 Variability of language. Individual and dialectic variations. Time variation or "drift. " How dialects arise. Linguistic stocks. Direction or "slope" of linguistic drift. Tendencies illustrated in an English sentence. Hesitations of usage as symptomatic of the direction of drift. Leveling tendencies in English. Weakening of case elements. Tendency to fixed position in the sentence. Drift toward the invariable word. 

 VIII. LANGUAGE AS A HISTORICAL PRODUCT: PHONETIC LAW

 Parallels in drift in related languages. Phonetic law as illustrated in the history of certain English and German vowels and consonants. Regularity of phonetic law. Shifting of sounds without destruction of phonetic pattern. Difficulty of explaining the nature of phonetic drifts. Vowel mutation in English and German. Morphological influence on phonetic change. Analogical levelings to offset irregularities produced by phonetic laws. New morphological features due to phonetic change. 

 IX. HOW LANGUAGES INFLUENCE EACH OTHER

 Linguistic influences due to cultural contact. Borrowing of words. Resistances to borrowing. Phonetic modification of borrowed words. Phonetic interinfluencings of neighboring languages. Morphological borrowings. Morphological resemblances as vestiges of genetic relationship. 

 X. LANGUAGE, RACE, AND CULTURE

 Naïve tendency to consider linguistic, racial, and cultural groupings as congruent. Race and language need not correspond. Cultural and linguistic boundaries not identical. Coincidences between linguistic cleavages and those of language and culture due to historical, not intrinsic psychological, causes. Language does not in any deep sense "reflect" culture. 

 XL LANGUAGE AND LITERATURE

 Language as the material or medium of literature. Literature may move on the generalized linguistic plane or may be inseparable from specific linguistic conditions. Language as a collective art. Necessary esthetic advantages or limitations in any language. Style as conditioned by inherent features of the language. Prosody as conditioned by the phonetic dynamics of a language. 

INDEX

I

INTRODUCTORY: LANGUAGE DEFINED

Speech is so familiar a feature of daily life that we rarely pause todefine it. It seems as natural to man as walking, and only less so thanbreathing. Yet it needs but a moment's reflection to convince us thatthis naturalness of speech is but an illusory feeling. The process ofacquiring speech is, in sober fact, an utterly different sort of thingfrom the process of learning to walk. In the case of the latterfunction, culture, in other words, the traditional body of social usage, is not seriously brought into play. The child is individually equipped, by the complex set of factors that we term biological heredity, to makeall the needed muscular and nervous adjustments that result in walking. Indeed, the very conformation of these muscles and of the appropriateparts of the nervous system may be said to be primarily adapted to themovements made in walking and in similar activities. In a very realsense the normal human being is predestined to walk, not because hiselders will assist him to learn the art, but because his organism isprepared from birth, or even from the moment of conception, to take onall those expenditures of nervous energy and all those muscularadaptations that result in walking. To put it concisely, walking is aninherent, biological function of man. 

Not so language. It is of course true that in a certain sense theindividual is predestined to talk, but that is due entirely to thecircumstance that he is born not merely in nature, but in the lap of asociety that is certain, reasonably certain, to lead him to itstraditions. Eliminate society and there is every reason to believe thathe will learn to walk, if, indeed, he survives at all. But it is just ascertain that he will never learn to talk, that is, to communicate ideasaccording to the traditional system of a particular society. Or, again, remove the new-born individual from the social environment into which hehas come and transplant him to an utterly alien one. He will develop theart of walking in his new environment very much as he would havedeveloped it in the old. But his speech will be completely at variancewith the speech of his native environment. Walking, then, is a generalhuman activity that varies only within circumscribed limits as we passfrom individual to individual. Its variability is involuntary andpurposeless. Speech is a human activity that varies without assignablelimit as we pass from social group to social group, because it is apurely historical heritage of the group, the product of long-continuedsocial usage. It varies as all creative effort varies--not asconsciously, perhaps, but none the less as truly as do the religions, the beliefs, the customs, and the arts of different peoples. Walking isan organic, an instinctive, function (not, of course, itself aninstinct); speech is a non-instinctive, acquired, "cultural" function. 

There is one fact that has frequently tended to prevent the recognitionof language as a merely conventional system of sound symbols, that hasseduced the popular mind into attributing to it an instinctive basisthat it does not really possess. This is the well-known observation thatunder the stress of emotion, say of a sudden twinge of pain or ofunbridled joy, we do involuntarily give utterance to sounds that thehearer interprets as indicative of the emotion itself. But there is allthe difference in the world between such involuntary expression offeeling and the normal type of communication of ideas that is speech. The former kind of utterance is indeed instinctive, but it isnon-symbolic; in other words, the sound of pain or the sound of joy doesnot, as such, indicate the emotion, it does not stand aloof, as it were, and announce that such and such an emotion is being felt. What it doesis to serve as a more or less automatic overflow of the emotionalenergy; in a sense, it is part and parcel of the emotion itself. Moreover, such instinctive cries hardly constitute communication in anystrict sense. They are not addressed to any one, they are merelyoverheard, if heard at all, as the bark of a dog, the sound ofapproaching footsteps, or the rustling of the wind is heard. If theyconvey certain ideas to the hearer, it is only in the very general sensein which any and every sound or even any phenomenon in our environmentmay be said to convey an idea to the perceiving mind. If the involuntarycry of pain which is conventionally represented by "Oh!" be looked uponas a true speech symbol equivalent to some such idea as "I am in greatpain, " it is just as allowable to interpret the appearance of clouds asan equivalent symbol that carries the definite message "It is likely torain. " A definition of language, however, that is so extended as tocover every type of inference becomes utterly meaningless. 

The mistake must not be made of identifying our conventionalinterjections (our oh! and ah! and sh!) with the instinctive criesthemselves. These interjections are merely conventional fixations of thenatural sounds. They therefore differ widely in various languages inaccordance with the specific phonetic genius of each of these. As suchthey may be considered an integral portion of speech, in the properlycultural sense of the term, being no more identical with the instinctivecries themselves than such words as "cuckoo" and "kill-deer" areidentical with the cries of the birds they denote or than Rossini'streatment of a storm in the overture to "William Tell" is in fact astorm. In other words, the interjections and sound-imitative words ofnormal speech are related to their natural prototypes as is art, apurely social or cultural thing, to nature. It may be objected that, though the interjections differ somewhat as we pass from language tolanguage, they do nevertheless offer striking family resemblances andmay therefore be looked upon as having grown up out of a commoninstinctive base. But their case is nowise different from that, say, ofthe varying national modes of pictorial representation. A Japanesepicture of a hill both differs from and resembles a typical modernEuropean painting of the same kind of hill. Both are suggested by andboth "imitate" the same natural feature. Neither the one nor the otheris the same thing as, or, in any intelligible sense, a direct outgrowthof, this natural feature. The two modes of representation are notidentical because they proceed from differing historical traditions, areexecuted with differing pictorial techniques. The interjections ofJapanese and English are, just so, suggested by a common naturalprototype, the instinctive cries, and are thus unavoidably suggestive ofeach other. They differ, now greatly, now but little, because they arebuilded out of historically diverse materials or techniques, therespective linguistic traditions, phonetic systems, speech habits of thetwo peoples. Yet the instinctive cries as such are practically identicalfor all humanity, just as the human skeleton or nervous system is to allintents and purposes a "fixed, " that is, an only slightly and"accidentally" variable, feature of man's organism. 

Interjections are among the least important of speech elements. Theirdiscussion is valuable mainly because it can be shown that even they, avowedly the nearest of all language sounds to instinctive utterance, are only superficially of an instinctive nature. Were it thereforepossible to demonstrate that the whole of language is traceable, in itsultimate historical and psychological foundations, to the interjections, it would still not follow that language is an instinctive activity. But, as a matter of fact, all attempts so to explain the origin of speechhave been fruitless. There is no tangible evidence, historical orotherwise, tending to show that the mass of speech elements and speechprocesses has evolved out of the interjections. These are a very smalland functionally insignificant proportion of the vocabulary of language;at no time and in no linguistic province that we have record of do wesee a noticeable tendency towards their elaboration into the primarywarp and woof of language. They are never more, at best, than adecorative edging to the ample, complex fabric. 

What applies to the interjections applies with even greater force to thesound-imitative words. Such words as "whippoorwill, " "to mew, " "to caw"are in no sense natural sounds that man has instinctively orautomatically reproduced. They are just as truly creations of the humanmind, flights of the human fancy, as anything else in language. They donot directly grow out of nature, they are suggested by it and play withit. Hence the onomatopoetic theory of the origin of speech, the theorythat would explain all speech as a gradual evolution from sounds of animitative character, really brings us no nearer to the instinctive levelthan is language as we know it to-day. As to the theory itself, it isscarcely more credible than its interjectional counterpart. It is truethat a number of words which we do not now feel to have asound-imitative value can be shown to have once had a phonetic form thatstrongly suggests their origin as imitations of natural sounds. Such isthe English word "to laugh. " For all that, it is quite impossible toshow, nor does it seem intrinsically reasonable to suppose, that morethan a negligible proportion of the elements of speech or anything atall of its formal apparatus is derivable from an onomatopoetic source. However much we may be disposed on general principles to assign afundamental importance in the languages of primitive peoples to theimitation of natural sounds, the actual fact of the matter is that theselanguages show no particular preference for imitative words. Among themost primitive peoples of aboriginal America, the Athabaskan tribes ofthe Mackenzie River speak languages in which such words seem to benearly or entirely absent, while they are used freely enough inlanguages as sophisticated as English and German. Such an instance showshow little the essential nature of speech is concerned with the mereimitation of things. 

The way is now cleared for a serviceable definition of language. Language is a purely human and non-instinctive method of communicatingideas, emotions, and desires by means of a system of voluntarilyproduced symbols. These symbols are, in the first instance, auditory andthey are produced by the so-called "organs of speech. " There is nodiscernible instinctive basis in human speech as such, however muchinstinctive expressions and the natural environment may serve as astimulus for the development of certain elements of speech, however muchinstinctive tendencies, motor and other, may give a predetermined rangeor mold to linguistic expression. Such human or animal communication, if"communication" it may be called, as is brought about by involuntary, instinctive cries is not, in our sense, language at all. 

I have just referred to the "organs of speech, " and it would seem atfirst blush that this is tantamount to an admission that speech itselfis an instinctive, biologically predetermined activity. We must not bemisled by the mere term. There are, properly speaking, no organs ofspeech; there are only organs that are incidentally useful in theproduction of speech sounds. The lungs, the larynx, the palate, thenose, the tongue, the teeth, and the lips, are all so utilized, but theyare no more to be thought of as primary organs of speech than are thefingers to be considered as essentially organs of piano-playing or theknees as organs of prayer. Speech is not a simple activity that iscarried on by one or more organs biologically adapted to the purpose. Itis an extremely complex and ever-shifting network of adjustments--in thebrain, in the nervous system, and in the articulating and auditoryorgans--tending towards the desired end of communication. The lungsdeveloped, roughly speaking, in connection with the necessarybiological function known as breathing; the nose, as an organ of smell;the teeth, as organs useful in breaking up food before it was ready fordigestion. If, then, these and other organs are being constantlyutilized in speech, it is only because any organ, once existent and inso far as it is subject to voluntary control, can be utilized by man forsecondary purposes. Physiologically, speech is an overlaid function, or, to be more precise, a group of overlaid functions. It gets what serviceit can out of organs and functions, nervous and muscular, that have comeinto being and are maintained for very different ends than its own. 

It is true that physiological psychologists speak of the localization ofspeech in the brain. This can only mean that the sounds of speech arelocalized in the auditory tract of the brain, or in some circumscribedportion of it, precisely as other classes of sounds are localized; andthat the motor processes involved in speech (such as the movements ofthe glottal cords in the larynx, the movements of the tongue required topronounce the vowels, lip movements required to articulate certainconsonants, and numerous others) are localized in the motor tractprecisely as are all other impulses to special motor activities. In thesame way control is lodged in the visual tract of the brain over allthose processes of visual recognition involved in reading. Naturally theparticular points or clusters of points of localization in the severaltracts that refer to any element of language are connected in the brainby paths of association, so that the outward, or psycho-physical, aspectof language, is of a vast network of associated localizations in thebrain and lower nervous tracts, the auditory localizations being withoutdoubt the most fundamental of all for speech. However, a speechsoundlocalized in the brain, even when associated with the particularmovements of the "speech organs" that are required to produce it, isvery far from being an element of language. It must be furtherassociated with some element or group of elements of experience, say avisual image or a class of visual images or a feeling of relation, before it has even rudimentary linguistic significance. This "element"of experience is the content or "meaning" of the linguistic unit; theassociated auditory, motor, and other cerebral processes that lieimmediately back of the act of speaking and the act of hearing speechare merely a complicated symbol of or signal for these "meanings, " ofwhich more anon. We see therefore at once that language as such is notand cannot be definitely localized, for it consists of a peculiarsymbolic relation--physiologically an arbitrary one--between allpossible elements of consciousness on the one hand and certain selectedelements localized in the auditory, motor, and other cerebral andnervous tracts on the other. If language can be said to be definitely"localized" in the brain, it is only in that general and rather uselesssense in which all aspects of consciousness, all human interest andactivity, may be said to be "in the brain. " Hence, we have no recoursebut to accept language as a fully formed functional system within man'spsychic or "spiritual" constitution. We cannot define it as an entity inpsycho-physical terms alone, however much the psycho-physical basis isessential to its functioning in the individual. 

From the physiologist's or psychologist's point of view we may seem tobe making an unwarrantable abstraction in desiring to handle the subjectof speech without constant and explicit reference to that basis. However, such an abstraction is justifiable. We can profitably discussthe intention, the form, and the history of speech, precisely as wediscuss the nature of any other phase of human culture--say art orreligion--as an institutional or cultural entity, leaving the organicand psychological mechanisms back of it as something to be taken forgranted. Accordingly, it must be clearly understood that thisintroduction to the study of speech is not concerned with those aspectsof physiology and of physiological psychology that underlie speech. Ourstudy of language is not to be one of the genesis and operation of aconcrete mechanism; it is, rather, to be an inquiry into the functionand form of the arbitrary systems of symbolism that we term languages. 

I have already pointed out that the essence of language consists in theassigning of conventional, voluntarily articulated, sounds, or of theirequivalents, to the diverse elements of experience. The word "house" isnot a linguistic fact if by it is meant merely the acoustic effectproduced on the ear by its constituent consonants and vowels, pronouncedin a certain order; nor the motor processes and tactile feelings whichmake up the articulation of the word; nor the visual perception on thepart of the hearer of this articulation; nor the visual perception ofthe word "house" on the written or printed page; nor the motor processesand tactile feelings which enter into the writing of the word; nor thememory of any or all of these experiences. It is only when these, andpossibly still other, associated experiences are automaticallyassociated with the image of a house that they begin to take on thenature of a symbol, a word, an element of language. But the mere fact ofsuch an association is not enough. One might have heard a particularword spoken in an individual house under such impressive circumstancesthat neither the word nor the image of the house ever recur inconsciousness without the other becoming present at the same time. Thistype of association does not constitute speech. The association must bea purely symbolic one; in other words, the word must denote, tag off, the image, must have no other significance than to serve as a counter torefer to it whenever it is necessary or convenient to do so. Such anassociation, voluntary and, in a sense, arbitrary as it is, demands aconsiderable exercise of self-conscious attention. At least to beginwith, for habit soon makes the association nearly as automatic as anyand more rapid than most. 

But we have traveled a little too fast. Were the symbol "house"--whetheran auditory, motor, or visual experience or image--attached but to thesingle image of a particular house once seen, it might perhaps, by anindulgent criticism, be termed an element of speech, yet it is obviousat the outset that speech so constituted would have little or no valuefor purposes of communication. The world of our experiences must beenormously simplified and generalized before it is possible to make asymbolic inventory of all our experiences of things and relations; andthis inventory is imperative before we can convey ideas. The elements oflanguage, the symbols that ticket off experience, must therefore beassociated with whole groups, delimited classes, of experience ratherthan with the single experiences themselves. Only so is communicationpossible, for the single experience lodges in an individualconsciousness and is, strictly speaking, incommunicable. To becommunicated it needs to be referred to a class which is tacitlyaccepted by the community as an identity. Thus, the single impressionwhich I have had of a particular house must be identified with all myother impressions of it. Further, my generalized memory or my "notion"of this house must be merged with the notions that all other individualswho have seen the house have formed of it. The particular experiencethat we started with has now been widened so as to embrace all possibleimpressions or images that sentient beings have formed or may form ofthe house in question. This first simplification of experience is at thebottom of a large number of elements of speech, the so-called propernouns or names of single individuals or objects. It is, essentially, thetype of simplification which underlies, or forms the crude subject of, history and art. But we cannot be content with this measure of reductionof the infinity of experience. We must cut to the bone of things, wemust more or less arbitrarily throw whole masses of experience togetheras similar enough to warrant their being looked upon--mistakenly, butconveniently--as identical. This house and that house and thousands ofother phenomena of like character are thought of as having enough incommon, in spite of great and obvious differences of detail, to beclassed under the same heading. In other words, the speech element"house" is the symbol, first and foremost, not of a single perception, nor even of the notion of a particular object, but of a "concept, " inother words, of a convenient capsule of thought that embraces thousandsof distinct experiences and that is ready to take in thousands more. Ifthe single significant elements of speech are the symbols of concepts, the actual flow of speech may be interpreted as a record of the settingof these concepts into mutual relations. 

The question has often been raised whether thought is possible withoutspeech; further, if speech and thought be not but two facets of the samepsychic process. The question is all the more difficult because it hasbeen hedged about by misunderstandings. In the first place, it is wellto observe that whether or not thought necessitates symbolism, that isspeech, the flow of language itself is not always indicative of thought. We have seen that the typical linguistic element labels a concept. Itdoes not follow from this that the use to which language is put isalways or even mainly conceptual. We are not in ordinary life so muchconcerned with concepts as such as with concrete particularities andspecific relations. When I say, for instance, "I had a good breakfastthis morning, " it is clear that I am not in the throes of laboriousthought, that what I have to transmit is hardly more than a pleasurablememory symbolically rendered in the grooves of habitual expression. Eachelement in the sentence defines a separate concept or conceptualrelation or both combined, but the sentence as a whole has no conceptualsignificance whatever. It is somewhat as though a dynamo capable ofgenerating enough power to run an elevator were operated almostexclusively to feed an electric door-bell. The parallel is moresuggestive than at first sight appears. Language may be looked upon asan instrument capable of running a gamut of psychic uses. Its flow notonly parallels that of the inner content of consciousness, but parallelsit on different levels, ranging from the state of mind that is dominatedby particular images to that in which abstract concepts and theirrelations are alone at the focus of attention and which is ordinarilytermed reasoning. Thus the outward form only of language is constant;its inner meaning, its psychic value or intensity, varies freely withattention or the selective interest of the mind, also, needless to say, with the mind's general development. From the point of view oflanguage, thought may be defined as the highest latent or potentialcontent of speech, the content that is obtained by interpreting each ofthe elements in the flow of language as possessed of its very fullestconceptual value. From this it follows at once that language and thoughtare not strictly coterminous. At best language can but be the outwardfacet of thought on the highest, most generalized, level of symbolicexpression. To put our viewpoint somewhat differently, language isprimarily a pre-rational function. It humbly works up to the thoughtthat is latent in, that may eventually be read into, its classificationsand its forms; it is not, as is generally but naïvely assumed, the finallabel put upon, the finished thought. 

Most people, asked if they can think without speech, would probablyanswer, "Yes, but it is not easy for me to do so. Still I know it can bedone. " Language is but a garment! But what if language is not so much agarment as a prepared road or groove? It is, indeed, in the highestdegree likely that language is an instrument originally put to useslower than the conceptual plane and that thought arises as a refinedinterpretation of its content. The product grows, in other words, withthe instrument, and thought may be no more conceivable, in its genesisand daily practice, without speech than is mathematical reasoningpracticable without the lever of an appropriate mathematical symbolism. No one believes that even the most difficult mathematical proposition isinherently dependent on an arbitrary set of symbols, but it isimpossible to suppose that the human mind is capable of arriving at orholding such a proposition without the symbolism. The writer, for one, is strongly of the opinion that the feeling entertained by so many thatthey can think, or even reason, without language is an illusion. Theillusion seems to be due to a number of factors. The simplest of theseis the failure to distinguish between imagery and thought. As a matterof fact, no sooner do we try to put an image into conscious relationwith another than we find ourselves slipping into a silent flow ofwords. Thought may be a natural domain apart from the artificial one ofspeech, but speech would seem to be the only road we know of that leadsto it. A still more fruitful source of the illusive feeling thatlanguage may be dispensed with in thought is the common failure torealize that language is not identical with its auditory symbolism. Theauditory symbolism may be replaced, point for point, by a motor or by avisual symbolism (many people can read, for instance, in a purely visualsense, that is, without the intermediating link of an inner flow of theauditory images that correspond to the printed or written words) or bystill other, more subtle and elusive, types of transfer that are not soeasy to define. Hence the contention that one thinks without languagemerely because he is not aware of a coexisting auditory imagery is veryfar indeed from being a valid one. One may go so far as to suspect thatthe symbolic expression of thought may in some cases run along outsidethe fringe of the conscious mind, so that the feeling of a free, nonlinguistic stream of thought is for minds of a certain type arelatively, but only a relatively, justified one. Psycho-physically, this would mean that the auditory or equivalent visual or motor centersin the brain, together with the appropriate paths of association, thatare the cerebral equivalent of speech, are touched off so lightly duringthe process of thought as not to rise into consciousness at all. Thiswould be a limiting case--thought riding lightly on the submerged crestsof speech, instead of jogging along with it, hand in hand. The modernpsychology has shown us how powerfully symbolism is at work in theunconscious mind. It is therefore easier to understand at the presenttime than it would have been twenty years ago that the most rarefiedthought may be but the conscious counterpart of an unconsciouslinguistic symbolism. 

One word more as to the relation between language and thought. The pointof view that we have developed does not by any means preclude thepossibility of the growth of speech being in a high degree dependent onthe development of thought. We may assume that language arosepre-rationally--just how and on what precise level of mental activity wedo not know--but we must not imagine that a highly developed system ofspeech symbols worked itself out before the genesis of distinct conceptsand of thinking, the handling of concepts. We must rather imagine thatthought processes set in, as a kind of psychic overflow, almost at thebeginning of linguistic expression; further, that the concept, oncedefined, necessarily reacted on the life of its linguistic symbol, encouraging further linguistic growth. We see this complex process ofthe interaction of language and thought actually taking place under oureyes. The instrument makes possible the product, the product refines theinstrument. The birth of a new concept is invariably foreshadowed by amore or less strained or extended use of old linguistic material; theconcept does not attain to individual and independent life until it hasfound a distinctive linguistic embodiment. In most cases the new symbolis but a thing wrought from linguistic material already in existence inways mapped out by crushingly despotic precedents. As soon as the wordis at hand, we instinctively feel, with something of a sigh of relief, that the concept is ours for the handling. Not until we own the symboldo we feel that we hold a key to the immediate knowledge orunderstanding of the concept. Would we be so ready to die for "liberty, "to struggle for "ideals, " if the words themselves were not ringingwithin us? And the word, as we know, is not only a key; it may also be afetter. 

Language is primarily an auditory system of symbols. In so far as it isarticulated it is also a motor system, but the motor aspect of speech isclearly secondary to the auditory. In normal individuals the impulse tospeech first takes effect in the sphere of auditory imagery and is thentransmitted to the motor nerves that control the organs of speech. Themotor processes and the accompanying motor feelings are not, however, the end, the final resting point. They are merely a means and a controlleading to auditory perception in both speaker and hearer. Communication, which is the very object of speech, is successfullyeffected only when the hearer's auditory perceptions are translated intothe appropriate and intended flow of imagery or thought or bothcombined. Hence the cycle of speech, in so far as we may look upon it asa purely external instrument, begins and ends in the realm of sounds. The concordance between the initial auditory imagery and the finalauditory perceptions is the social seal or warrant of the successfulissue of the process. As we have already seen, the typical course ofthis process may undergo endless modifications or transfers intoequivalent systems without thereby losing its essential formalcharacteristics. 

The most important of these modifications is the abbreviation of thespeech process involved in thinking. This has doubtless many forms, according to the structural or functional peculiarities of theindividual mind. The least modified form is that known as "talking toone's self" or "thinking aloud. " Here the speaker and the hearer areidentified in a single person, who may be said to communicate withhimself. More significant is the still further abbreviated form in whichthe sounds of speech are not articulated at all. To this belong all thevarieties of silent speech and of normal thinking. The auditory centersalone may be excited; or the impulse to linguistic expression may becommunicated as well to the motor nerves that communicate with theorgans of speech but be inhibited either in the muscles of these organsor at some point in the motor nerves themselves; or, possibly, theauditory centers may be only slightly, if at all, affected, the speechprocess manifesting itself directly in the motor sphere. There must bestill other types of abbreviation. How common is the excitation of themotor nerves in silent speech, in which no audible or visiblearticulations result, is shown by the frequent experience of fatigue inthe speech organs, particularly in the larynx, after unusuallystimulating reading or intensive thinking. 

All the modifications so far considered are directly patterned on thetypical process of normal speech. Of very great interest and importanceis the possibility of transferring the whole system of speech symbolisminto other terms than those that are involved in the typical process. This process, as we have seen, is a matter of sounds and of movementsintended to produce these sounds. The sense of vision is not broughtinto play. But let us suppose that one not only hears the articulatedsounds but sees the articulations themselves as they are being executedby the speaker. Clearly, if one can only gain a sufficiently high degreeof adroitness in perceiving these movements of the speech organs, theway is opened for a new type of speech symbolism--that in which thesound is replaced by the visual image of the articulations thatcorrespond to the sound. This sort of system has no great value for mostof us because we are already possessed of the auditory-motor system ofwhich it is at best but an imperfect translation, not all thearticulations being visible to the eye. However, it is well known whatexcellent use deaf-mutes can make of "reading from the lips" as asubsidiary method of apprehending speech. The most important of allvisual speech symbolisms is, of course, that of the written or printedword, to which, on the motor side, corresponds the system of delicatelyadjusted movements which result in the writing or typewriting or othergraphic method of recording speech. The significant feature for ourrecognition in these new types of symbolism, apart from the fact thatthey are no longer a by-product of normal speech itself, is that eachelement (letter or written word) in the system corresponds to a specificelement (sound or sound-group or spoken word) in the primary system. Written language is thus a point-to-point equivalence, to borrow amathematical phrase, to its spoken counterpart. The written forms aresecondary symbols of the spoken ones--symbols of symbols--yet so closeis the correspondence that they may, not only in theory but in theactual practice of certain eye-readers and, possibly, in certain typesof thinking, be entirely substituted for the spoken ones. Yet theauditory-motor associations are probably always latent at the least, that is, they are unconsciously brought into play. Even those who readand think without the slightest use of sound imagery are, at lastanalysis, dependent on it. They are merely handling the circulatingmedium, the money, of visual symbols as a convenient substitute for theeconomic goods and services of the fundamental auditory symbols. 

The possibilities of linguistic transfer are practically unlimited. Afamiliar example is the Morse telegraph code, in which the letters ofwritten speech are represented by a conventionally fixed sequence oflonger or shorter ticks. Here the transfer takes place from the writtenword rather than directly from the sounds of spoken speech. The letterof the telegraph code is thus a symbol of a symbol of a symbol. It doesnot, of course, in the least follow that the skilled operator, in orderto arrive at an understanding of a telegraphic message, needs totranspose the individual sequence of ticks into a visual image of theword before he experiences its normal auditory image. The precise methodof reading off speech from the telegraphic communication undoubtedlyvaries widely with the individual. It is even conceivable, if notexactly likely, that certain operators may have learned to thinkdirectly, so far as the purely conscious part of the process of thoughtis concerned, in terms of the tick-auditory symbolism or, if they happento have a strong natural bent toward motor symbolism, in terms of thecorrelated tactile-motor symbolism developed in the sending oftelegraphic messages. 

Still another interesting group of transfers are the different gesturelanguages, developed for the use of deaf-mutes, of Trappist monks vowedto perpetual silence, or of communicating parties that are within seeingdistance of each other but are out of earshot. Some of these systems areone-to-one equivalences of the normal system of speech; others, likemilitary gesture-symbolism or the gesture language of the Plains Indiansof North America (understood by tribes of mutually unintelligible formsof speech) are imperfect transfers, limiting themselves to the renderingof such grosser speech elements as are an imperative minimum underdifficult circumstances. In these latter systems, as in such still moreimperfect symbolisms as those used at sea or in the woods, it may becontended that language no longer properly plays a part but that theideas are directly conveyed by an utterly unrelated symbolic process orby a quasi-instinctive imitativeness. Such an interpretation would beerroneous. The intelligibility of these vaguer symbolisms can hardly bedue to anything but their automatic and silent translation into theterms of a fuller flow of speech. 

We shall no doubt conclude that all voluntary communication of ideas, aside from normal speech, is either a transfer, direct or indirect, fromthe typical symbolism of language as spoken and heard or, at the least, involves the intermediary of truly linguistic symbolism. This is a factof the highest importance. Auditory imagery and the correlated motorimagery leading to articulation are, by whatever devious ways we followthe process, the historic fountain-head of all speech and of allthinking. One other point is of still greater importance. The ease withwhich speech symbolism can be transferred from one sense to another, from technique to technique, itself indicates that the mere sounds ofspeech are not the essential fact of language, which lies rather in theclassification, in the formal patterning, and in the relating ofconcepts. Once more, language, as a structure, is on its inner face themold of thought. It is this abstracted language, rather more than thephysical facts of speech, that is to concern us in our inquiry. 

There is no more striking general fact about language than itsuniversality. One may argue as to whether a particular tribe engages inactivities that are worthy of the name of religion or of art, but weknow of no people that is not possessed of a fully developed language. The lowliest South African Bushman speaks in the forms of a richsymbolic system that is in essence perfectly comparable to the speech ofthe cultivated Frenchman. It goes without saying that the more abstractconcepts are not nearly so plentifully represented in the language ofthe savage, nor is there the rich terminology and the finer definitionof nuances that reflect the higher culture. Yet the sort of linguisticdevelopment that parallels the historic growth of culture and which, inits later stages, we associate with literature is, at best, but asuperficial thing. The fundamental groundwork of language--thedevelopment of a clear-cut phonetic system, the specific association ofspeech elements with concepts, and the delicate provision for the formalexpression of all manner of relations--all this meets us rigidlyperfected and systematized in every language known to us. Many primitivelanguages have a formal richness, a latent luxuriance of expression, that eclipses anything known to the languages of modern civilization. Even in the mere matter of the inventory of speech the layman must beprepared for strange surprises. Popular statements as to the extremepoverty of expression to which primitive languages are doomed are simplymyths. Scarcely less impressive than the universality of speech is itsalmost incredible diversity. Those of us that have studied French orGerman, or, better yet, Latin or Greek, know in what varied forms athought may run. The formal divergences between the English plan and theLatin plan, however, are comparatively slight in the perspective of whatwe know of more exotic linguistic patterns. The universality and thediversity of speech lead to a significant inference. We are forced tobelieve that language is an immensely ancient heritage of the humanrace, whether or not all forms of speech are the historical outgrowth ofa single pristine form. It is doubtful if any other cultural asset ofman, be it the art of drilling for fire or of chipping stone, may layclaim to a greater age. I am inclined to believe that it antedated eventhe lowliest developments of material culture, that these developments, in fact, were not strictly possible until language, the tool ofsignificant expression, had itself taken shape. 

II

THE ELEMENTS OF SPEECH

We have more than once referred to the "elements of speech, " by which weunderstood, roughly speaking, what are ordinarily called "words. " Wemust now look more closely at these elements and acquaint ourselves withthe stuff of language. The very simplest element of speech--and by"speech" we shall hence-forth mean the auditory system of speechsymbolism, the flow of spoken words--is the individual sound, though, aswe shall see later on, the sound is not itself a simple structure butthe resultant of a series of independent, yet closely correlated, adjustments in the organs of speech. And yet the individual sound isnot, properly considered, an element of speech at all, for speech is asignificant function and the sound as such has no significance. Ithappens occasionally that the single sound is an independentlysignificant element (such as French _a_ "has" and _à_ "to" or Latin _i_"go!"), but such cases are fortuitous coincidences between individualsound and significant word. The coincidence is apt to be fortuitous notonly in theory but in point of actual historic fact; thus, the instancescited are merely reduced forms of originally fuller phoneticgroups--Latin _habet_ and _ad_ and Indo-European _ei_ respectively. Iflanguage is a structure and if the significant elements of language arethe bricks of the structure, then the sounds of speech can only becompared to the unformed and unburnt clay of which the bricks arefashioned. In this chapter we shall have nothing further to do withsounds as sounds. 

The true, significant elements of language are generally sequences ofsounds that are either words, significant parts of words, or wordgroupings. What distinguishes each of these elements is that it is theoutward sign of a specific idea, whether of a single concept or image orof a number of such concepts or images definitely connected into awhole. The single word may or may not be the simplest significantelement we have to deal with. The English words _sing_, _sings_, _singing_, _singer_ each conveys a perfectly definite and intelligibleidea, though the idea is disconnected and is therefore functionally ofno practical value. We recognize immediately that these words are of twosorts. The first word, _sing_, is an indivisible phonetic entityconveying the notion of a certain specific activity. The other words allinvolve the same fundamental notion but, owing to the addition of otherphonetic elements, this notion is given a particular twist that modifiesor more closely defines it. They represent, in a sense, compoundedconcepts that have flowered from the fundamental one. We may, therefore, analyze the words _sings_, _singing_, and _singer_ as binary expressionsinvolving a fundamental concept, a concept of subject matter (_sing_), and a further concept of more abstract order--one of person, number, time, condition, function, or of several of these combined. 

If we symbolize such a term as _sing_ by the algebraic formula A, weshall have to symbolize such terms as _sings_ and _singer_ by theformula A + b. [1] The element A may be either a complete and independentword (_sing_) or the fundamental substance, the so-called root orstem[2] or "radical element" (_sing-_) of a word. The element b (_-s_, _-ing_, _-er_) is the indicator of a subsidiary and, as a rule, a moreabstract concept; in the widest sense of the word "form, " it puts uponthe fundamental concept a formal limitation. We may term it a"grammatical element" or affix. As we shall see later on, thegrammatical element or the grammatical increment, as we had better putit, need not be suffixed to the radical element. It may be a prefixedelement (like the _un-_ of _unsingable_), it may be inserted into thevery body of the stem (like the _n_ of the Latin _vinco_ "I conquer" ascontrasted with its absence in _vici_ "I have conquered"), it may be thecomplete or partial repetition of the stem, or it may consist of somemodification of the inner form of the stem (change of vowel, as in_sung_ and _song_; change of consonant as in _dead_ and _death_; changeof accent; actual abbreviation). Each and every one of these types ofgrammatical element or modification has this peculiarity, that it maynot, in the vast majority of cases, be used independently but needs tobe somehow attached to or welded with a radical element in order toconvey an intelligible notion. We had better, therefore, modify ourformula, A + b, to A + (b), the round brackets symbolizing theincapacity of an element to stand alone. The grammatical element, moreover, is not only non-existent except as associated with a radicalone, it does not even, as a rule, obtain its measure of significanceunless it is associated with a particular class of radical elements. Thus, the _-s_ of English _he hits_ symbolizes an utterly differentnotion from the _-s_ of _books_, merely because _hit_ and _book_ aredifferently classified as to function. We must hasten to observe, however, that while the radical element may, on occasion, be identicalwith the word, it does not follow that it may always, or evencustomarily, be used as a word. Thus, the _hort-_ "garden" of such Latinforms as _hortus_, _horti_, and _horto_ is as much of an abstraction, though one yielding a more easily apprehended significance, than the_-ing_ of _singing_. Neither exists as an independently intelligible andsatisfying element of speech. Both the radical element, as such, and thegrammatical element, therefore, are reached only by a process ofabstraction. It seemed proper to symbolize _sing-er_ as A + (b);_hort-us_ must be symbolized as (A) + (b). 

[Footnote 1: We shall reserve capitals for radical elements. ]

[Footnote 2: These words are not here used in a narrowly technicalsense. ]

So far, the first speech element that we have found which we can sayactually "exists" is the word. Before defining the word, however, wemust look a little more closely at the type of word that is illustratedby _sing_. Are we, after all, justified in identifying it with a radicalelement? Does it represent a simple correspondence between concept andlinguistic expression? Is the element _sing-_, that we have abstractedfrom _sings_, _singing_, and _singer_ and to which we may justly ascribea general unmodified conceptual value, actually the same linguistic factas the word _sing_? It would almost seem absurd to doubt it, yet alittle reflection only is needed to convince us that the doubt isentirely legitimate. The word _sing_ cannot, as a matter of fact, befreely used to refer to its own conceptual content. The existence ofsuch evidently related forms as _sang_ and _sung_ at once shows that itcannot refer to past time, but that, for at least an important part ofits range of usage, it is limited to the present. On the other hand, theuse of _sing_ as an "infinitive" (in such locutions as _to sing_ and _hewill sing_) does indicate that there is a fairly strong tendency for theword _sing_ to represent the full, untrammeled amplitude of a specificconcept. Yet if _sing_ were, in any adequate sense, the fixedexpression of the unmodified concept, there should be no room for suchvocalic aberrations as we find in _sang_ and _sung_ and _song_, norshould we find _sing_ specifically used to indicate present time for allpersons but one (third person singular _sings_). 

The truth of the matter is that _sing_ is a kind of twilight word, trembling between the status of a true radical element and that of amodified word of the type of _singing_. Though it has no outward sign toindicate that it conveys more than a generalized idea, we do feel thatthere hangs about it a variable mist of added value. The formula A doesnot seem to represent it so well as A + (0). We might suspect _sing_ ofbelonging to the A + (b) type, with the reservation that the (b) hadvanished. This report of the "feel" of the word is far from fanciful, for historical evidence does, in all earnest, show that _sing_ is inorigin a number of quite distinct words, of type A + (b), that havepooled their separate values. The (b) of each of these has gone as atangible phonetic element; its force, however, lingers on in weakenedmeasure. The _sing_ of _I sing_ is the correspondent of the Anglo-Saxon_singe_; the infinitive _sing_, of _singan_; the imperative _sing_ of_sing_. Ever since the breakdown of English forms that set in about thetime of the Norman Conquest, our language has been straining towards thecreation of simple concept-words, unalloyed by formal connotations, butit has not yet succeeded in this, apart, possibly, from isolated adverbsand other elements of that sort. Were the typical unanalyzable word ofthe language truly a pure concept-word (type A) instead of being of astrangely transitional type (type A + [0]), our _sing_ and _work_ and_house_ and thousands of others would compare with the genuineradical-words of numerous other languages. [3] Such a radical-word, totake a random example, is the Nootka[4] word _hamot_ "bone. " Our Englishcorrespondent is only superficially comparable. _Hamot_ means "bone" ina quite indefinite sense; to our English word clings the notion ofsingularity. The Nootka Indian can convey the idea of plurality, in oneof several ways, if he so desires, but he does not need to; _hamot_ maydo for either singular or plural, should no interest happen to attach tothe distinction. As soon as we say "bone" (aside from its secondaryusage to indicate material), we not merely specify the nature of theobject but we imply, whether we will or no, that there is but one ofthese objects to be considered. And this increment of value makes allthe difference. 

[Footnote 3: It is not a question of the general isolating character ofsuch languages as Chinese (see Chapter VI). Radical-words may and dooccur in languages of all varieties, many of them of a high degree ofcomplexity. ]

[Footnote 4: Spoken by a group of Indian tribes in Vancouver Island. ]

We now know of four distinct formal types of word: A (Nootka _hamot_);A + (0) (_sing_, _bone_); A + (b) (_singing_); (A) + (b) (Latin_hortus_). There is but one other type that is fundamentally possible:A + B, the union of two (or more) independently occurring radicalelements into a single term. Such a word is the compound _fire-engine_or a Sioux form equivalent to _eat-stand_ (i. E. , "to eat whilestanding"). It frequently happens, however, that one of the radicalelements becomes functionally so subordinated to the other that it takeson the character of a grammatical element. We may symbolize this byA + b, a type that may gradually, by loss of external connection betweenthe subordinated element b and its independent counterpart B merge withthe commoner type A + (b). A word like _beautiful_ is an example ofA + b, the _-ful_ barely preserving the impress of its lineage. A wordlike _homely_, on the other hand, is clearly of the type A + (b), for noone but a linguistic student is aware of the connection between the_-ly_ and the independent word _like_. 

In actual use, of course, these five (or six) fundamental types may beindefinitely complicated in a number of ways. The (0) may have amultiple value; in other words, the inherent formal modification of thebasic notion of the word may affect more than one category. In such aLatin word as _cor_ "heart, " for instance, not only is a concreteconcept conveyed, but there cling to the form, which is actually shorterthan its own radical element (_cord-_), the three distinct, yetintertwined, formal concepts of singularity, gender classification(neuter), and case (subjective-objective). The complete grammaticalformula for _cor_ is, then, A + (0) + (0) + (0), though the merelyexternal, phonetic formula would be (A)--, (A) indicating the abstracted"stem" _cord-_, the minus sign a loss of material. The significant thingabout such a word as _cor_ is that the three conceptual limitations arenot merely expressed by implication as the word sinks into place in asentence; they are tied up, for good and all, within the very vitals ofthe word and cannot be eliminated by any possibility of usage. 

Other complications result from a manifolding of parts. In a given wordthere may be several elements of the order A (we have already symbolizedthis by the type A + B), of the order (A), of the order b, and of theorder (b). Finally, the various types may be combined among themselvesin endless ways. A comparatively simple language like English, or evenLatin, illustrates but a modest proportion of these theoreticalpossibilities. But if we take our examples freely from the vaststorehouse of language, from languages exotic as well as from those thatwe are more familiar with, we shall find that there is hardly apossibility that is not realized in actual usage. One example will dofor thousands, one complex type for hundreds of possible types. I selectit from Paiute, the language of the Indians of the arid plateaus ofsouthwestern Utah. The word_wii-to-kuchum-punku-rügani-yugwi-va-ntü-m(ü)_[5] is of unusual lengtheven for its own language, but it is no psychological monster for allthat. It means "they who are going to sit and cut up with a knife ablack cow (_or_ bull), " or, in the order of the Indian elements, "knife-black-buffalo-pet-cut up-sit(plur. )-future-participle-animateplur. " The formula for this word, in accordance with our symbolism, would be (F) + (E) + C + d + A + B + (g) + (h) + (i) + (0). It is theplural of the future participle of a compound verb "to sit and cutup"--A + B. The elements (g)--which denotes futurity--, (h)--aparticipial suffix--, and (i)--indicating the animate plural--aregrammatical elements which convey nothing when detached. The formula (0)is intended to imply that the finished word conveys, in addition to whatis definitely expressed, a further relational idea, that ofsubjectivity; in other words, the form can only be used as the subjectof a sentence, not in an objective or other syntactic relation. Theradical element A ("to cut up"), before entering into combination withthe coördinate element B ("to sit"), is itself compounded with twonominal elements or element-groups--an instrumentally used stem (F)("knife"), which may be freely used as the radical element of nounforms but cannot be employed as an absolute noun in its given form, andan objectively used group--(E) + C + d ("black cow _or_ bull"). Thisgroup in turn consists of an adjectival radical element (E) ("black"), which cannot be independently employed (the absolute notion of "black"can be rendered only as the participle of a verb: "black-be-ing"), andthe compound noun C + d ("buffalo-pet"). The radical element C properlymeans "buffalo, " but the element d, properly an independently occurringnoun meaning "horse" (originally "dog" or "domesticated animal" ingeneral), is regularly used as a quasi-subordinate element indicatingthat the animal denoted by the stem to which it is affixed is owned by ahuman being. It will be observed that the whole complex(F) + (E) + C + d + A + B is functionally no more than a verbal base, corresponding to the _sing-_ of an English form like _singing_; thatthis complex remains verbal in force on the addition of the temporalelement (g)--this (g), by the way, must not be understood as appended toB alone, but to the whole basic complex as a unit--; and that theelements (h) + (i) + (0) transform the verbal expression into a formallywell-defined noun. 

[Footnote 5: In this and other examples taken from exotic languages I amforced by practical considerations to simplify the actual phoneticforms. This should not matter perceptibly, as we are concerned with formas such, not with phonetic content. ]

It is high time that we decided just what is meant by a word. Our firstimpulse, no doubt, would have been to define the word as the symbolic, linguistic counterpart of a single concept. We now know that such adefinition is impossible. In truth it is impossible to define the wordfrom a functional standpoint at all, for the word may be anything fromthe expression of a single concept--concrete or abstract or purelyrelational (as in _of_ or _by_ or _and_)--to the expression of acomplete thought (as in Latin _dico_ "I say" or, with greaterelaborateness of form, in a Nootka verb form denoting "I have beenaccustomed to eat twenty round objects [e. G. , apples] while engaged in[doing so and so]"). In the latter case the word becomes identical withthe sentence. The word is merely a form, a definitely molded entity thattakes in as much or as little of the conceptual material of the wholethought as the genius of the language cares to allow. Thus it is thatwhile the single radical elements and grammatical elements, the carriersof isolated concepts, are comparable as we pass from language tolanguage, the finished words are not. Radical (or grammatical) elementand sentence--these are the primary _functional_ units of speech, theformer as an abstracted minimum, the latter as the estheticallysatisfying embodiment of a unified thought. The actual _formal_ units ofspeech, the words, may on occasion identify themselves with either ofthe two functional units; more often they mediate between the twoextremes, embodying one or more radical notions and also one or moresubsidiary ones. We may put the whole matter in a nutshell by sayingthat the radical and grammatical elements of language, abstracted asthey are from the realities of speech, respond to the conceptual worldof science, abstracted as it is from the realities of experience, andthat the word, the existent unit of living speech, responds to the unitof actually apprehended experience, of history, of art. The sentence isthe logical counterpart of the complete thought only if it be felt asmade up of the radical and grammatical elements that lurk in therecesses of its words. It is the psychological counterpart ofexperience, of art, when it is felt, as indeed it normally is, as thefinished play of word with word. As the necessity of defining thoughtsolely and exclusively for its own sake becomes more urgent, the wordbecomes increasingly irrelevant as a means. We can therefore easilyunderstand why the mathematician and the symbolic logician are driven todiscard the word and to build up their thought with the help of symbolswhich have, each of them, a rigidly unitary value. 

But is not the word, one may object, as much of an abstraction as theradical element? Is it not as arbitrarily lifted out of the livingsentence as is the minimum conceptual element out of the word? Somestudents of language have, indeed, looked upon the word as such anabstraction, though with very doubtful warrant, it seems to me. It istrue that in particular cases, especially in some of the highlysynthetic languages of aboriginal America, it is not always easy to saywhether a particular element of language is to be interpreted as anindependent word or as part of a larger word. These transitional cases, puzzling as they may be on occasion, do not, however, materially weakenthe case for the psychological validity of the word. Linguisticexperience, both as expressed in standardized, written form and astested in daily usage, indicates overwhelmingly that there is not, as arule, the slightest difficulty in bringing the word to consciousness asa psychological reality. No more convincing test could be desired thanthis, that the naive Indian, quite unaccustomed to the concept of thewritten word, has nevertheless no serious difficulty in dictating a textto a linguistic student word by word; he tends, of course, to run hiswords together as in actual speech, but if he is called to a halt and ismade to understand what is desired, he can readily isolate the words assuch, repeating them as units. He regularly refuses, on the other hand, to isolate the radical or grammatical element, on the ground that it"makes no sense. "[6] What, then, is the objective criterion of the word?The speaker and hearer feel the word, let us grant, but how shall wejustify their feeling? If function is not the ultimate criterion of theword, what is?

[Footnote 6: These oral experiences, which I have had time and again asa field student of American Indian languages, are very neatly confirmedby personal experiences of another sort. Twice I have taught intelligentyoung Indians to write their own languages according to the phoneticsystem which I employ. They were taught merely how to render accuratelythe sounds as such. Both had some difficulty in learning to break up aword into its constituent sounds, but none whatever in determining thewords. This they both did with spontaneous and complete accuracy. In thehundreds of pages of manuscript Nootka text that I have obtained fromone of these young Indians the words, whether abstract relationalentities like English _that_ and _but_ or complex sentence-words likethe Nootka example quoted above, are, practically without exception, isolated precisely as I or any other student would have isolated them. Such experiences with naïve speakers and recorders do more to convinceone of the definitely plastic unity of the word than any amount ofpurely theoretical argument. ]

It is easier to ask the question than to answer it. The best that we cando is to say that the word is one of the smallest, completely satisfyingbits of isolated "meaning" into which the sentence resolves itself. Itcannot be cut into without a disturbance of meaning, one or the other orboth of the severed parts remaining as a helpless waif on our hands. Inpractice this unpretentious criterion does better service than might besupposed. In such a sentence as _It is unthinkable_, it is simplyimpossible to group the elements into any other and smaller "words" thanthe three indicated. _Think_ or _thinkable_ might be isolated, but asneither _un-_ nor _-able_ nor _is-un_ yields a measurable satisfaction, we are compelled to leave _unthinkable_ as an integral whole, aminiature bit of art. Added to the "feel" of the word are frequently, but by no means invariably, certain external phonetic characteristics. Chief of these is accent. In many, perhaps in most, languages the singleword is marked by a unifying accent, an emphasis on one of thesyllables, to which the rest are subordinated. The particular syllablethat is to be so distinguished is dependent, needless to say, on thespecial genius of the language. The importance of accent as a unifyingfeature of the word is obvious in such English examples as_unthinkable_, _characterizing_. The long Paiute word that we haveanalyzed is marked as a rigid phonetic unit by several features, chiefof which are the accent on its second syllable (_wii'_-"knife") and theslurring ("unvoicing, " to use the technical phonetic term) of its finalvowel (_-mü_, animate plural). Such features as accent, cadence, and thetreatment of consonants and vowels within the body of a word are oftenuseful as aids in the external demarcation of the word, but they must byno means be interpreted, as is sometimes done, as themselves responsiblefor its psychological existence. They at best but strengthen a feelingof unity that is already present on other grounds. 

We have already seen that the major functional unit of speech, thesentence, has, like the word, a psychological as well as a merelylogical or abstracted existence. Its definition is not difficult. It isthe linguistic expression of a proposition. It combines a subject ofdiscourse with a statement in regard to this subject. Subject and"predicate" may be combined in a single word, as in Latin _dico_; eachmay be expressed independently, as in the English equivalent, _I say_;each or either may be so qualified as to lead to complex propositions ofmany sorts. No matter how many of these qualifying elements (words orfunctional parts of words) are introduced, the sentence does not loseits feeling of unity so long as each and every one of them falls inplace as contributory to the definition of either the subject ofdiscourse or the core of the predicate[7]. Such a sentence as _The mayorof New York is going to deliver a speech of welcome in French_ isreadily felt as a unified statement, incapable of reduction by thetransfer of certain of its elements, in their given form, to thepreceding or following sentences. The contributory ideas of _of NewYork_, _of welcome_, and _in French_ may be eliminated without hurtingthe idiomatic flow of the sentence. _The mayor is going to deliver aspeech_ is a perfectly intelligible proposition. But further than thiswe cannot go in the process of reduction. We cannot say, for instance, _Mayor is going to deliver_. [8] The reduced sentence resolves itselfinto the subject of discourse--_the mayor_--and the predicate--_is goingto deliver a speech_. It is customary to say that the true subject ofsuch a sentence is _mayor_, the true predicate _is going_ or even _is_, the other elements being strictly subordinate. Such an analysis, however, is purely schematic and is without psychological value. It ismuch better frankly to recognize the fact that either or both of the twoterms of the sentence-proposition may be incapable of expression in theform of single words. There are languages that can convey all that isconveyed by _The-mayor is-going-to-deliver-a-speech_ in two words, asubject word and a predicate word, but English is not so highlysynthetic. The point that we are really making here is that underlyingthe finished sentence is a living sentence type, of fixed formalcharacteristics. These fixed types or actual sentence-groundworks may befreely overlaid by such additional matter as the speaker or writer caresto put on, but they are themselves as rigidly "given" by tradition asare the radical and grammatical elements abstracted from the finishedword. New words may be consciously created from these fundamentalelements on the analogy of old ones, but hardly new types of words. Inthe same way new sentences are being constantly created, but always onstrictly traditional lines. The enlarged sentence, however, allows as arule of considerable freedom in the handling of what may be called"unessential" parts. It is this margin of freedom which gives us theopportunity of individual style. 

[Footnote 7: "Coordinate sentences" like _I shall remain but you may go_may only doubtfully be considered as truly unified predications, as truesentences. They are sentences in a stylistic sense rather than from thestrictly formal linguistic standpoint. The orthography _I shall remain. But you may go_ is as intrinsically justified as _I shall remain. Nowyou may go_. The closer connection in sentiment between the first twopropositions has led to a conventional visual representation that mustnot deceive the analytic spirit. ]

[Footnote 8: Except, possibly, in a newspaper headline. Such headlines, however, are language only in a derived sense. ]

The habitual association of radical elements, grammatical elements, words, and sentences with concepts or groups of concepts related intowholes is the fact itself of language. It is important to note thatthere is in all languages a certain randomness of association. Thus, theidea of "hide" may be also expressed by the word "conceal, " the notionof "three times" also by "thrice. " The multiple expression of a singleconcept is universally felt as a source of linguistic strength andvariety, not as a needless extravagance. More irksome is a randomcorrespondence between idea and linguistic expression in the field ofabstract and relational concepts, particularly when the concept isembodied in a grammatical element. Thus, the randomness of theexpression of plurality in such words as _books_, _oxen_, _sheep_, and_geese_ is felt to be rather more, I fancy, an unavoidable andtraditional predicament than a welcome luxuriance. It is obvious that alanguage cannot go beyond a certain point in this randomness. Manylanguages go incredibly far in this respect, it is true, but linguistichistory shows conclusively that sooner or later the less frequentlyoccurring associations are ironed out at the expense of the more vitalones. In other words, all languages have an inherent tendency to economyof expression. Were this tendency entirely inoperative, there would beno grammar. The fact of grammar, a universal trait of language, issimply a generalized expression of the feeling that analogous conceptsand relations are most conveniently symbolized in analogous forms. Werea language ever completely "grammatical, " it would be a perfect engineof conceptual expression. Unfortunately, or luckily, no language istyrannically consistent. All grammars leak. 

Up to the present we have been assuming that the material of languagereflects merely the world of concepts and, on what I have ventured tocall the "pre-rational" plane, of images, which are the raw material ofconcepts. We have, in other words, been assuming that language movesentirely in the ideational or cognitive sphere. It is time that weamplified the picture. The volitional aspect of consciousness also is tosome extent explicitly provided for in language. Nearly all languageshave special means for the expression of commands (in the imperativeforms of the verb, for example) and of desires, unattained orunattainable (_Would he might come!_ or _Would he were here!_) Theemotions, on the whole, seem to be given a less adequate outlet. Emotion, indeed, is proverbially inclined to speechlessness. Most, ifnot all, the interjections are to be put to the credit of emotionalexpression, also, it may be, a number of linguistic elements expressingcertain modalities, such as dubitative or potential forms, which may beinterpreted as reflecting the emotional states of hesitation ordoubt--attenuated fear. On the whole, it must be admitted that ideationreigns supreme in language, that volition and emotion come in asdistinctly secondary factors. This, after all, is perfectlyintelligible. The world of image and concept, the endless andever-shifting picture of objective reality, is the unavoidablesubject-matter of human communication, for it is only, or mainly, interms of this world that effective action is possible. Desire, purpose, emotion are the personal color of the objective world; they are appliedprivately by the individual soul and are of relatively little importanceto the neighboring one. All this does not mean that volition and emotionare not expressed. They are, strictly speaking, never absent from normalspeech, but their expression is not of a truly linguistic nature. Thenuances of emphasis, tone, and phrasing, the varying speed andcontinuity of utterance, the accompanying bodily movements, all theseexpress something of the inner life of impulse and feeling, but as thesemeans of expression are, at last analysis, but modified forms of theinstinctive utterance that man shares with the lower animals, theycannot be considered as forming part of the essential culturalconception of language, however much they may be inseparable from itsactual life. And this instinctive expression of volition and emotion is, for the most part, sufficient, often more than sufficient, for thepurposes of communication. 

There are, it is true, certain writers on the psychology of language[9]who deny its prevailingly cognitive character but attempt, on thecontrary, to demonstrate the origin of most linguistic elements withinthe domain of feeling. I confess that I am utterly unable to followthem. What there is of truth in their contentions may be summed up, itseems to me, by saying that most words, like practically all elements ofconsciousness, have an associated feeling-tone, a mild, yet none theless real and at times insidiously powerful, derivative of pleasure orpain. This feeling-tone, however, is not as a rule an inherent value inthe word itself; it is rather a sentimental growth on the word's truebody, on its conceptual kernel. Not only may the feeling-tone changefrom one age to another (this, of course, is true of the conceptualcontent as well), but it varies remarkably from individual to individualaccording to the personal associations of each, varies, indeed, fromtime to time in a single individual's consciousness as his experiencesmold him and his moods change. To be sure, there are socially acceptedfeeling-tones, or ranges of feeling-tone, for many words over and abovethe force of individual association, but they are exceedingly variableand elusive things at best. They rarely have the rigidity of thecentral, primary fact. We all grant, for instance, that _storm_, _tempest_, and _hurricane_, quite aside from their slight differences ofactual meaning, have distinct feeling-tones, tones that are felt by allsensitive speakers and readers of English in a roughly equivalentfashion. _Storm_, we feel, is a more general and a decidedly less"magnificent" word than the other two; _tempest_ is not only associatedwith the sea but is likely, in the minds of many, to have obtained asoftened glamour from a specific association with Shakespeare's greatplay; _hurricane_ has a greater forthrightness, a directer ruthlessnessthan its synonyms. Yet the individual's feeling-tones for these wordsare likely to vary enormously. To some _tempest_ and _hurricane_ mayseem "soft, " literary words, the simpler _storm_ having a fresh, ruggedvalue which the others do not possess (think of _storm and stress_). Ifwe have browsed much in our childhood days in books of the Spanish Main, _hurricane_ is likely to have a pleasurably bracing tone; if we have hadthe misfortune to be caught in one, we are not unlikely to feel the wordas cold, cheerless, sinister. 

[Footnote 9: E. G. , the brilliant Dutch writer, Jac van Ginneken. ]

The feeling-tones of words are of no use, strictly speaking, to science;the philosopher, if he desires to arrive at truth rather than merely topersuade, finds them his most insidious enemies. But man is rarelyengaged in pure science, in solid thinking. Generally his mentalactivities are bathed in a warm current of feeling and he seizes uponthe feeling-tones of words as gentle aids to the desired excitation. They are naturally of great value to the literary artist. It isinteresting to note, however, that even to the artist they are a danger. A word whose customary feeling-tone is too unquestioningly acceptedbecomes a plushy bit of furniture, a _cliché_. Every now and then theartist has to fight the feeling-tone, to get the word to mean what itnakedly and conceptually should mean, depending for the effect offeeling on the creative power of an individual juxtaposition of conceptsor images. 

III

THE SOUNDS OF LANGUAGE

We have seen that the mere phonetic framework of speech does notconstitute the inner fact of language and that the single sound ofarticulated speech is not, as such, a linguistic element at all. For allthat, speech is so inevitably bound up with sounds and theirarticulation that we can hardly avoid giving the subject of phoneticssome general consideration. Experience has shown that neither the purelyformal aspects of a language nor the course of its history can be fullyunderstood without reference to the sounds in which this form and thishistory are embodied. A detailed survey of phonetics would be both tootechnical for the general reader and too loosely related to our maintheme to warrant the needed space, but we can well afford to consider afew outstanding facts and ideas connected with the sounds of language. 

The feeling that the average speaker has of his language is that it isbuilt up, acoustically speaking, of a comparatively small number ofdistinct sounds, each of which is rather accurately provided for in thecurrent alphabet by one letter or, in a few cases, by two or morealternative letters. As for the languages of foreigners, he generallyfeels that, aside from a few striking differences that cannot escapeeven the uncritical ear, the sounds they use are the same as those he isfamiliar with but that there is a mysterious "accent" to these foreignlanguages, a certain unanalyzed phonetic character, apart from thesounds as such, that gives them their air of strangeness. This naïvefeeling is largely illusory on both scores. Phonetic analysis convincesone that the number of clearly distinguishable sounds and nuances ofsounds that are habitually employed by the speakers of a language is fargreater than they themselves recognize. Probably not one English speakerout of a hundred has the remotest idea that the _t_ of a word like_sting_ is not at all the same sound as the _t_ of _teem_, the latter_t_ having a fullness of "breath release" that is inhibited in theformer case by the preceding _s_; that the _ea_ of _meat_ is ofperceptibly shorter duration than the _ea_ of _mead_; or that the final_s_ of a word like _heads_ is not the full, buzzing _z_ sound of the _s_in such a word as _please_. It is the frequent failure of foreigners, who have acquired a practical mastery of English and who have eliminatedall the cruder phonetic shortcomings of their less careful brethren, toobserve such minor distinctions that helps to give their Englishpronunciation the curiously elusive "accent" that we all vaguely feel. We do not diagnose the "accent" as the total acoustic effect produced bya series of slight but specific phonetic errors for the very good reasonthat we have never made clear to ourselves our own phonetic stock intrade. If two languages taken at random, say English and Russian, arecompared as to their phonetic systems, we are more apt than not to findthat very few of the phonetic elements of the one find an exact analoguein the other. Thus, the _t_ of a Russian word like _tam_ "there" isneither the English _t_ of _sting_ nor the English _t_ of _teem_. Itdiffers from both in its "dental" articulation, in other words, in beingproduced by contact of the tip of the tongue with the upper teeth, not, as in English, by contact of the tongue back of the tip with the gumridge above the teeth; moreover, it differs from the _t_ of _teem_ alsoin the absence of a marked "breath release" before the following vowelis attached, so that its acoustic effect is of a more precise, "metallic" nature than in English. Again, the English _l_ is unknown inRussian, which possesses, on the other hand, two distinct _l_-soundsthat the normal English speaker would find it difficult exactly toreproduce--a "hollow, " guttural-like _l_ and a "soft, " palatalized_l_-sound that is only very approximately rendered, in English terms, as_ly_. Even so simple and, one would imagine, so invariable a sound as_m_ differs in the two languages. In a Russian word like _most_ "bridge"the _m_ is not the same as the _m_ of the English word _most_; the lipsare more fully rounded during its articulation, so that it makes aheavier, more resonant impression on the ear. The vowels, needless tosay, differ completely in English and Russian, hardly any two of thembeing quite the same. 

I have gone into these illustrative details, which are of little or nospecific interest for us, merely in order to provide something of anexperimental basis to convince ourselves of the tremendous variabilityof speech sounds. Yet a complete inventory of the acoustic resources ofall the European languages, the languages nearer home, whileunexpectedly large, would still fall far short of conveying a just ideaof the true range of human articulation. In many of the languages ofAsia, Africa, and aboriginal America there are whole classes of soundsthat most of us have no knowledge of. They are not necessarily moredifficult of enunciation than sounds more familiar to our ears; theymerely involve such muscular adjustments of the organs of speech as wehave never habituated ourselves to. It may be safely said that the totalnumber of possible sounds is greatly in excess of those actually inuse. Indeed, an experienced phonetician should have no difficulty ininventing sounds that are unknown to objective investigation. One reasonwhy we find it difficult to believe that the range of possible speechsounds is indefinitely large is our habit of conceiving the sound as asimple, unanalyzable impression instead of as the resultant of a numberof distinct muscular adjustments that take place simultaneously. Aslight change in any one of these adjustments gives us a new sound whichis akin to the old one, because of the continuance of the otheradjustments, but which is acoustically distinct from it, so sensitivehas the human ear become to the nuanced play of the vocal mechanism. Another reason for our lack of phonetic imagination is the fact that, while our ear is delicately responsive to the sounds of speech, themuscles of our speech organs have early in life become exclusivelyaccustomed to the particular adjustments and systems of adjustment thatare required to produce the traditional sounds of the language. All ornearly all other adjustments have become permanently inhibited, whetherthrough inexperience or through gradual elimination. Of course the powerto produce these inhibited adjustments is not entirely lost, but theextreme difficulty we experience in learning the new sounds of foreignlanguages is sufficient evidence of the strange rigidity that has set infor most people in the voluntary control of the speech organs. The pointmay be brought home by contrasting the comparative lack of freedom ofvoluntary speech movements with the all but perfect freedom of voluntarygesture. [10] Our rigidity in articulation is the price we have had topay for easy mastery of a highly necessary symbolism. One cannot be bothsplendidly free in the random choice of movements and selective withdeadly certainty. [11]

[Footnote 10: Observe the "voluntary. " When we shout or grunt orotherwise allow our voices to take care of themselves, as we are likelyto do when alone in the country on a fine spring day, we are no longerfixing vocal adjustments by voluntary control. Under these circumstanceswe are almost certain to hit on speech sounds that we could never learnto control in actual speech. ]

[Footnote 11: If speech, in its acoustic and articulatory aspect, isindeed a rigid system, how comes it, one may plausibly object, that notwo people speak alike? The answer is simple. All that part of speechwhich falls out of the rigid articulatory framework is not speech inidea, but is merely a superadded, more or less instinctively determinedvocal complication inseparable from speech in practice. All theindividual color of speech--personal emphasis, speed, personal cadence, personal pitch--is a non-linguistic fact, just as the incidentalexpression of desire and emotion are, for the most part, alien tolinguistic expression. Speech, like all elements of culture, demandsconceptual selection, inhibition of the randomness of instinctivebehavior. That its "idea" is never realized as such in practice, itscarriers being instinctively animated organisms, is of course true ofeach and every aspect of culture. ]

There are, then, an indefinitely large number of articulated soundsavailable for the mechanics of speech; any given language makes use ofan explicit, rigidly economical selection of these rich resources; andeach of the many possible sounds of speech is conditioned by a number ofindependent muscular adjustments that work together simultaneouslytowards its production. A full account of the activity of each of theorgans of speech--in so far as its activity has a bearing onlanguage--is impossible here, nor can we concern ourselves in asystematic way with the classification of sounds on the basis of theirmechanics. [12] A few bold outlines are all that we can attempt. Theorgans of speech are the lungs and bronchial tubes; the throat, particularly that part of it which is known as the larynx or, in popularparlance, the "Adam's apple"; the nose; the uvula, which is the soft, pointed, and easily movable organ that depends from the rear of thepalate; the palate, which is divided into a posterior, movable "softpalate" or velum and a "hard palate"; the tongue; the teeth; and thelips. The palate, lower palate, tongue, teeth, and lips may be lookedupon as a combined resonance chamber, whose constantly varying shape, chiefly due to the extreme mobility of the tongue, is the main factor ingiving the outgoing breath its precise quality[13] of sound. 

[Footnote 12: Purely acoustic classifications, such as more easilysuggest themselves to a first attempt at analysis, are now in less favoramong students of phonetics than organic classifications. The latterhave the advantage of being more objective. Moreover, the acousticquality of a sound is dependent on the articulation, even though inlinguistic consciousness this quality is the primary, not the secondary, fact. ]

[Footnote 13: By "quality" is here meant the inherent nature andresonance of the sound as such. The general "quality" of theindividual's voice is another matter altogether. This is chieflydetermined by the individual anatomical characteristics of the larynxand is of no linguistic interest whatever. ]

The lungs and bronchial tubes are organs of speech only in so far asthey supply and conduct the current of outgoing air without whichaudible articulation is impossible. They are not responsible for anyspecific sound or acoustic feature of sounds except, possibly, accent orstress. It may be that differences of stress are due to slightdifferences in the contracting force of the lung muscles, but even thisinfluence of the lungs is denied by some students, who explain thefluctuations of stress that do so much to color speech by reference tothe more delicate activity of the glottal cords. These glottal cords aretwo small, nearly horizontal, and highly sensitive membranes within thelarynx, which consists, for the most part, of two large and severalsmaller cartilages and of a number of small muscles that control theaction of the cords. 

The cords, which are attached to the cartilages, are to the human speechorgans what the two vibrating reeds are to a clarinet or the strings toa violin. They are capable of at least three distinct types of movement, each of which is of the greatest importance for speech. They may bedrawn towards or away from each other, they may vibrate like reeds orstrings, and they may become lax or tense in the direction of theirlength. The last class of these movements allows the cords to vibrate atdifferent "lengths" or degrees of tenseness and is responsible for thevariations in pitch which are present not only in song but in the moreelusive modulations of ordinary speech. The two other types of glottalaction determine the nature of the voice, "voice" being a convenientterm for breath as utilized in speech. If the cords are well apart, allowing the breath to escape in unmodified form, we have the conditiontechnically known as "voicelessness. " All sounds produced under thesecircumstances are "voiceless" sounds. Such are the simple, unmodifiedbreath as it passes into the mouth, which is, at least approximately, the same as the sound that we write _h_, also a large number of specialarticulations in the mouth chamber, like _p_ and _s_. On the other hand, the glottal cords may be brought tight together, without vibrating. Whenthis happens, the current of breath is checked for the time being. Theslight choke or "arrested cough" that is thus made audible is notrecognized in English as a definite sound but occurs nevertheless notinfrequently. [14] This momentary check, technically known as a "glottalstop, " is an integral element of speech in many languages, as Danish, Lettish, certain Chinese dialects, and nearly all American Indianlanguages. Between the two extremes of voicelessness, that ofcompletely open breath and that of checked breath, lies the position oftrue voice. In this position the cords are close together, but not sotightly as to prevent the air from streaming through; the cords are setvibrating and a musical tone of varying pitch results. A tone soproduced is known as a "voiced sound. " It may have an indefinite numberof qualities according to the precise position of the upper organs ofspeech. Our vowels, nasals (such as _m_ and _n_), and such sounds as_b_, _z_, and _l_ are all voiced sounds. The most convenient test of avoiced sound is the possibility of pronouncing it on any given pitch, inother words, of singing on it. [15] The voiced sounds are the mostclearly audible elements of speech. As such they are the carriers ofpractically all significant differences in stress, pitch, andsyllabification. The voiceless sounds are articulated noises that breakup the stream of voice with fleeting moments of silence. Acousticallyintermediate between the freely unvoiced and the voiced sounds are anumber of other characteristic types of voicing, such as murmuring andwhisper. [16] These and still other types of voice are relativelyunimportant in English and most other European languages, but there arelanguages in which they rise to some prominence in the normal flow ofspeech. 

[Footnote 14: As at the end of the snappily pronounced _no!_ (sometimeswritten _nope!_) or in the over-carefully pronounced _at all_, where onemay hear a slight check between the _t_ and the _a_. ]

[Footnote 15: "Singing" is here used in a wide sense. One cannot singcontinuously on such a sound as _b_ or _d_, but one may easily outline atune on a series of _b_'s or _d_'s in the manner of the plucked"pizzicato" on stringed instruments. A series of tones executed oncontinuant consonants, like _m_, _z_, or _l_, gives the effect ofhumming, droning, or buzzing. The sound of "humming, " indeed, is nothingbut a continuous voiced nasal, held on one pitch or varying in pitch, asdesired. ]

[Footnote 16: The whisper of ordinary speech is a combination ofunvoiced sounds and "whispered" sounds, as the term is understood inphonetics. ]

The nose is not an active organ of speech, but it is highly important asa resonance chamber. It may be disconnected from the mouth, which isthe other great resonance chamber, by the lifting of the movable part ofthe soft palate so as to shut off the passage of the breath into thenasal cavity; or, if the soft palate is allowed to hang down freely andunobstructively, so that the breath passes into both the nose and themouth, these make a combined resonance chamber. Such sounds as _b_ and_a_ (as in _father_) are voiced "oral" sounds, that is, the voicedbreath does not receive a nasal resonance. As soon as the soft palate islowered, however, and the nose added as a participating resonancechamber, the sounds _b_ and _a_ take on a peculiar "nasal" quality andbecome, respectively, _m_ and the nasalized vowel written _an_ in French(e. G. , _sang_, _tant_). The only English sounds[17] that normallyreceive a nasal resonance are _m_, _n_, and the _ng_ sound of _sing_. Practically all sounds, however, may be nasalized, not only thevowels--nasalized vowels are common in all parts of the world--but suchsounds as _l_ or _z_. Voiceless nasals are perfectly possible. Theyoccur, for instance, in Welsh and in quite a number of American Indianlanguages. 

[Footnote 17: Aside from the involuntary nasalizing of all voiced soundsin the speech of those that talk with a "nasal twang. "]

The organs that make up the oral resonance chamber may articulate in twoways. The breath, voiced or unvoiced, nasalized or unnasalized, may beallowed to pass through the mouth without being checked or impeded atany point; or it may be either momentarily checked or allowed to streamthrough a greatly narrowed passage with resulting air friction. Thereare also transitions between the two latter types of articulation. Theunimpeded breath takes on a particular color or quality in accordancewith the varying shape of the oral resonance chamber. This shape ischiefly determined by the position of the movable parts--the tongue andthe lips. As the tongue is raised or lowered, retracted or broughtforward, held tense or lax, and as the lips are pursed ("rounded") invarying degree or allowed to keep their position of rest, a large numberof distinct qualities result. These oral qualities are the vowels. Intheory their number is infinite, in practice the ear can differentiateonly a limited, yet a surprisingly large, number of resonance positions. Vowels, whether nasalized or not, are normally voiced sounds; in not afew languages, however, "voiceless vowels"[18] also occur. 

[Footnote 18: These may be also defined as free unvoiced breath withvarying vocalic timbres. In the long Paiute word quoted on page 31 thefirst _u_ and the final _ü_ are pronounced without voice. ]

[Transcriber's note: Footnote 18 refers to line 1014. ]

The remaining oral sounds are generally grouped together as"consonants. " In them the stream of breath is interfered with in someway, so that a lesser resonance results, and a sharper, more incisivequality of tone. There are four main types of articulation generallyrecognized within the consonantal group of sounds. The breath may becompletely stopped for a moment at some definite point in the oralcavity. Sounds so produced, like _t_ or _d_ or _p_, are known as "stops"or "explosives. "[19] Or the breath may be continuously obstructedthrough a narrow passage, not entirely checked. Examples of such"spirants" or "fricatives, " as they are called, are _s_ and _z_ and _y_. The third class of consonants, the "laterals, " are semi-stopped. Thereis a true stoppage at the central point of articulation, but the breathis allowed to escape through the two side passages or through one ofthem. Our English _d_, for instance, may be readily transformed into_l_, which has the voicing and the position of _d_, merely bydepressing the sides of the tongue on either side of the point ofcontact sufficiently to allow the breath to come through. Laterals arepossible in many distinct positions. They may be unvoiced (the Welsh_ll_ is an example) as well as voiced. Finally, the stoppage of thebreath may be rapidly intermittent; in other words, the active organ ofcontact--generally the point of the tongue, less often theuvula[20]--may be made to vibrate against or near the point of contact. These sounds are the "trills" or "rolled consonants, " of which thenormal English _r_ is a none too typical example. They are welldeveloped in many languages, however, generally in voiced form, sometimes, as in Welsh and Paiute, in unvoiced form as well. 

[Footnote 19: Nasalized stops, say _m_ or _n_, can naturally not betruly "stopped, " as there is no way of checking the stream of breath inthe nose by a definite articulation. ]

[Footnote 20: The lips also may theoretically so articulate. "Labialtrills, " however, are certainly rare in natural speech. ]

The oral manner of articulation is naturally not sufficient to define aconsonant. The place of articulation must also be considered. Contactsmay be formed at a large number of points, from the root of the tongueto the lips. It is not necessary here to go at length into this somewhatcomplicated matter. The contact is either between the root of the tongueand the throat, [21] some part of the tongue and a point on the palate(as in _k_ or _ch_ or _l_), some part of the tongue and the teeth (as inthe English _th_ of _thick_ and _then_), the teeth and one of the lips(practically always the upper teeth and lower lip, as in _f_), or thetwo lips (as in _p_ or English _w_). The tongue articulations are themost complicated of all, as the mobility of the tongue allows variouspoints on its surface, say the tip, to articulate against a number ofopposed points of contact. Hence arise many positions of articulationthat we are not familiar with, such as the typical "dental" position ofRussian or Italian _t_ and _d_; or the "cerebral" position of Sanskritand other languages of India, in which the tip of the tongue articulatesagainst the hard palate. As there is no break at any point between therims of the teeth back to the uvula nor from the tip of the tongue backto its root, it is evident that all the articulations that involve thetongue form a continuous organic (and acoustic) series. The positionsgrade into each other, but each language selects a limited number ofclearly defined positions as characteristic of its consonantal system, ignoring transitional or extreme positions. Frequently a language allowsa certain latitude in the fixing of the required position. This is true, for instance, of the English _k_ sound, which is articulated muchfurther to the front in a word like _kin_ than in _cool_. We ignore thisdifference, psychologically, as a non-essential, mechanical one. Anotherlanguage might well recognize the difference, or only a slightly greaterone, as significant, as paralleling the distinction in position betweenthe _k_ of _kin_ and the _t_ of _tin_. 

[Footnote 21: This position, known as "faucal, " is not common. ]

The organic classification of speech sounds is a simple matter afterwhat we have learned of their production. Any such sound may be put intoits proper place by the appropriate answer to four main questions:--Whatis the position of the glottal cords during its articulation? Does thebreath pass into the mouth alone or is it also allowed to stream intothe nose? Does the breath pass freely through the mouth or is it impededat some point and, if so, in what manner? What are the precise points ofarticulation in the mouth?[22] This fourfold classification of sounds, worked out in all its detailed ramifications, [23] is sufficient toaccount for all, or practically all, the sounds of language. [24]

[Footnote 22: "Points of articulation" must be understood to includetongue and lip positions of the vowels. ]

[Footnote 23: Including, under the fourth category, a number of specialresonance adjustments that we have not been able to take upspecifically. ]

[Footnote 24: In so far, it should be added, as these sounds areexpiratory, i. E. , pronounced with the outgoing breath. Certainlanguages, like the South African Hottentot and Bushman, have also anumber of inspiratory sounds, pronounced by sucking in the breath atvarious points of oral contact. These are the so-called "clicks. "]

The phonetic habits of a given language are not exhaustively defined bystating that it makes use of such and such particular sounds out of theall but endless gamut that we have briefly surveyed. There remains theimportant question of the dynamics of these phonetic elements. Twolanguages may, theoretically, be built up of precisely the same seriesof consonants and vowels and yet produce utterly different acousticeffects. One of them may not recognize striking variations in thelengths or "quantities" of the phonetic elements, the other may notesuch variations most punctiliously (in probably the majority oflanguages long and short vowels are distinguished; in many, as inItalian or Swedish or Ojibwa, long consonants are recognized as distinctfrom short ones). Or the one, say English, may be very sensitive torelative stresses, while in the other, say French, stress is a veryminor consideration. Or, again, the pitch differences which areinseparable from the actual practice of language may not affect the wordas such, but, as in English, may be a more or less random or, at best, but a rhetorical phenomenon, while in other languages, as in Swedish, Lithuanian, Chinese, Siamese, and the majority of African languages, they may be more finely graduated and felt as integral characteristicsof the words themselves. Varying methods of syllabifying are alsoresponsible for noteworthy acoustic differences. Most important of all, perhaps, are the very different possibilities of combining the phoneticelements. Each language has its peculiarities. The _ts_ combination, forinstance, is found in both English and German, but in English it canonly occur at the end of a word (as in _hats_), while it occurs freelyin German as the psychological equivalent of a single sound (as in_Zeit_, _Katze_). Some languages allow of great heapings of consonantsor of vocalic groups (diphthongs), in others no two consonants or no twovowels may ever come together. Frequently a sound occurs only in aspecial position or under special phonetic circumstances. In English, for instance, the _z_-sound of _azure_ cannot occur initially, while thepeculiar quality of the _t_ of _sting_ is dependent on its beingpreceded by the _s_. These dynamic factors, in their totality, are asimportant for the proper understanding of the phonetic genius of alanguage as the sound system itself, often far more so. 

We have already seen, in an incidental way, that phonetic elements orsuch dynamic features as quantity and stress have varying psychological"values. " The English _ts_ of _fiats_ is merely a _t_ followed by afunctionally independent _s_, the _ts_ of the German word _Zeit_ has anintegral value equivalent, say, to the _t_ of the English word _tide_. Again, the _t_ of _time_ is indeed noticeably distinct from that of_sting_, but the difference, to the consciousness of an English-speakingperson, is quite irrelevant. It has no "value. " If we compare the_t_-sounds of Haida, the Indian language spoken in the Queen CharlotteIslands, we find that precisely the same difference of articulation hasa real value. In such a word as _sting_ "two, " the _t_ is pronouncedprecisely as in English, but in _sta_ "from" the _t_ is clearly"aspirated, " like that of _time_. In other words, an objectivedifference that is irrelevant in English is of functional value inHaida; from its own psychological standpoint the _t_ of _sting_ is asdifferent from that of _sta_ as, from our standpoint, is the _t_ of_time_ from the _d_ of _divine_. Further investigation would yield theinteresting result that the Haida ear finds the difference between theEnglish _t_ of _sting_ and the _d_ of _divine_ as irrelevant as thenaïve English ear finds that of the _t_-sounds of _sting_ and _time_. The objective comparison of sounds in two or more languages is, then, ofno psychological or historical significance unless these sounds arefirst "weighted, " unless their phonetic "values" are determined. Thesevalues, in turn, flow from the general behavior and functioning of thesounds in actual speech. 

These considerations as to phonetic value lead to an importantconception. Back of the purely objective system of sounds that ispeculiar to a language and which can be arrived at only by a painstakingphonetic analysis, there is a more restricted "inner" or "ideal" systemwhich, while perhaps equally unconscious as a system to the naïvespeaker, can far more readily than the other be brought to hisconsciousness as a finished pattern, a psychological mechanism. Theinner sound-system, overlaid though it may be by the mechanical or theirrelevant, is a real and an immensely important principle in the lifeof a language. It may persist as a pattern, involving number, relation, and functioning of phonetic elements, long after its phonetic content ischanged. Two historically related languages or dialects may not have asound in common, but their ideal sound-systems may be identicalpatterns. I would not for a moment wish to imply that this pattern maynot change. It may shrink or expand or change its functionalcomplexion, but its rate of change is infinitely less rapid than that ofthe sounds as such. Every language, then, is characterized as much byits ideal system of sounds and by the underlying phonetic pattern(system, one might term it, of symbolic atoms) as by a definitegrammatical structure. Both the phonetic and conceptual structures showthe instinctive feeling of language for form. [25]

[Footnote 25: The conception of the ideal phonetic system, the phoneticpattern, of a language is not as well understood by linguistic studentsas it should be. In this respect the unschooled recorder of language, provided he has a good ear and a genuine instinct for language, is oftenat a great advantage as compared with the minute phonetician, who is aptto be swamped by his mass of observations. I have already employed myexperience in teaching Indians to write their own language for itstesting value in another connection. It yields equally valuable evidencehere. I found that it was difficult or impossible to teach an Indian tomake phonetic distinctions that did not correspond to "points in thepattern of his language, " however these differences might strike ourobjective ear, but that subtle, barely audible, phonetic differences, ifonly they hit the "points in the pattern, " were easily and voluntarilyexpressed in writing. In watching my Nootka interpreter write hislanguage, I often had the curious feeling that he was transcribing anideal flow of phonetic elements which he heard, inadequately from apurely objective standpoint, as the intention of the actual rumble ofspeech. ]

IV

FORM IN LANGUAGE: GRAMMATICAL PROCESSES

The question of form in language presents itself under two aspects. Wemay either consider the formal methods employed by a language, its"grammatical processes, " or we may ascertain the distribution ofconcepts with reference to formal expression. What are the formalpatterns of the language? And what types of concepts make up the contentof these formal patterns? The two points of view are quite distinct. TheEnglish word _unthinkingly_ is, broadly speaking, formally parallel tothe word _reformers_, each being built up on a radical element which mayoccur as an independent verb (_think_, _form_), this radical elementbeing preceded by an element (_un-_, _re-_) that conveys a definite andfairly concrete significance but that cannot be used independently, andfollowed by two elements (_-ing_, _-ly_; _-er_, _-s_) that limit theapplication of the radical concept in a relational sense. This formalpattern--(b) + A + (c) + (d)[26]--is a characteristic feature of thelanguage. A countless number of functions may be expressed by it; inother words, all the possible ideas conveyed by such prefixed andsuffixed elements, while tending to fall into minor groups, do notnecessarily form natural, functional systems. There is no logicalreason, for instance, why the numeral function of _-s_ should beformally expressed in a manner that is analogous to the expression ofthe idea conveyed by _-ly_. It is perfectly conceivable that in anotherlanguage the concept of manner (_-ly_) may be treated according to anentirely different pattern from that of plurality. The former might haveto be expressed by an independent word (say, _thus unthinking_), thelatter by a prefixed element (say, _plural[27]-reform-er_). There are, of course, an unlimited number of other possibilities. Even within theconfines of English alone the relative independence of form and functioncan be made obvious. Thus, the negative idea conveyed by _un-_ can bejust as adequately expressed by a suffixed element (_-less_) in such aword as _thoughtlessly_. Such a twofold formal expression of thenegative function would be inconceivable in certain languages, sayEskimo, where a suffixed element would alone be possible. Again, theplural notion conveyed by the _-s_ of _reformers_ is just as definitelyexpressed in the word _geese_, where an utterly distinct methodis employed. Furthermore, the principle of vocalic change(_goose_--_geese_) is by no means confined to the expression of the ideaof plurality; it may also function as an indicator of difference of time(e. G. , _sing_--_sang_, _throw_--_threw_). But the expression in Englishof past time is not by any means always bound up with a change of vowel. In the great majority of cases the same idea is expressed by means of adistinct suffix (_die-d_, _work-ed_). Functionally, _died_ and _sang_are analogous; so are _reformers_ and _geese_. Formally, we must arrangethese words quite otherwise. Both _die-d_ and _re-form-er-s_ employ themethod of suffixing grammatical elements; both _sang_ and _geese_ havegrammatical form by virtue of the fact that their vowels differ from thevowels of other words with which they are closely related in form andmeaning (_goose_; _sing_, _sung_). 

[Footnote 26: For the symbolism, see chapter II. ]

[Footnote 27: "_Plural_" is here a symbol for any prefix indicatingplurality. ]

Every language possesses one or more formal methods or indicating therelation of a secondary concept to the main concept of the radicalelement. Some of these grammatical processes, like suffixing, areexceedingly wide-spread; others, like vocalic change, are less commonbut far from rare; still others, like accent and consonantal change, aresomewhat exceptional as functional processes. Not all languages are asirregular as English in the assignment of functions to its stock ofgrammatical processes. As a rule, such basic concepts as those ofplurality and time are rendered by means of one or other method alone, but the rule has so many exceptions that we cannot safely lay it down asa principle. Wherever we go we are impressed by the fact that pattern isone thing, the utilization of pattern quite another. A few furtherexamples of the multiple expression of identical functions in otherlanguages than English may help to make still more vivid this idea ofthe relative independence of form and function. 

In Hebrew, as in other Semitic languages, the verbal idea as such isexpressed by three, less often by two or four, characteristicconsonants. Thus, the group _sh-m-r_ expresses the idea of "guarding, "the group _g-n-b_ that of "stealing, " _n-t-n_ that of "giving. "Naturally these consonantal sequences are merely abstracted from theactual forms. The consonants are held together in different forms bycharacteristic vowels that vary according to the idea that it is desiredto express. Prefixed and suffixed elements are also frequently used. Themethod of internal vocalic change is exemplified in _shamar_ "he hasguarded, " _shomer_ "guarding, " _shamur_ "being guarded, " _shmor_ "(to)guard. " Analogously, _ganab_ "he has stolen, " _goneb_ "stealing, "_ganub_ "being stolen, " _gnob_ "(to) steal. " But not all infinitives areformed according to the type of _shmor_ and _gnob_ or of other types ofinternal vowel change. Certain verbs suffix a _t_-element for theinfinitive, e. G. , _ten-eth_ "to give, " _heyo-th_ "to be. " Again, thepronominal ideas may be expressed by independent words (e. G. , _anoki_"I"), by prefixed elements (e. G. , _e-shmor_ "I shall guard"), or bysuffixed elements (e. G. , _shamar-ti_ "I have guarded"). In Nass, anIndian language of British Columbia, plurals are formed by four distinctmethods. Most nouns (and verbs) are reduplicated in the plural, that is, part of the radical element is repeated, e. G. , _gyat_ "person, "_gyigyat_ "people. " A second method is the use of certain characteristicprefixes, e. G. , _an'on_ "hand, " _ka-an'on_ "hands"; _wai_ "one paddles, "_lu-wai_ "several paddle. " Still other plurals are formed by means ofinternal vowel change, e. G. , _gwula_ "cloak, " _gwila_ "cloaks. " Finally, a fourth class of plurals is constituted by such nouns as suffix agrammatical element, e. G. , _waky_ "brother, " _wakykw_ "brothers. "

From such groups of examples as these--and they might be multiplied _adnauseam_--we cannot but conclude that linguistic form may and should bestudied as types of patterning, apart from the associated functions. Weare the more justified in this procedure as all languages evince acurious instinct for the development of one or more particulargrammatical processes at the expense of others, tending always to losesight of any explicit functional value that the process may have had inthe first instance, delighting, it would seem, in the sheer play of itsmeans of expression. It does not matter that in such a case as theEnglish _goose_--_geese_, _foul_--_defile_, _sing_--_sang_--_sung_ wecan prove that we are dealing with historically distinct processes, that the vocalic alternation of _sing_ and _sang_, for instance, iscenturies older as a specific type of grammatical process than theoutwardly parallel one of _goose_ and _geese_. It remains true thatthere is (or was) an inherent tendency in English, at the time suchforms as _geese_ came into being, for the utilization of vocalic changeas a significant linguistic method. Failing the precedent set by suchalready existing types of vocalic alternation as _sing_--_sang_--_sung_, it is highly doubtful if the detailed conditions that brought about theevolution of forms like _teeth_ and _geese_ from _tooth_ and _goose_would have been potent enough to allow the native linguistic feeling towin through to an acceptance of these new types of plural formation aspsychologically possible. This feeling for form as such, freelyexpanding along predetermined lines and greatly inhibited in certaindirections by the lack of controlling types of patterning, should bemore clearly understood than it seems to be. A general survey of manydiverse types of languages is needed to give us the proper perspectiveon this point. We saw in the preceding chapter that every language hasan inner phonetic system of definite pattern. We now learn that it hasalso a definite feeling for patterning on the level of grammaticalformation. Both of these submerged and powerfully controlling impulsesto definite form operate as such, regardless of the need for expressingparticular concepts or of giving consistent external shape to particulargroups of concepts. It goes without saying that these impulses can findrealization only in concrete functional expression. We must saysomething to be able to say it in a certain manner. 

Let us now take up a little more systematically, however briefly, thevarious grammatical processes that linguistic research has established. They may be grouped into six main types: word order; composition;affixation, including the use of prefixes, suffixes, and infixes;internal modification of the radical or grammatical element, whetherthis affects a vowel or a consonant; reduplication; and accentualdifferences, whether dynamic (stress) or tonal (pitch). There are alsospecial quantitative processes, like vocalic lengthening or shorteningand consonantal doubling, but these may be looked upon as particularsub-types of the process of internal modification. Possibly still otherformal types exist, but they are not likely to be of importance in ageneral survey. It is important to bear in mind that a linguisticphenomenon cannot be looked upon as illustrating a definite "process"unless it has an inherent functional value. The consonantal change inEnglish, for instance, of _book-s_ and _bag-s_ (_s_ in the former, _z_in the latter) is of no functional significance. It is a purelyexternal, mechanical change induced by the presence of a precedingvoiceless consonant, _k_, in the former case, of a voiced consonant, _g_, in the latter. This mechanical alternation is objectively the sameas that between the noun _house_ and the verb _to house_. In the lattercase, however, it has an important grammatical function, that oftransforming a noun into a verb. The two alternations belong, then, toentirely different psychological categories. Only the latter is a trueillustration of consonantal modification as a grammatical process. 

The simplest, at least the most economical, method of conveying somesort of grammatical notion is to juxtapose two or more words in adefinite sequence without making any attempt by inherent modification ofthese words to establish a connection between them. Let us put down twosimple English words at random, say _sing praise_. This conveys nofinished thought in English, nor does it clearly establish a relationbetween the idea of singing and that of praising. Nevertheless, it ispsychologically impossible to hear or see the two words juxtaposedwithout straining to give them some measure of coherent significance. The attempt is not likely to yield an entirely satisfactory result, butwhat is significant is that as soon as two or more radical concepts areput before the human mind in immediate sequence it strives to bind themtogether with connecting values of some sort. In the case of _singpraise_ different individuals are likely to arrive at differentprovisional results. Some of the latent possibilities of thejuxtaposition, expressed in currently satisfying form, are: _sing praise(to him)!_ or _singing praise, praise expressed in a song_ or _to singand praise_ or _one who sings a song of praise_ (compare such Englishcompounds as _killjoy_, i. E. , _one who kills joy_) or _he sings a songof praise (to him)_. The theoretical possibilities in the way ofrounding out these two concepts into a significant group of concepts oreven into a finished thought are indefinitely numerous. None of themwill quite work in English, but there are numerous languages where oneor other of these amplifying processes is habitual. It depends entirelyon the genius of the particular language what function is inherentlyinvolved in a given sequence of words. 

Some languages, like Latin, express practically all relations by meansof modifications within the body of the word itself. In these, sequenceis apt to be a rhetorical rather than a strictly grammatical principle. Whether I say in Latin _hominem femina videt_ or _femina hominem videt_or _hominem videt femina_ or _videt femina hominem_ makes little or nodifference beyond, possibly, a rhetorical or stylistic one. _The womansees the man_ is the identical significance of each of these sentences. In Chinook, an Indian language of the Columbia River, one can be equallyfree, for the relation between the verb and the two nouns is asinherently fixed as in Latin. The difference between the two languagesis that, while Latin allows the nouns to establish their relation toeach other and to the verb, Chinook lays the formal burden entirely onthe verb, the full content of which is more or less adequately renderedby _she-him-sees_. Eliminate the Latin case suffixes (_-a_ and _-em_)and the Chinook pronominal prefixes (_she-him-_) and we cannot afford tobe so indifferent to our word order. We need to husband our resources. In other words, word order takes on a real functional value. Latin andChinook are at one extreme. Such languages as Chinese, Siamese, andAnnamite, in which each and every word, if it is to function properly, falls into its assigned place, are at the other extreme. But themajority of languages fall between these two extremes. In English, forinstance, it may make little grammatical difference whether I say_yesterday the man saw the dog_ or _the man saw the dog yesterday_, butit is not a matter of indifference whether I say _yesterday the man sawthe dog_ or _yesterday the dog saw the man_ or whether I say _he ishere_ or _is he here?_ In the one case, of the latter group of examples, the vital distinction of subject and object depends entirely on theplacing of certain words of the sentence, in the latter a slightdifference of sequence makes all the difference between statement andquestion. It goes without saying that in these cases the Englishprinciple of word order is as potent a means of expression as is theLatin use of case suffixes or of an interrogative particle. There ishere no question of functional poverty, but of formal economy. 

We have already seen something of the process of composition, theuniting into a single word of two or more radical elements. Psychologically this process is closely allied to that of word order inso far as the relation between the elements is implied, not explicitlystated. It differs from the mere juxtaposition of words in the sentencein that the compounded elements are felt as constituting but parts of asingle word-organism. Such languages as Chinese and English, in whichthe principle of rigid sequence is well developed, tend not infrequentlyalso to the development of compound words. It is but a step from such aChinese word sequence as _jin tak_ "man virtue, " i. E. , "the virtue ofmen, " to such more conventionalized and psychologically unifiedjuxtapositions as _t'ien tsz_ "heaven son, " i. E. , "emperor, " or _shuifu_ "water man, " i. E. , "water carrier. " In the latter case we may aswell frankly write _shui-fu_ as a single word, the meaning of thecompound as a whole being as divergent from the precise etymologicalvalues of its component elements as is that of our English word_typewriter_ from the merely combined values of _type_ and _writer_. InEnglish the unity of the word _typewriter_ is further safeguarded by apredominant accent on the first syllable and by the possibility ofadding such a suffixed element as the plural _-s_ to the whole word. Chinese also unifies its compounds by means of stress. However, then, inits ultimate origins the process of composition may go back to typicalsequences of words in the sentence, it is now, for the most part, aspecialized method of expressing relations. French has as rigid a wordorder as English but does not possess anything like its power ofcompounding words into more complex units. On the other hand, classicalGreek, in spite of its relative freedom in the placing of words, has avery considerable bent for the formation of compound terms. 

It is curious to observe how greatly languages differ in their abilityto make use of the process of composition. One would have thought ongeneral principles that so simple a device as gives us our _typewriter_and _blackbird_ and hosts of other words would be an all but universalgrammatical process. Such is not the case. There are a great manylanguages, like Eskimo and Nootka and, aside from paltry exceptions, theSemitic languages, that cannot compound radical elements. What is evenstranger is the fact that many of these languages are not in the leastaverse to complex word-formations, but may on the contrary effect asynthesis that far surpasses the utmost that Greek and Sanskrit arecapable of. Such a Nootka word, for instance, as "when, as they say, hehad been absent for four days" might be expected to embody at leastthree radical elements corresponding to the concepts of "absent, ""four, " and "day. " As a matter of fact the Nootka word is utterlyincapable of composition in our sense. It is invariably built up out ofa single radical element and a greater or less number of suffixedelements, some of which may have as concrete a significance as theradical element itself. In, the particular case we have cited theradical element conveys the idea of "four, " the notions of "day" and"absent" being expressed by suffixes that are as inseparable from theradical nucleus of the word as is an English element like _-er_ from the_sing_ or _hunt_ of such words as _singer_ and _hunter_. The tendency toword synthesis is, then, by no means the same thing as the tendency tocompounding radical elements, though the latter is not infrequently aready means for the synthetic tendency to work with. 

There is a bewildering variety of types of composition. These typesvary according to function, the nature of the compounded elements, andorder. In a great many languages composition is confined to what we maycall the delimiting function, that is, of the two or more compoundedelements one is given a more precisely qualified significance by theothers, which contribute nothing to the formal build of the sentence. InEnglish, for instance, such compounded elements as _red_ in _redcoat_ or_over_ in _overlook_ merely modify the significance of the dominant_coat_ or _look_ without in any way sharing, as such, in the predicationthat is expressed by the sentence. Some languages, however, such asIroquois and Nahuatl, [28] employ the method of composition for muchheavier work than this. In Iroquois, for instance, the composition of anoun, in its radical form, with a following verb is a typical method ofexpressing case relations, particularly of the subject or object. _I-meat-eat_ for instance, is the regular Iroquois method of expressingthe sentence _I am eating meat_. In other languages similar forms mayexpress local or instrumental or still other relations. Such Englishforms as _killjoy_ and _marplot_ also illustrate the compounding of averb and a noun, but the resulting word has a strictly nominal, not averbal, function. We cannot say _he marplots_. Some languages allow thecomposition of all or nearly all types of elements. Paiute, forinstance, may compound noun with noun, adjective with noun, verb withnoun to make a noun, noun with verb to make a verb, adverb with verb, verb with verb. Yana, an Indian language of California, can freelycompound noun with noun and verb with noun, but not verb with verb. On the other hand, Iroquois can compound only noun with verb, nevernoun and noun as in English or verb and verb as in so many otherlanguages. Finally, each language has its characteristic types of orderof composition. In English the qualifying element regularly precedes; incertain other languages it follows. Sometimes both types are used in thesame language, as in Yana, where "beef" is "bitter-venison" but"deer-liver" is expressed by "liver-deer. " The compounded object of averb precedes the verbal element in Paiute, Nahuatl, and Iroquois, follows it in Yana, Tsimshian, [29] and the Algonkin languages. 

[Footnote 28: The language of the Aztecs, still spoken in large parts ofMexico. ]

[Footnote 29: Indian language of British Columbia closely related to theNass already cited. ]

Of all grammatical processes affixing is incomparably the mostfrequently employed. There are languages, like Chinese and Siamese, thatmake no grammatical use of elements that do not at the same time possessan independent value as radical elements, but such languages areuncommon. Of the three types of affixing--the use of prefixes, suffixes, and infixes--suffixing is much the commonest. Indeed, it is a fair guessthat suffixes do more of the formative work of language than all othermethods combined. It is worth noting that there are not a few affixinglanguages that make absolutely no use of prefixed elements but possess acomplex apparatus of suffixes. Such are Turkish, Hottentot, Eskimo, Nootka, and Yana. Some of these, like the three last mentioned, havehundreds of suffixed elements, many of them of a concreteness ofsignificance that would demand expression in the vast majority oflanguages by means of radical elements. The reverse case, the use ofprefixed elements to the complete exclusion of suffixes, is far lesscommon. A good example is Khmer (or Cambodgian), spoken in FrenchCochin-China, though even here there are obscure traces of old suffixesthat have ceased to function as such and are now felt to form part ofthe radical element. 

A considerable majority of known languages are prefixing and suffixingat one and the same time, but the relative importance of the two groupsof affixed elements naturally varies enormously. In some languages, suchas Latin and Russian, the suffixes alone relate the word to the rest ofthe sentence, the prefixes being confined to the expression of suchideas as delimit the concrete significance of the radical elementwithout influencing its bearing in the proposition. A Latin form like_remittebantur_ "they were being sent back" may serve as an illustrationof this type of distribution of elements. The prefixed element _re-_"back" merely qualifies to a certain extent the inherent significance ofthe radical element _mitt-_ "send, " while the suffixes _-eba-_, _-nt-_, and _-ur_ convey the less concrete, more strictly formal, notions oftime, person, plurality, and passivity. 

On the other hand, there are languages, like the Bantu group of Africaor the Athabaskan languages[30] of North America, in which thegrammatically significant elements precede, those that follow theradical element forming a relatively dispensable class. The Hupa word_te-s-e-ya-te_ "I will go, " for example, consists of a radical element_-ya-_ "to go, " three essential prefixes and a formally subsidiarysuffix. The element _te-_ indicates that the act takes place here andthere in space or continuously over space; practically, it has noclear-cut significance apart from such verb stems as it is customary toconnect it with. The second prefixed element, _-s-_, is even less easyto define. All we can say is that it is used in verb forms of "definite"time and that it marks action as in progress rather than as beginning orcoming to an end. The third prefix, _-e-_, is a pronominal element, "I, "which can be used only in "definite" tenses. It is highly important tounderstand that the use of _-e-_ is conditional on that of _-s-_ or ofcertain alternative prefixes and that _te-_ also is in practice linkedwith _-s-_. The group _te-s-e-ya_ is a firmly knit grammatical unit. Thesuffix _-te_, which indicates the future, is no more necessary to itsformal balance than is the prefixed _re-_ of the Latin word; it is notan element that is capable of standing alone but its function ismaterially delimiting rather than strictly formal. [31]

[Footnote 30: Including such languages as Navaho, Apache, Hupa, Carrier, Chipewyan, Loucheux. ]

[Footnote 31: This may seem surprising to an English reader. Wegenerally think of time as a function that is appropriately expressed ina purely formal manner. This notion is due to the bias that Latingrammar has given us. As a matter of fact the English future (_I shallgo_) is not expressed by affixing at all; moreover, it may be expressedby the present, as in _to-morrow I leave this place_, where the temporalfunction is inherent in the independent adverb. Though in lesser degree, the Hupa _-te_ is as irrelevant to the vital word as is _to-morrow_ tothe grammatical "feel" of _I leave_. ]

It is not always, however, that we can clearly set off the suffixes of alanguage as a group against its prefixes. In probably the majority oflanguages that use both types of affixes each group has both delimitingand formal or relational functions. The most that we can say is that alanguage tends to express similar functions in either the one or theother manner. If a certain verb expresses a certain tense by suffixing, the probability is strong that it expresses its other tenses in ananalogous fashion and that, indeed, all verbs have suffixed tenseelements. Similarly, we normally expect to find the pronominal elements, so far as they are included in the verb at all, either consistentlyprefixed or suffixed. But these rules are far from absolute. We havealready seen that Hebrew prefixes its pronominal elements in certaincases, suffixes them in others. In Chimariko, an Indian language ofCalifornia, the position of the pronominal affixes depends on the verb;they are prefixed for certain verbs, suffixed for others. 

It will not be necessary to give many further examples of prefixing andsuffixing. One of each category will suffice to illustrate theirformative possibilities. The idea expressed in English by the sentence_I came to give it to her_ is rendered in Chinook[32] by_i-n-i-a-l-u-d-am_. This word--and it is a thoroughly unified word witha clear-cut accent on the first _a_--consists of a radical element, _-d-_ "to give, " six functionally distinct, if phonetically frail, prefixed elements, and a suffix. Of the prefixes, _i-_ indicatesrecently past time; _n-_, the pronominal subject "I"; _-i-_, thepronominal object "it";[33] _-a-_, the second pronominal object "her";_-l-_, a prepositional element indicating that the preceding pronominalprefix is to be understood as an indirect object (_-her-to-_, i. E. , "toher"); and _-u-_, an element that it is not easy to definesatisfactorily but which, on the whole, indicates movement away from thespeaker. The suffixed _-am_ modifies the verbal content in a localsense; it adds to the notion conveyed by the radical element that of"arriving" or "going (or coming) for that particular purpose. " It isobvious that in Chinook, as in Hupa, the greater part of the grammaticalmachinery resides in the prefixes rather than in the suffixes. 

[Footnote 32: Wishram dialect. ]

[Footnote 33: Really "him, " but Chinook, like Latin or French, possessesgrammatical gender. An object may be referred to as "he, " "she, " or"it, " according to the characteristic form of its noun. ]

A reverse case, one in which the grammatically significant elementscluster, as in Latin, at the end of the word is yielded by Fox, one ofthe better known Algonkin languages of the Mississippi Valley. We maytake the form _eh-kiwi-n-a-m-oht-ati-wa-ch(i)_ "then they together kept(him) in flight from them. " The radical element here is _kiwi-_, a verbstem indicating the general notion of "indefinite movement round about, here and there. " The prefixed element _eh-_ is hardly more than anadverbial particle indicating temporal subordination; it may beconveniently rendered as "then. " Of the seven suffixes included in thishighly-wrought word, _-n-_ seems to be merely a phonetic element servingto connect the verb stem with the following _-a-_;[34] _-a-_ is a"secondary stem"[35] denoting the idea of "flight, to flee"; _-m-_denotes causality with reference to an animate object;[36] _-o(ht)-_indicates activity done for the subject (the so-called "middle" or"medio-passive" voice of Greek); _-(a)ti-_ is a reciprocal element, "oneanother"; _-wa-ch(i)_ is the third person animate plural (_-wa-_, plural; _-chi_, more properly personal) of so-called "conjunctive"forms. The word may be translated more literally (and yet onlyapproximately as to grammatical feeling) as "then they (animate) causedsome animate being to wander about in flight from one another ofthemselves. " Eskimo, Nootka, Yana, and other languages have similarlycomplex arrays of suffixed elements, though the functions performed bythem and their principles of combination differ widely. 

[Footnote 34: This analysis is doubtful. It is likely that _-n-_possesses a function that still remains to be ascertained. The Algonkinlanguages are unusually complex and present many unsolved problems ofdetail. ]

[Footnote 35: "Secondary stems" are elements which are suffixes from aformal point of view, never appearing without the support of a trueradical element, but whose function is as concrete, to all intents andpurposes, as that of the radical element itself. Secondary verb stems ofthis type are characteristic of the Algonkin languages and of Yana. ]

[Footnote 36: In the Algonkin languages all persons and things areconceived of as either animate or inanimate, just as in Latin or Germanthey are conceived of as masculine, feminine, or neuter. ]

We have reserved the very curious type of affixation known as "infixing"for separate illustration. It is utterly unknown in English, unless weconsider the _-n-_ of _stand_ (contrast _stood_) as an infixed element. The earlier Indo-European languages, such as Latin, Greek and Sanskrit, made a fairly considerable use of infixed nasals to differentiate thepresent tense of a certain class of verbs from other forms (contrastLatin _vinc-o_ "I conquer" with _vic-i_ "I conquered"; Greek _lamb-an-o_"I take" with _e-lab-on_ "I took"). There are, however, more strikingexamples of the process, examples in which it has assumed a more clearlydefined function than in these Latin and Greek cases. It is particularlyprevalent in many languages of southeastern Asia and of the Malayarchipelago. Good examples from Khmer (Cambodgian) are _tmeu_ "one whowalks" and _daneu_ "walking" (verbal noun), both derived from _deu_ "towalk. " Further examples may be quoted from Bontoc Igorot, a Filipinolanguage. Thus, an infixed _-in-_ conveys the idea of the product of anaccomplished action, e. G. , _kayu_ "wood, " _kinayu_ "gathered wood. "Infixes are also freely used in the Bontoc Igorot verb. Thus, an infixed_-um-_ is characteristic of many intransitive verbs with personalpronominal suffixes, e. G. , _sad-_ "to wait, " _sumid-ak_ "I wait";_kineg_ "silent, " _kuminek-ak_ "I am silent. " In other verbs itindicates futurity, e. G. , _tengao-_ "to celebrate a holiday, "_tumengao-ak_ "I shall have a holiday. " The past tense is frequentlyindicated by an infixed _-in-_; if there is already an infixed _-um-_, the two elements combine to _-in-m-_, e. G. , _kinminek-ak_ "I am silent. "Obviously the infixing process has in this (and related) languages thesame vitality that is possessed by the commoner prefixes and suffixesof other languages. The process is also found in a number of aboriginalAmerican languages. The Yana plural is sometimes formed by an infixedelement, e. G. , _k'uruwi_ "medicine-men, " _k'uwi_ "medicine-man"; inChinook an infixed _-l-_ is used in certain verbs to indicate repeatedactivity, e. G. , _ksik'ludelk_ "she keeps looking at him, " _iksik'lutk_"she looked at him" (radical element _-tk_). A peculiarly interestingtype of infixation is found in the Siouan languages, in which certainverbs insert the pronominal elements into the very body of the radicalelement, e. G. , Sioux _cheti_ "to build a fire, " _chewati_ "I build afire"; _shuta_ "to miss, " _shuunta-pi_ "we miss. "

A subsidiary but by no means unimportant grammatical process is that ofinternal vocalic or consonantal change. In some languages, as in English(_sing_, _sang_, _sung_, _song_; _goose_, _geese_), the former of thesehas become one of the major methods of indicating fundamental changes ofgrammatical function. At any rate, the process is alive enough to leadour children into untrodden ways. We all know of the growing youngsterwho speaks of having _brung_ something, on the analogy of such forms as_sung_ and _flung_. In Hebrew, as we have seen, vocalic change is ofeven greater significance than in English. What is true of Hebrew is ofcourse true of all other Semitic languages. A few examples of so-called"broken" plurals from Arabic[37] will supplement the Hebrew verb formsthat I have given in another connection. The noun _balad_ "place" hasthe plural form _bilad_;[38] _gild_ "hide" forms the plural _gulud_;_ragil_ "man, " the plural _rigal_; _shibbak_ "window, " the plural_shababik_. Very similar phenomena are illustrated by the Hamiticlanguages of Northern Africa, e. G. , Shilh[39] _izbil_ "hair, " plural_izbel_; _a-slem_ "fish, " plural _i-slim-en_; _sn_ "to know, " _sen_ "tobe knowing"; _rmi_ "to become tired, " _rumni_ "to be tired"; _ttss_[40]"to fall asleep, " _ttoss_ "to sleep. " Strikingly similar to English andGreek alternations of the type _sing_--_sang_ and _leip-o_ "I leave, "_leloip-a_ "I have left, " are such Somali[41] cases as _al_ "I am, " _il_"I was"; _i-dah-a_ "I say, " _i-di_ "I said, " _deh_ "say!"

[Footnote 37: Egyptian dialect. ]

[Footnote 38: There are changes of accent and vocalic quantity in theseforms as well, but the requirements of simplicity force us to neglectthem. ]

[Footnote 39: A Berber language of Morocco. ]

[Footnote 40: Some of the Berber languages allow consonantalcombinations that seem unpronounceable to us. ]

[Footnote 41: One of the Hamitic languages of eastern Africa. ]

Vocalic change is of great significance also in a number of AmericanIndian languages. In the Athabaskan group many verbs change the qualityor quantity of the vowel of the radical element as it changes its tenseor mode. The Navaho verb for "I put (grain) into a receptacle" is_bi-hi-sh-ja_, in which _-ja_ is the radical element; the past tense, _bi-hi-ja'_, has a long _a_-vowel, followed by the "glottal stop"[42];the future is _bi-h-de-sh-ji_ with complete change of vowel. In othertypes of Navaho verbs the vocalic changes follow different lines, e. G. , _yah-a-ni-ye_ "you carry (a pack) into (a stable)"; past, _yah-i-ni-yin_(with long _i_ in _-yin_; _-n_ is here used to indicate nasalization);future, _yah-a-di-yehl_ (with long _e_). In another Indian language, Yokuts[43], vocalic modifications affect both noun and verb forms. Thus, _buchong_ "son" forms the plural _bochang-i_ (contrast the objective_buchong-a_); _enash_ "grandfather, " the plural _inash-a_; the verb_engtyim_ "to sleep" forms the continuative _ingetym-ad_ "to besleeping" and the past _ingetym-ash_. 

[Footnote 42: See page 49. ]

[Transcriber's note: Footnote 42 refers to the paragraph beginning online 1534. ]

[Footnote 43: Spoken in the south-central part of California. ]

Consonantal change as a functional process is probably far less commonthan vocalic modifications, but it is not exactly rare. There is aninteresting group of cases in English, certain nouns and correspondingverbs differing solely in that the final consonant is voiceless orvoiced. Examples are _wreath_ (with _th_ as in _think_), but _towreathe_ (with _th_ as in _then_); _house_, but _to house_ (with _s_pronounced like _z_). That we have a distinct feeling for theinterchange as a means of distinguishing the noun from the verb isindicated by the extension of the principle by many Americans to such anoun as _rise_ (e. G. , _the rise of democracy_)--pronounced like_rice_--in contrast to the verb _to rise_ (_s_ like _z_). 

In the Celtic languages the initial consonants undergo several types ofchange according to the grammatical relation that subsists between theword itself and the preceding word. Thus, in modern Irish, a word like_bo_ "ox" may under the appropriate circumstances, take the forms _bho_(pronounce _wo_) or _mo_ (e. G. , _an bo_ "the ox, " as a subject, but _tirna mo_ "land of the oxen, " as a possessive plural). In the verb theprinciple has as one of its most striking consequences the "aspiration"of initial consonants in the past tense. If a verb begins with _t_, say, it changes the _t_ to _th_ (now pronounced _h_) in forms of the past; ifit begins with _g_, the consonant changes, in analogous forms, to _gh_(pronounced like a voiced spirant[44] _g_ or like _y_, according to thenature of the following vowel). In modern Irish the principle ofconsonantal change, which began in the oldest period of the language asa secondary consequence of certain phonetic conditions, has become oneof the primary grammatical processes of the language. 

[Footnote 44: See page 50. ]

[Transcriber's note: Footnote 44 refers to the paragraph beginning online 1534. ]

Perhaps as remarkable as these Irish phenomena are the consonantalinterchanges of Ful, an African language of the Soudan. Here we findthat all nouns belonging to the personal class form the plural bychanging their initial _g_, _j_, _d_, _b_, _k_, _ch_, and _p_ to _y_ (or_w_), _y_, _r_, _w_, _h_, _s_ and _f_ respectively; e. G. , _jim-o_"companion, " _yim-'be_ "companions"; _pio-o_ "beater, " _fio-'be_"beaters. " Curiously enough, nouns that belong to the class of thingsform their singular and plural in exactly reverse fashion, e. G. , _yola-re_ "grass-grown place, " _jola-je_ "grass-grown places";_fitan-du_ "soul, " _pital-i_ "souls. " In Nootka, to refer to but oneother language in which the process is found, the _t_ or _tl_[45] ofmany verbal suffixes becomes _hl_ in forms denoting repetition, e. G. , _hita-'ato_ "to fall out, " _hita-'ahl_ "to keep falling out";_mat-achisht-utl_ "to fly on to the water, " _mat-achisht-ohl_ "to keepflying on to the water. " Further, the _hl_ of certain elements changesto a peculiar _h_-sound in plural forms, e. G. , _yak-ohl_ "sore-faced, "_yak-oh_ "sore-faced (people). "

[Footnote 45: These orthographies are but makeshifts for simple sounds. ]

Nothing is more natural than the prevalence of reduplication, in otherwords, the repetition of all or part of the radical element. The processis generally employed, with self-evident symbolism, to indicate suchconcepts as distribution, plurality, repetition, customary activity, increase of size, added intensity, continuance. Even in English it isnot unknown, though it is not generally accounted one of the typicalformative devices of our language. Such words as _goody-goody_ and _topooh-pooh_ have become accepted as part of our normal vocabulary, butthe method of duplication may on occasion be used more freely than isindicated by such stereotyped examples. Such locutions as _a big bigman_ or _Let it cool till it's thick thick_ are far more common, especially in the speech of women and children, than our linguistictext-books would lead one to suppose. In a class by themselves are thereally enormous number of words, many of them sound-imitative orcontemptuous in psychological tone, that consist of duplications witheither change of the vowel or change of the initial consonant--words ofthe type _sing-song_, _riff-raff_, _wishy-washy_, _harum-skarum_, _roly-poly_. Words of this type are all but universal. Such examples asthe Russian _Chudo-Yudo_ (a dragon), the Chinese _ping-pang_ "rattlingof rain on the roof, "[46] the Tibetan _kyang-kyong_ "lazy, " and theManchu _porpon parpan_ "blear-eyed" are curiously reminiscent, both inform and in psychology, of words nearer home. But it can hardly be saidthat the duplicative process is of a distinctively grammaticalsignificance in English. We must turn to other languages forillustration. Such cases as Hottentot _go-go_ "to look at carefully"(from _go_ "to see"), Somali _fen-fen_ "to gnaw at on all sides" (from_fen_ "to gnaw at"), Chinook _iwi iwi_ "to look about carefully, toexamine" (from _iwi_ "to appear"), or Tsimshian _am'am_ "several (are)good" (from _am_ "good") do not depart from the natural and fundamentalrange of significance of the process. A more abstract function isillustrated in Ewe, [47] in which both infinitives and verbal adjectivesare formed from verbs by duplication; e. G. , _yi_ "to go, " _yiyi_ "to go, act of going"; _wo_ "to do, " _wowo_[48] "done"; _mawomawo_ "not to do"(with both duplicated verb stem and duplicated negative particle). Causative duplications are characteristic of Hottentot, e. G. , _gam-gam_[49] "to cause to tell" (from _gam_ "to tell"). Or the processmay be used to derive verbs from nouns, as in Hottentot _khoe-khoe_ "totalk Hottentot" (from _khoe-b_ "man, Hottentot"), or as in Kwakiutl_metmat_ "to eat clams" (radical element _met-_ "clam"). 

[Footnote 46: Whence our _ping-pong_. ]

[Footnote 47: An African language of the Guinea Coast. ]

[Footnote 48: In the verbal adjective the tone of the second syllablediffers from that of the first. ]

[Footnote 49: Initial "click" (see page 55, note 15) omitted. ]

[Transcriber's note: Footnote 49 refers to Footnote 24, beginning online 1729. ]

The most characteristic examples of reduplication are such as repeatonly part of the radical element. It would be possible to demonstratethe existence of a vast number of formal types of such partialduplication, according to whether the process makes use of one or moreof the radical consonants, preserves or weakens or alters the radicalvowel, or affects the beginning, the middle, or the end of the radicalelement. The functions are even more exuberantly developed than withsimple duplication, though the basic notion, at least in origin, isnearly always one of repetition or continuance. Examples illustratingthis fundamental function can be quoted from all parts of the globe. Initially reduplicating are, for instance, Shilh _ggen_ "to be sleeping"(from _gen_ "to sleep"); Ful _pepeu-'do_ "liar" (i. E. , "one who alwayslies"), plural _fefeu-'be_ (from _fewa_ "to lie"); Bontoc Igorot _anak_"child, " _ananak_ "children"; _kamu-ek_ "I hasten, " _kakamu-ek_ "Ihasten more"; Tsimshian _gyad_ "person, " _gyigyad_ "people"; Nass_gyibayuk_ "to fly, " _gyigyibayuk_ "one who is flying. " Psychologicallycomparable, but with the reduplication at the end, are Somali _ur_"body, " plural _urar_; Hausa _suna_ "name, " plural _sunana-ki;_Washo[50] _gusu_ "buffalo, " _gususu_ "buffaloes"; Takelma[51] _himi-d-_"to talk to, " _himim-d-_ "to be accustomed to talk to. " Even morecommonly than simple duplication, this partial duplication of theradical element has taken on in many languages functions that seem in noway related to the idea of increase. The best known examples areprobably the initial reduplication of our older Indo-European languages, which helps to form the perfect tense of many verbs (e. G. , Sanskrit_dadarsha_ "I have seen, " Greek _leloipa_ "I have left, " Latin _tetigi_"I have touched, " Gothic _lelot_ "I have let"). In Nootka reduplicationof the radical element is often employed in association with certainsuffixes; e. G. , _hluch-_ "woman" forms _hluhluch-'ituhl_ "to dream of awoman, " _hluhluch-k'ok_ "resembling a woman. " Psychologically similar tothe Greek and Latin examples are many Takelma cases of verbs thatexhibit two forms of the stem, one employed in the present or past, theother in the future and in certain modes and verbal derivatives. Theformer has final reduplication, which is absent in the latter; e. G. , _al-yebeb-i'n_ "I show (or showed) to him, " _al-yeb-in_ "I shall showhim. "

[Footnote 50: An Indian language of Nevada. ]

[Footnote 51: An Indian language of Oregon. ]

We come now to the subtlest of all grammatical processes, variations inaccent, whether of stress or pitch. The chief difficulty in isolatingaccent as a functional process is that it is so often combined withalternations in vocalic quantity or quality or complicated by thepresence of affixed elements that its grammatical value appears as asecondary rather than as a primary feature. In Greek, for instance, itis characteristic of true verbal forms that they throw the accent backas far as the general accentual rules will permit, while nouns may bemore freely accented. There is thus a striking accentual differencebetween a verbal form like _eluthemen_ "we were released, " accented onthe second syllable of the word, and its participial derivative_lutheis_ "released, " accented on the last. The presence of thecharacteristic verbal elements _e-_ and _-men_ in the first case and ofthe nominal _-s_ in the second tends to obscure the inherent value ofthe accentual alternation. This value comes out very neatly in suchEnglish doublets as _to refund_ and _a refund_, _to extract_ and _anextract, to come down_ and _a come down_, _to lack luster_ and_lack-luster eyes_, in which the difference between the verb and thenoun is entirely a matter of changing stress. In the Athabaskanlanguages there are not infrequently significant alternations of accent, as in Navaho _ta-di-gis_ "you wash yourself" (accented on the secondsyllable), _ta-di-gis_ "he washes himself" (accented on the first). [52]

[Footnote 52: It is not unlikely, however, that these Athabaskanalternations are primarily tonal in character. ]

Pitch accent may be as functional as stress and is perhaps more oftenso. The mere fact, however, that pitch variations are phoneticallyessential to the language, as in Chinese (e. G. , _feng_ "wind" with alevel tone, _feng_ "to serve" with a falling tone) or as in classicalGreek (e. G. , _lab-on_ "having taken" with a simple or high tone on thesuffixed participial _-on_, _gunaik-on_ "of women" with a compound orfalling tone on the case suffix _-on_) does not necessarily constitute afunctional, or perhaps we had better say grammatical, use of pitch. Insuch cases the pitch is merely inherent in the radical element or affix, as any vowel or consonant might be. It is different with such Chinesealternations as _chung_ (level) "middle" and _chung_ (falling) "to hitthe middle"; _mai_ (rising) "to buy" and _mai_ (falling) "to sell";_pei_ (falling) "back" and _pei_ (level) "to carry on the back. "Examples of this type are not exactly common in Chinese and the languagecannot be said to possess at present a definite feeling for tonaldifferences as symbolic of the distinction between noun and verb. 

There are languages, however, in which such differences are of the mostfundamental grammatical importance. They are particularly common in theSoudan. In Ewe, for instance, there are formed from _subo_ "to serve"two reduplicated forms, an infinitive _subosubo_ "to serve, " with a lowtone on the first two syllables and a high one on the last two, and anadjectival _subosubo_ "serving, " in which all the syllables have a hightone. Even more striking are cases furnished by Shilluk, one of thelanguages of the headwaters of the Nile. The plural of the noun oftendiffers in tone from the singular, e. G. , _yit_ (high) "ear" but _yit_(low) "ears. " In the pronoun three forms may be distinguished by tonealone; _e_ "he" has a high tone and is subjective, _-e_ "him" (e. G. , _achwol-e_ "he called him") has a low tone and is objective, _-e_ "his"(e. G. , _wod-e_ "his house") has a middle tone and is possessive. Fromthe verbal element _gwed-_ "to write" are formed _gwed-o_ "(he) writes"with a low tone, the passive _gwet_ "(it was) written" with a fallingtone, the imperative _gwet_ "write!" with a rising tone, and the verbalnoun _gwet_ "writing" with a middle tone. In aboriginal America alsopitch accent is known to occur as a grammatical process. A good exampleof such a pitch language is Tlingit, spoken by the Indians of thesouthern coast of Alaska. In this language many verbs vary the tone ofthe radical element according to tense; _hun_ "to sell, " _sin_ "tohide, " _tin_ "to see, " and numerous other radical elements, iflow-toned, refer to past time, if high-toned, to the future. Anothertype of function is illustrated by the Takelma forms _hel_ "song, " withfalling pitch, but _hel_ "sing!" with a rising inflection; parallel tothese forms are _sel_ (falling) "black paint, " _sel_ (rising) "paintit!" All in all it is clear that pitch accent, like stress and vocalicor consonantal modifications, is far less infrequently employed as agrammatical process than our own habits of speech would prepare us tobelieve probable. 

V

FORM IN LANGUAGE: GRAMMATICAL CONCEPTS

We have seen that the single word expresses either a simple concept or acombination of concepts so interrelated as to form a psychologicalunity. We have, furthermore, briefly reviewed from a strictly formalstandpoint the main processes that are used by all known languages toaffect the fundamental concepts--those embodied in unanalyzable words orin the radical elements of words--by the modifying or formativeinfluence of subsidiary concepts. In this chapter we shall look a littlemore closely into the nature of the world of concepts, in so far as thatworld is reflected and systematized in linguistic structure. 

Let us begin with a simple sentence that involves various kinds ofconcepts--_the farmer kills the duckling_. A rough and ready analysisdiscloses here the presence of three distinct and fundamental conceptsthat are brought into connection with each other in a number of ways. These three concepts are "farmer" (the subject of discourse), "kill"(defining the nature of the activity which the sentence informs usabout), and "duckling" (another subject[53] of discourse that takes animportant though somewhat passive part in this activity). We canvisualize the farmer and the duckling and we have also no difficulty inconstructing an image of the killing. In other words, the elements_farmer_, _kill_, and _duckling_ define concepts of a concrete order. 

[Footnote 53: Not in its technical sense. ]

But a more careful linguistic analysis soon brings us to see that thetwo subjects of discourse, however simply we may visualize them, are notexpressed quite as directly, as immediately, as we feel them. A "farmer"is in one sense a perfectly unified concept, in another he is "one whofarms. " The concept conveyed by the radical element (_farm-_) is not oneof personality at all but of an industrial activity (_to farm_), itselfbased on the concept of a particular type of object (_a farm_). Similarly, the concept of _duckling_ is at one remove from that which isexpressed by the radical element of the word, _duck_. This element, which may occur as an independent word, refers to a whole class ofanimals, big and little, while _duckling_ is limited in its applicationto the young of that class. The word _farmer_ has an "agentive" suffix_-er_ that performs the function of indicating the one that carries outa given activity, in this case that of farming. It transforms the verb_to farm_ into an agentive noun precisely as it transforms the verbs _tosing_, _to paint_, _to teach_ into the corresponding agentive nouns_singer_, _painter_, _teacher_. The element _-ling_ is not so freelyused, but its significance is obvious. It adds to the basic concept thenotion of smallness (as also in _gosling_, _fledgeling_) or the somewhatrelated notion of "contemptible" (as in _weakling_, _princeling_, _hireling_). The agentive _-er_ and the diminutive _-ling_ both conveyfairly concrete ideas (roughly those of "doer" and "little"), but theconcreteness is not stressed. They do not so much define distinctconcepts as mediate between concepts. The _-er_ of _farmer_ does notquite say "one who (farms)" it merely indicates that the sort of personwe call a "farmer" is closely enough associated with activity on a farmto be conventionally thought of as always so occupied. He may, as amatter of fact, go to town and engage in any pursuit but farming, yethis linguistic label remains "farmer. " Language here betrays a certainhelplessness or, if one prefers, a stubborn tendency to look away fromthe immediately suggested function, trusting to the imagination and tousage to fill in the transitions of thought and the details ofapplication that distinguish one concrete concept (_to farm_) fromanother "derived" one (_farmer_). It would be impossible for anylanguage to express every concrete idea by an independent word orradical element. The concreteness of experience is infinite, theresources of the richest language are strictly limited. It must perforcethrow countless concepts under the rubric of certain basic ones, usingother concrete or semi-concrete ideas as functional mediators. The ideasexpressed by these mediating elements--they may be independent words, affixes, or modifications of the radical element--may be called"derivational" or "qualifying. " Some concrete concepts, such as _kill_, are expressed radically; others, such as _farmer_ and _duckling_, areexpressed derivatively. Corresponding to these two modes of expressionwe have two types of concepts and of linguistic elements, radical(_farm_, _kill_, _duck_) and derivational (_-er_, _-ling_). When a word(or unified group of words) contains a derivational element (or word)the concrete significance of the radical element (_farm-_, _duck-_)tends to fade from consciousness and to yield to a new concreteness(_farmer_, _duckling_) that is synthetic in expression rather than inthought. In our sentence the concepts of _farm_ and _duck_ are notreally involved at all; they are merely latent, for formal reasons, inthe linguistic expression. 

Returning to this sentence, we feel that the analysis of _farmer_ and_duckling_ are practically irrelevant to an understanding of its contentand entirely irrelevant to a feeling for the structure of the sentenceas a whole. From the standpoint of the sentence the derivationalelements _-er_ and _-ling_ are merely details in the local economy oftwo of its terms (_farmer_, _duckling_) that it accepts as units ofexpression. This indifference of the sentence as such to some part ofthe analysis of its words is shown by the fact that if we substitutesuch radical words as _man_ and _chick_ for _farmer_ and _duckling_, weobtain a new material content, it is true, but not in the least a newstructural mold. We can go further and substitute another activity forthat of "killing, " say "taking. " The new sentence, _the man takes thechick_, is totally different from the first sentence in what it conveys, not in how it conveys it. We feel instinctively, without the slightestattempt at conscious analysis, that the two sentences fit precisely thesame pattern, that they are really the same fundamental sentence, differing only in their material trappings. In other words, they expressidentical relational concepts in an identical manner. The manner is herethreefold--the use of an inherently relational word (_the_) in analogouspositions, the analogous sequence (subject; predicate, consisting ofverb and object) of the concrete terms of the sentence, and the use ofthe suffixed element _-s_ in the verb. 

Change any of these features of the sentence and it becomes modified, slightly or seriously, in some purely relational, non-material regard. If _the_ is omitted (_farmer kills duckling_, _man takes chick_), thesentence becomes impossible; it falls into no recognized formal patternand the two subjects of discourse seem to hang incompletely in the void. We feel that there is no relation established between either of themand what is already in the minds of the speaker and his auditor. As soonas a _the_ is put before the two nouns, we feel relieved. We know thatthe farmer and duckling which the sentence tells us about are the samefarmer and duckling that we had been talking about or hearing about orthinking about some time before. If I meet a man who is not looking atand knows nothing about the farmer in question, I am likely to be staredat for my pains if I announce to him that "the farmer [what farmer?]kills the duckling [didn't know he had any, whoever he is]. " If the factnevertheless seems interesting enough to communicate, I should becompelled to speak of "_a farmer_ up my way" and of "_a duckling_ ofhis. " These little words, _the_ and _a_, have the important function ofestablishing a definite or an indefinite reference. 

If I omit the first _the_ and also leave out the suffixed _-s_, I obtainan entirely new set of relations. _Farmer, kill the duckling_ impliesthat I am now speaking to the farmer, not merely about him; further, that he is not actually killing the bird, but is being ordered by me todo so. The subjective relation of the first sentence has become avocative one, one of address, and the activity is conceived in terms ofcommand, not of statement. We conclude, therefore, that if the farmer isto be merely talked about, the little _the_ must go back into its placeand the _-s_ must not be removed. The latter element clearly defines, orrather helps to define, statement as contrasted with command. I find, moreover, that if I wish to speak of several farmers, I cannot say _thefarmers kills the duckling_, but must say _the farmers kill theduckling_. Evidently _-s_ involves the notion of singularity in thesubject. If the noun is singular, the verb must have a form tocorrespond; if the noun is plural, the verb has another, correspondingform. [54] Comparison with such forms as _I kill_ and _you kill_ shows, moreover, that the _-s_ has exclusive reference to a person other thanthe speaker or the one spoken to. We conclude, therefore, that itconnotes a personal relation as well as the notion of singularity. Andcomparison with a sentence like _the farmer killed the duckling_indicates that there is implied in this overburdened _-s_ a distinctreference to present time. Statement as such and personal reference maywell be looked upon as inherently relational concepts. Number isevidently felt by those who speak English as involving a necessaryrelation, otherwise there would be no reason to express the concepttwice, in the noun and in the verb. Time also is clearly felt as arelational concept; if it were not, we should be allowed to say _thefarmer killed-s_ to correspond to _the farmer kill-s_. Of the fourconcepts inextricably interwoven in the _-s_ suffix, all are felt asrelational, two necessarily so. The distinction between a trulyrelational concept and one that is so felt and treated, though it neednot be in the nature of things, will receive further attention in amoment. 

[Footnote 54: It is, of course, an "accident" that _-s_ denotesplurality in the noun, singularity in the verb. ]

Finally, I can radically disturb the relational cut of the sentence bychanging the order of its elements. If the positions of _farmer_ and_kills_ are interchanged, the sentence reads _kills the farmer theduckling_, which is most naturally interpreted as an unusual but notunintelligible mode of asking the question, _does the farmer kill theduckling?_ In this new sentence the act is not conceived as necessarilytaking place at all. It may or it may not be happening, the implicationbeing that the speaker wishes to know the truth of the matter and thatthe person spoken to is expected to give him the information. Theinterrogative sentence possesses an entirely different "modality" fromthe declarative one and implies a markedly different attitude of thespeaker towards his companion. An even more striking change in personalrelations is effected if we interchange _the farmer_ and _the duckling_. _The duckling kills the farmer_ involves precisely the same subjects ofdiscourse and the same type of activity as our first sentence, but theroles of these subjects of discourse are now reversed. The duckling hasturned, like the proverbial worm, or, to put it in grammaticalterminology, what was "subject" is now "object, " what was object is nowsubject. 

The following tabular statement analyzes the sentence from the point ofview of the concepts expressed in it and of the grammatical processesemployed for their expression. 

 I. CONCRETE CONCEPTS: 1. First subject of discourse: _farmer_ 2. Second subject of discourse: _duckling_ 3. Activity: _kill_ ---- analyzable into: A. RADICAL CONCEPTS: 1. Verb: _(to) farm_ 2. Noun: _duck_ 3. Verb: _kill_ B. DERIVATIONAL CONCEPTS: 1. Agentive: expressed by suffix _-er_ 2. Diminutive: expressed by suffix _-ling_II. RELATIONAL CONCEPTS: Reference: 1. Definiteness of reference to first subject of discourse: expressed by first _the_, which has preposed position 2. Definiteness of reference to second subject of discourse: expressed by second _the_, which has preposed position Modality: 3. Declarative: expressed by sequence of "subject" plus verb; and implied by suffixed _-s_ Personal relations: 4. Subjectivity of _farmer_: expressed by position of _farmer_ before kills; and by suffixed _-s_ 5. Objectivity of _duckling_: expressed by position of _duckling_ after _kills_ Number: 6. Singularity of first subject of discourse: expressed by lack of plural suffix in _farmer_; and by suffix _-s_ in following verb 7. Singularity of second subject of discourse: expressed by lack of plural suffix in _duckling_ Time: 8. Present: expressed by lack of preterit suffix in verb; and by suffixed _-s_

In this short sentence of five words there are expressed, therefore, thirteen distinct concepts, of which three are radical and concrete, twoderivational, and eight relational. Perhaps the most striking result ofthe analysis is a renewed realization of the curious lack of accord inour language between function and form. The method of suffixing is usedboth for derivational and for relational elements; independent words orradical elements express both concrete ideas (objects, activities, qualities) and relational ideas (articles like _the_ and _a_; wordsdefining case relations, like _of_, _to_, _for_, _with_, _by_; wordsdefining local relations, like _in_, _on_, _at_); the same relationalconcept may be expressed more than once (thus, the singularity of_farmer_ is both negatively expressed in the noun and positively in theverb); and one element may convey a group of interwoven concepts ratherthan one definite concept alone (thus the _-s_ of _kills_ embodies noless than four logically independent relations). 

Our analysis may seem a bit labored, but only because we are soaccustomed to our own well-worn grooves of expression that they havecome to be felt as inevitable. Yet destructive analysis of the familiaris the only method of approach to an understanding of fundamentallydifferent modes of expression. When one has learned to feel what isfortuitous or illogical or unbalanced in the structure of his ownlanguage, he is already well on the way towards a sympathetic grasp ofthe expression of the various classes of concepts in alien types ofspeech. Not everything that is "outlandish" is intrinsically illogicalor far-fetched. It is often precisely the familiar that a widerperspective reveals as the curiously exceptional. From a purely logicalstandpoint it is obvious that there is no inherent reason why theconcepts expressed in our sentence should have been singled out, treated, and grouped as they have been and not otherwise. The sentenceis the outgrowth of historical and of unreasoning psychological forcesrather than of a logical synthesis of elements that have been clearlygrasped in their individuality. This is the case, to a greater or lessdegree, in all languages, though in the forms of many we find a morecoherent, a more consistent, reflection than in our English forms ofthat unconscious analysis into individual concepts which is neverentirely absent from speech, however it may be complicated with oroverlaid by the more irrational factors. 

A cursory examination of other languages, near and far, would soon showthat some or all of the thirteen concepts that our sentence happens toembody may not only be expressed in different form but that they may bedifferently grouped among themselves; that some among them may bedispensed with; and that other concepts, not considered worth expressingin English idiom, may be treated as absolutely indispensable to theintelligible rendering of the proposition. First as to a differentmethod of handling such concepts as we have found expressed in theEnglish sentence. If we turn to German, we find that in the equivalentsentence (_Der Bauer tötet das Entelein_) the definiteness of referenceexpressed by the English _the_ is unavoidably coupled with three otherconcepts--number (both _der_ and _das_ are explicitly singular), case(_der_ is subjective; _das_ is subjective or objective, by eliminationtherefore objective), and gender, a new concept of the relational orderthat is not in this case explicitly involved in English (_der_ ismasculine, _das_ is neuter). Indeed, the chief burden of the expressionof case, gender, and number is in the German sentence borne by theparticles of reference rather than by the words that express theconcrete concepts (_Bauer_, _Entelein_) to which these relationalconcepts ought logically to attach themselves. In the sphere of concreteconcepts too it is worth noting that the German splits up the idea of"killing" into the basic concept of "dead" (_tot_) and the derivationalone of "causing to do (or be) so and so" (by the method of vocalicchange, _töt-_); the German _töt-et_ (analytically _tot-_+vowelchange+_-et_) "causes to be dead" is, approximately, the formalequivalent of our _dead-en-s_, though the idiomatic application of thislatter word is different. [55]

[Footnote 55: "To cause to be dead" or "to cause to die" in the sense of"to kill" is an exceedingly wide-spread usage. It is found, forinstance, also in Nootka and Sioux. ]

Wandering still further afield, we may glance at the Yana method ofexpression. Literally translated, the equivalent Yana sentence wouldread something like "kill-s he farmer[56] he to duck-ling, " in which"he" and "to" are rather awkward English renderings of a general thirdpersonal pronoun (_he_, _she_, _it_, or _they_) and an objectiveparticle which indicates that the following noun is connected with theverb otherwise than as subject. The suffixed element in "kill-s"corresponds to the English suffix with the important exceptions that itmakes no reference to the number of the subject and that the statementis known to be true, that it is vouched for by the speaker. Number isonly indirectly expressed in the sentence in so far as there is nospecific verb suffix indicating plurality of the subject nor specificplural elements in the two nouns. Had the statement been made onanother's authority, a totally different "tense-modal" suffix would havehad to be used. The pronouns of reference ("he") imply nothing bythemselves as to number, gender, or case. Gender, indeed, is completelyabsent in Yana as a relational category. 

[Footnote 56: Agriculture was not practised by the Yana. The verbal ideaof "to farm" would probably be expressed in some such synthetic manneras "to dig-earth" or "to grow-cause. " There are suffixed elementscorresponding to _-er_ and _-ling_. ]

The Yana sentence has already illustrated the point that certain of oursupposedly essential concepts may be ignored; both the Yana and theGerman sentence illustrate the further point that certain concepts mayneed expression for which an English-speaking person, or rather theEnglish-speaking habit, finds no need whatever. One could go on and giveendless examples of such deviations from English form, but we shall haveto content ourselves with a few more indications. In the Chinesesentence "Man kill duck, " which may be looked upon as the practicalequivalent of "The man kills the duck, " there is by no means presentfor the Chinese consciousness that childish, halting, empty feelingwhich we experience in the literal English translation. The threeconcrete concepts--two objects and an action--are each directlyexpressed by a monosyllabic word which is at the same time a radicalelement; the two relational concepts--"subject" and "object"--areexpressed solely by the position of the concrete words before and afterthe word of action. And that is all. Definiteness or indefiniteness ofreference, number, personality as an inherent aspect of the verb, tense, not to speak of gender--all these are given no expression in theChinese sentence, which, for all that, is a perfectly adequatecommunication--provided, of course, there is that context, thatbackground of mutual understanding that is essential to the completeintelligibility of all speech. Nor does this qualification impair ourargument, for in the English sentence too we leave unexpressed a largenumber of ideas which are either taken for granted or which have beendeveloped or are about to be developed in the course of theconversation. Nothing has been said, for example, in the English, German, Yana, or Chinese sentence as to the place relations of thefarmer, the duck, the speaker, and the listener. Are the farmer and theduck both visible or is one or the other invisible from the point ofview of the speaker, and are both placed within the horizon of thespeaker, the listener, or of some indefinite point of reference "offyonder"? In other words, to paraphrase awkwardly certain latent"demonstrative" ideas, does this farmer (invisible to us but standingbehind a door not far away from me, you being seated yonder well out ofreach) kill that duckling (which belongs to you)? or does that farmer(who lives in your neighborhood and whom we see over there) kill thatduckling (that belongs to him)? This type of demonstrative elaborationis foreign to our way of thinking, but it would seem very natural, indeed unavoidable, to a Kwakiutl Indian. 

What, then, are the absolutely essential concepts in speech, theconcepts that must be expressed if language is to be a satisfactorymeans of communication? Clearly we must have, first of all, a largestock of basic or radical concepts, the concrete wherewithal of speech. We must have objects, actions, qualities to talk about, and these musthave their corresponding symbols in independent words or in radicalelements. No proposition, however abstract its intent, is humanlypossible without a tying on at one or more points to the concrete worldof sense. In every intelligible proposition at least two of theseradical ideas must be expressed, though in exceptional cases one or evenboth may be understood from the context. And, secondly, such relationalconcepts must be expressed as moor the concrete concepts to each otherand construct a definite, fundamental form of proposition. In thisfundamental form there must be no doubt as to the nature of therelations that obtain between the concrete concepts. We must know whatconcrete concept is directly or indirectly related to what other, andhow. If we wish to talk of a thing and an action, we must know if theyare coördinately related to each other (e. G. , "He is fond of _wine andgambling_"); or if the thing is conceived of as the starting point, the"doer" of the action, or, as it is customary to say, the "subject" ofwhich the action is predicated; or if, on the contrary, it is the endpoint, the "object" of the action. If I wish to communicate anintelligible idea about a farmer, a duckling, and the act of killing, itis not enough to state the linguistic symbols for these concrete ideasin any order, higgledy-piggledy, trusting that the hearer may constructsome kind of a relational pattern out of the general probabilities ofthe case. The fundamental syntactic relations must be unambiguouslyexpressed. I can afford to be silent on the subject of time and placeand number and of a host of other possible types of concepts, but I canfind no way of dodging the issue as to who is doing the killing. Thereis no known language that can or does dodge it, any more than itsucceeds in saying something without the use of symbols for the concreteconcepts. 

We are thus once more reminded of the distinction between essential orunavoidable relational concepts and the dispensable type. The former areuniversally expressed, the latter are but sparsely developed in somelanguages, elaborated with a bewildering exuberance in others. But whatprevents us from throwing in these "dispensable" or "secondary"relational concepts with the large, floating group of derivational, qualifying concepts that we have already discussed? Is there, after allis said and done, a fundamental difference between a qualifying conceptlike the negative in _unhealthy_ and a relational one like the numberconcept in _books_? If _unhealthy_ may be roughly paraphrased as _nothealthy_, may not _books_ be just as legitimately paraphrased, barringthe violence to English idiom, as _several book?_ There are, indeed, languages in which the plural, if expressed at all, is conceived of inthe same sober, restricted, one might almost say casual, spirit in whichwe feel the negative in _unhealthy_. For such languages the numberconcept has no syntactic significance whatever, is not essentiallyconceived of as defining a relation, but falls into the group ofderivational or even of basic concepts. In English, however, as inFrench, German, Latin, Greek--indeed in all the languages that we havemost familiarity with--the idea of number is not merely appended to agiven concept of a thing. It may have something of this merelyqualifying value, but its force extends far beyond. It infects much elsein the sentence, molding other concepts, even such as have nointelligible relation to number, into forms that are said to correspondto or "agree with" the basic concept to which it is attached in thefirst instance. If "a man falls" but "men fall" in English, it is notbecause of any inherent change that has taken place in the nature of theaction or because the idea of plurality inherent in "men" must, in thevery nature of ideas, relate itself also to the action performed bythese men. What we are doing in these sentences is what most languages, in greater or less degree and in a hundred varying ways, are in thehabit of doing--throwing a bold bridge between the two basicallydistinct types of concept, the concrete and the abstractly relational, infecting the latter, as it were, with the color and grossness of theformer. By a certain violence of metaphor the material concept is forcedto do duty for (or intertwine itself with) the strictly relational. 

The case is even more obvious if we take gender as our text. In the twoEnglish phrases, "The white woman that comes" and "The white men thatcome, " we are not reminded that gender, as well as number, may beelevated into a secondary relational concept. It would seem a littlefar-fetched to make of masculinity and femininity, crassly material, philosophically accidental concepts that they are, a means of relatingquality and person, person and action, nor would it easily occur to us, if we had not studied the classics, that it was anything but absurd toinject into two such highly attenuated relational concepts as areexpressed by "the" and "that" the combined notions of number and sex. Yet all this, and more, happens in Latin. _Illa alba femina quae venit_and _illi albi homines qui veniunt_, conceptually translated, amount tothis: _that_-one-feminine-doer[57] one-feminine-_white_-doerfeminine-doing-one-_woman_ _which_-one-feminine-doerother[58]-one-now-_come_; and: _that_-several-masculine-doerseveral-masculine-_white_-doer masculine-doing-several-_man__which_-several-masculine-doer other-several-now-_come_. Each wordinvolves no less than four concepts, a radical concept (either properlyconcrete--_white_, _man_, _woman_, _come_--or demonstrative--_that_, _which_) and three relational concepts, selected from the categories ofcase, number, gender, person, and tense. Logically, only case[59] (therelation of _woman_ or _men_ to a following verb, of _which_ to itsantecedent, of _that_ and _white_ to _woman_ or _men_, and of _which_ to_come_) imperatively demands expression, and that only in connectionwith the concepts directly affected (there is, for instance, no need tobe informed that the whiteness is a doing or doer's whiteness[60]). Theother relational concepts are either merely parasitic (genderthroughout; number in the demonstrative, the adjective, the relative, and the verb) or irrelevant to the essential syntactic form of thesentence (number in the noun; person; tense). An intelligent andsensitive Chinaman, accustomed as he is to cut to the very bone oflinguistic form, might well say of the Latin sentence, "How pedanticallyimaginative!" It must be difficult for him, when first confronted by theillogical complexities of our European languages, to feel at home in anattitude that so largely confounds the subject-matter of speech with itsformal pattern or, to be more accurate, that turns certain fundamentallyconcrete concepts to such attenuated relational uses. 

[Footnote 57: "Doer, " not "done to. " This is a necessarily clumsy tag torepresent the "nominative" (subjective) in contrast to the "accusative"(objective). ]

[Footnote 58: I. E. , not you or I. ]

[Footnote 59: By "case" is here meant not only the subjective-objectiverelation but also that of attribution. ]

[Footnote 60: Except in so far as Latin uses this method as a ratherawkward, roundabout method of establishing the attribution of the colorto the particular object or person. In effect one cannot in Latindirectly say that a person is white, merely that what is white isidentical with the person who is, acts, or is acted upon in such andsuch a manner. In origin the feel of the Latin _illa alba femina_ isreally "that-one, the-white-one, (namely) the-woman"--three substantiveideas that are related to each other by a juxtaposition intended toconvey an identity. English and Chinese express the attribution directlyby means of order. In Latin the _illa_ and _alba_ may occupy almost anyposition in the sentence. It is important to observe that the subjectiveform of _illa_ and _alba_, does not truly define a relation of thesequalifying concepts to _femina_. Such a relation might be formallyexpressed _via_ an attributive case, say the genitive (_woman ofwhiteness_). In Tibetan both the methods of order and of true caserelation may be employed: _woman white_ (i. E. , "white woman") or_white-of woman_ (i. E. , "woman of whiteness, woman who is white, whitewoman"). ]

I have exaggerated somewhat the concreteness of our subsidiary or rathernon-syntactical relational concepts In order that the essential factsmight come out in bold relief. It goes without saying that a Frenchmanhas no clear sex notion in his mind when he speaks of _un arbre_("a-masculine tree") or of _une pomme_ ("a-feminine apple"). Nor havewe, despite the grammarians, a very vivid sense of the present ascontrasted with all past and all future time when we say _He comes_. [61]This is evident from our use of the present to indicate both future time("He comes to-morrow") and general activity unspecified as to time("Whenever he comes, I am glad to see him, " where "comes" refers to pastoccurrences and possible future ones rather than to present activity). In both the French and English instances the primary ideas of sex andtime have become diluted by form-analogy and by extensions into therelational sphere, the concepts ostensibly indicated being now sovaguely delimited that it is rather the tyranny of usage than the needof their concrete expression that sways us in the selection of this orthat form. If the thinning-out process continues long enough, we mayeventually be left with a system of forms on our hands from which allthe color of life has vanished and which merely persist by inertia, duplicating each other's secondary, syntactic functions with endlessprodigality. Hence, in part, the complex conjugational systems of somany languages, in which differences of form are attended by noassignable differences of function. There must have been a time, forinstance, though it antedates our earliest documentary evidence, whenthe type of tense formation represented by _drove_ or _sank_ differed inmeaning, in however slightly nuanced a degree, from the type (_killed_, _worked_) which has now become established in English as the prevailingpreterit formation, very much as we recognize a valuable distinction atpresent between both these types and the "perfect" (_has driven, haskilled_) but may have ceased to do so at some point in the future. [62]Now form lives longer than its own conceptual content. Both areceaselessly changing, but, on the whole, the form tends to linger onwhen the spirit has flown or changed its being. Irrational form, formfor form's sake--however we term this tendency to hold on to formaldistinctions once they have come to be--is as natural to the life oflanguage as is the retention of modes of conduct that have long outlivedthe meaning they once had. 

[Footnote 61: Aside, naturally, from the life and imminence that may becreated for such a sentence by a particular context. ]

[Footnote 62: This has largely happened in popular French and German, where the difference is stylistic rather than functional. The preteritsare more literary or formal in tone than the perfects. ]

There is another powerful tendency which makes for a formal elaborationthat does not strictly correspond to clear-cut conceptual differences. This is the tendency to construct schemes of classification into whichall the concepts of language must be fitted. Once we have made up ourminds that all things are either definitely good or bad or definitelyblack or white, it is difficult to get into the frame of mind thatrecognizes that any particular thing may be both good and bad (in otherwords, indifferent) or both black and white (in other words, gray), still more difficult to realize that the good-bad or black-whitecategories may not apply at all. Language is in many respects asunreasonable and stubborn about its classifications as is such a mind. It must have its perfectly exclusive pigeon-holes and will tolerate noflying vagrants. Any concept that asks for expression must submit to theclassificatory rules of the game, just as there are statistical surveysin which even the most convinced atheist must perforce be labeledCatholic, Protestant, or Jew or get no hearing. In English we have madeup our minds that all action must be conceived of in reference to threestandard times. If, therefore, we desire to state a proposition that isas true to-morrow as it was yesterday, we have to pretend that thepresent moment may be elongated fore and aft so as to take in alleternity. [63] In French we know once for all that an object is masculineor feminine, whether it be living or not; just as in many American andEast Asiatic languages it must be understood to belong to a certainform-category (say, ring-round, ball-round, long and slender, cylindrical, sheet-like, in mass like sugar) before it can be enumerated(e. G. , "two ball-class potatoes, " "three sheet-class carpets") or evensaid to "be" or "be handled in a definite way" (thus, in the Athabaskanlanguages and in Yana, "to carry" or "throw" a pebble is quite anotherthing than to carry or throw a log, linguistically no less than in termsof muscular experience). Such instances might be multiplied at will. Itis almost as though at some period in the past the unconscious mind ofthe race had made a hasty inventory of experience, committed itself to apremature classification that allowed of no revision, and saddled theinheritors of its language with a science that they no longer quitebelieved in nor had the strength to overthrow. Dogma, rigidly prescribedby tradition, stiffens into formalism. Linguistic categories make up asystem of surviving dogma--dogma of the unconscious. They are often buthalf real as concepts; their life tends ever to languish away into formfor form's sake. 

[Footnote 63: Hence, "the square root of 4 _is_ 2, " precisely as "myuncle _is_ here now. " There are many "primitive" languages that are morephilosophical and distinguish between a true "present" and a "customary"or "general" tense. ]

There is still a third cause for the rise of this non-significant form, or rather of non-significant differences of form. This is the mechanicaloperation of phonetic processes, which may bring about formaldistinctions that have not and never had a corresponding functionaldistinction. Much of the irregularity and general formal complexity ofour declensional and conjugational systems is due to this process. Theplural of _hat_ is _hats_, the plural of _self_ is _selves_. In theformer case we have a true _-s_ symbolizing plurality, in the latter a_z_-sound coupled with a change in the radical element of the word of_f_ to _v_. Here we have not a falling together of forms thatoriginally stood for fairly distinct concepts--as we saw was presumablythe case with such parallel forms as _drove_ and _worked_--but a merelymechanical manifolding of the same formal element without acorresponding growth of a new concept. This type of form development, therefore, while of the greatest interest for the general history oflanguage, does not directly concern us now in our effort to understandthe nature of grammatical concepts and their tendency to degenerate intopurely formal counters. 

We may now conveniently revise our first classification of concepts asexpressed in language and suggest the following scheme:

 I. _Basic (Concrete) Concepts_ (such as objects, actions, qualities): normally expressed by independent words or radical elements; involve no relation as such[64]

 II. _Derivational Concepts_ (less concrete, as a rule, than I, more so than III): normally expressed by affixing non-radical elements to radical elements or by inner modification of these; differ from type I in defining ideas that are irrelevant to the proposition as a whole but that give a radical element a particular increment of significance and that are thus inherently related in a specific way to concepts of type I[65]

III. _Concrete Relational Concepts_ (still more abstract, yet not entirely devoid of a measure of concreteness): normally expressed by affixing non-radical elements to radical elements, but generally at a greater remove from these than is the case with elements of type II, or by inner modification of radical elements; differ fundamentally from type II in indicating or implying relations that transcend the particular word to which they are immediately attached, thus leading over to

 IV. _Pure Relational Concepts_ (purely abstract): normally expressed by affixing non-radical elements to radical elements (in which case these concepts are frequently intertwined with those of type III) or by their inner modification, by independent words, or by position; serve to relate the concrete elements of the proposition to each other, thus giving it definite syntactic form. 

[Footnote 64: Except, of course, the fundamental selection and contrastnecessarily implied in defining one concept as against another. "Man"and "white" possess an inherent relation to "woman" and "black, " but itis a relation of conceptual content only and is of no direct interest togrammar. ]

[Footnote 65: Thus, the _-er_ of _farmer_ may he defined as indicatingthat particular substantive concept (object or thing) that serves as thehabitual subject of the particular verb to which it is affixed. Thisrelation of "subject" (_a farmer farms_) is inherent in and specific tothe word; it does not exist for the sentence as a whole. In the same waythe _-ling_ of _duckling_ defines a specific relation of attributionthat concerns only the radical element, not the sentence. ]

The nature of these four classes of concepts as regards theirconcreteness or their power to express syntactic relations may be thussymbolized: _ Material _/ I. Basic Concepts Content \_ II. Derivational Concepts _ Relation _/ III. Concrete Relational Concepts \_ IV. Pure Relational Concepts

These schemes must not be worshipped as fetiches. In the actual work ofanalysis difficult problems frequently arise and we may well be in doubtas to how to group a given set of concepts. This is particularly apt tobe the case in exotic languages, where we may be quite sure of theanalysis of the words in a sentence and yet not succeed in acquiringthat inner "feel" of its structure that enables us to tell infalliblywhat is "material content" and what is "relation. " Concepts of class Iare essential to all speech, also concepts of class IV. Concepts II andIII are both common, but not essential; particularly group III, whichrepresents, in effect, a psychological and formal confusion of types IIand IV or of types I and IV, is an avoidable class of concepts. Logically there is an impassable gulf between I and IV, but theillogical, metaphorical genius of speech has wilfully spanned the gulfand set up a continuous gamut of concepts and forms that leadsimperceptibly from the crudest of materialities ("house" or "JohnSmith") to the most subtle of relations. It is particularly significantthat the unanalyzable independent word belongs in most cases to eithergroup I or group IV, rather less commonly to II or III. It is possiblefor a concrete concept, represented by a simple word, to lose itsmaterial significance entirely and pass over directly into therelational sphere without at the same time losing its independence as aword. This happens, for instance, in Chinese and Cambodgian when theverb "give" is used in an abstract sense as a mere symbol of the"indirect objective" relation (e. G. , Cambodgian "We make story this giveall that person who have child, " i. E. , "We have made this story _for_all those that have children"). 

There are, of course, also not a few instances of transitions betweengroups I and II and I and III, as well as of the less radical onebetween II and III. To the first of these transitions belongs that wholeclass of examples in which the independent word, after passing throughthe preliminary stage of functioning as the secondary or qualifyingelement in a compound, ends up by being a derivational affix pure andsimple, yet without losing the memory of its former independence. Suchan element and concept is the _full_ of _teaspoonfull_, which hoverspsychologically between the status of an independent, radical concept(compare _full_) or of a subsidiary element in a compound (cf. _brim-full_) and that of a simple suffix (cf. _dutiful_) in which theprimary concreteness is no longer felt. In general, the more highlysynthetic our linguistic type, the more difficult and even arbitrary itbecomes to distinguish groups I and II. 

Not only is there a gradual loss of the concrete as we pass through fromgroup I to group IV, there is also a constant fading away of the feelingof sensible reality within the main groups of linguistic conceptsthemselves. In many languages it becomes almost imperative, therefore, to make various sub-classifications, to segregate, for instance, themore concrete from the more abstract concepts of group II. Yet we mustalways beware of reading into such abstracter groups that purely formal, relational feeling that we can hardly help associating with certain ofthe abstracter concepts which, with us, fall in group III, unless, indeed, there is clear evidence to warrant such a reading in. An exampleor two should make clear these all-important distinctions. [66] In Nootkawe have an unusually large number of derivational affixes (expressingconcepts of group II). Some of these are quite material in content(e. G. , "in the house, " "to dream of"), others, like an element denotingplurality and a diminutive affix, are far more abstract in content. Theformer type are more closely welded with the radical element than thelatter, which can only be suffixed to formations that have the value ofcomplete words. If, therefore, I wish to say "the small fires in thehouse"--and I can do this in one word--I must form the word"fire-in-the-house, " to which elements corresponding to "small, " ourplural, and "the" are appended. The element indicating the definitenessof reference that is implied in our "the" comes at the very end of theword. So far, so good. "Fire-in-the-house-the" is an intelligiblecorrelate of our "the house-fire. "[67] But is the Nootka correlate of"the small fires in the house" the true equivalent of an English "_thehouse-firelets_"?[68] By no means. First of all, the plural elementprecedes the diminutive in Nootka: "fire-in-the-house-plural-small-the, "in other words "the house-fires-let, " which at once reveals theimportant fact that the plural concept is not as abstractly, asrelationally, felt as in English. A more adequate rendering would be"the house-fire-several-let, " in which, however, "several" is too grossa word, "-let" too choice an element ("small" again is too gross). Intruth we cannot carry over into English the inherent feeling of theNootka word, which seems to hover somewhere between "the house-firelets"and "the house-fire-several-small. " But what more than anything elsecuts off all possibility of comparison between the English _-s_ of"house-firelets" and the "-several-small" of the Nootka word is this, that in Nootka neither the plural nor the diminutive affix correspondsor refers to anything else in the sentence. In English "thehouse-firelets burn" (not "burns"), in Nootka neither verb, noradjective, nor anything else in the proposition is in the leastconcerned with the plurality or the diminutiveness of the fire. Hence, while Nootka recognizes a cleavage between concrete and less concreteconcepts within group II, the less concrete do not transcend the groupand lead us into that abstracter air into which our plural _-s_ carriesus. But at any rate, the reader may object, it is something that theNootka plural affix is set apart from the concreter group of affixes;and may not the Nootka diminutive have a slenderer, a more elusivecontent than our _-let_ or _-ling_ or the German _-chen_ or _-lein?_[69]

[Footnote 66: It is precisely the failure to feel the "value" or "tone, "as distinct from the outer significance, of the concept expressed by agiven grammatical element that has so often led students tomisunderstand the nature of languages profoundly alien to their own. Noteverything that calls itself "tense" or "mode" or "number" or "gender"or "person" is genuinely comparable to what we mean by these terms inLatin or French. ]

[Footnote 67: Suffixed articles occur also in Danish and Swedish and innumerous other languages. The Nootka element for "in the house" differsfrom our "house-" in that it is suffixed and cannot occur as anindependent word; nor is it related to the Nootka word for "house. "]

[Footnote 68: Assuming the existence of a word "firelet. "]

[Footnote 69: The Nootka diminutive is doubtless more of afeeling-element, an element of nuance, than our _-ling_. This is shownby the fact that it may be used with verbs as well as with nouns. Inspeaking to a child, one is likely to add the diminutive to any word inthe sentence, regardless of whether there is an inherent diminutivemeaning in the word or not. ]

Can such a concept as that of plurality ever be classified with the morematerial concepts of group II? Indeed it can be. In Yana the thirdperson of the verb makes no formal distinction between singular andplural. Nevertheless the plural concept can be, and nearly always is, expressed by the suffixing of an element (_-ba-_) to the radical elementof the verb. "It burns in the east" is rendered by the verb _ya-hau-si_"burn-east-s. "[70] "They burn in the east" is _ya-ba-hau-si_. Note thatthe plural affix immediately follows the radical element (_ya-_), disconnecting it from the local element (_-hau-_). It needs no laboredargument to prove that the concept of plurality is here hardly lessconcrete than that of location "in the east, " and that the Yana formcorresponds in feeling not so much to our "They burn in the east"(_ardunt oriente_) as to a "Burn-several-east-s, it plurally burns inthe east, " an expression which we cannot adequately assimilate for lackof the necessary form-grooves into which to run it. 

[Footnote 70: _-si_ is the third person of the present tense. _-hau-_"east" is an affix, not a compounded radical element. ]

But can we go a step farther and dispose of the category of plurality asan utterly material idea, one that would make of "books" a "pluralbook, " in which the "plural, " like the "white" of "white book, " fallscontentedly into group I? Our "many books" and "several books" areobviously not cases in point. Even if we could say "many book" and"several book" (as we can say "many a book" and "each book"), the pluralconcept would still not emerge as clearly as it should for our argument;"many" and "several" are contaminated by certain notions of quantity orscale that are not essential to the idea of plurality itself. We mustturn to central and eastern Asia for the type of expression we areseeking. In Tibetan, for instance, _nga-s mi mthong_[71] "I-by man see, by me a man is seen, I see a man" may just as well be understood to mean"I see men, " if there happens to be no reason to emphasize the fact ofplurality. [72] If the fact is worth expressing, however, I can say_nga-s mi rnams mthong_ "by me man plural see, " where _rnams_ is theperfect conceptual analogue of _-s_ in _books_, divested of allrelational strings. _Rnams_ follows its noun as would any otherattributive word--"man plural" (whether two or a million) like "manwhite. " No need to bother about his plurality any more than about hiswhiteness unless we insist on the point. 

[Footnote 71: These are classical, not modern colloquial, forms. ]

[Footnote 72: Just as in English "He has written books" makes nocommitment on the score of quantity ("a few, several, many"). ]

What is true of the idea of plurality is naturally just as true of agreat many other concepts. They do not necessarily belong where we whospeak English are in the habit of putting them. They may be shiftedtowards I or towards IV, the two poles of linguistic expression. Nordare we look down on the Nootka Indian and the Tibetan for theirmaterial attitude towards a concept which to us is abstract andrelational, lest we invite the reproaches of the Frenchman who feels asubtlety of relation in _femme blanche_ and _homme blanc_ that he missesin the coarser-grained _white woman_ and _white man_. But the BantuNegro, were he a philosopher, might go further and find it strange thatwe put in group II a category, the diminutive, which he strongly feelsto belong to group III and which he uses, along with a number of otherclassificatory concepts, [73] to relate his subjects and objects, attributes and predicates, as a Russian or a German handles his gendersand, if possible, with an even greater finesse. 

[Footnote 73: Such as person class, animal class, instrument class, augmentative class. ]

It is because our conceptual scheme is a sliding scale rather than aphilosophical analysis of experience that we cannot say in advance justwhere to put a given concept. We must dispense, in other words, with awell-ordered classification of categories. What boots it to put tenseand mode here or number there when the next language one handles putstense a peg "lower down" (towards I), mode and number a peg "higher up"(towards IV)? Nor is there much to be gained in a summary work of thiskind from a general inventory of the types of concepts generally foundin groups II, III, and IV. There are too many possibilities. It would beinteresting to show what are the most typical noun-forming andverb-forming elements of group II; how variously nouns may be classified(by gender; personal and non-personal; animate and inanimate; by form;common and proper); how the concept of number is elaborated (singularand plural; singular, dual, and plural; singular, dual, trial, andplural; single, distributive, and collective); what tense distinctionsmay be made in verb or noun (the "past, " for instance, may be anindefinite past, immediate, remote, mythical, completed, prior); howdelicately certain languages have developed the idea of "aspect"[74](momentaneous, durative, continuative, inceptive, cessative, durative-inceptive, iterative, momentaneous-iterative, durative-iterative, resultative, and still others); what modalities maybe recognized (indicative, imperative, potential, dubitative, optative, negative, and a host of others[75]); what distinctions of person arepossible (is "we, " for instance, conceived of as a plurality of "I" oris it as distinct from "I" as either is from "you" or "he"?--bothattitudes are illustrated in language; moreover, does "we" include youto whom I speak or not?--"inclusive" and "exclusive" forms); what may bethe general scheme of orientation, the so-called demonstrativecategories ("this" and "that" in an endless procession of nuances);[76]how frequently the form expresses the source or nature of the speaker'sknowledge (known by actual experience, by hearsay, [77] by inference);how the syntactic relations may be expressed in the noun (subjective andobjective; agentive, instrumental, and person affected;[78] varioustypes of "genitive" and indirect relations) and, correspondingly, in theverb (active and passive; active and static; transitive andintransitive; impersonal, reflexive, reciprocal, indefinite as toobject, and many other special limitations on the starting-point andend-point of the flow of activity). These details, important as many ofthem are to an understanding of the "inner form" of language, yield ingeneral significance to the more radical group-distinctions that we haveset up. It is enough for the general reader to feel that languagestruggles towards two poles of linguistic expression--material contentand relation--and that these poles tend to be connected by a long seriesof transitional concepts. 

[Footnote 74: A term borrowed from Slavic grammar. It indicates thelapse of action, its nature from the standpoint of continuity. Our "cry"is indefinite as to aspect, "be crying" is durative, "cry put" ismomentaneous, "burst into tears" is inceptive, "keep crying" iscontinuative, "start in crying" is durative-inceptive, "cry now andagain" is iterative, "cry out every now and then" or "cry in fits andstarts" is momentaneous-iterative. "To put on a coat" is momentaneous, "to wear a coat" is resultative. As our examples show, aspect isexpressed in English by all kinds of idiomatic turns rather than by aconsistently worked out set of grammatical forms. In many languagesaspect is of far greater formal significance than tense, with which thenaive student is apt to confuse it. ]

[Footnote 75: By "modalities" I do not mean the matter of factstatement, say, of negation or uncertainty as such, rather theirimplication in terms of form. There are languages, for instance, whichhave as elaborate an apparatus of negative forms for the verb as Greekhas of the optative or wish-modality. ]

[Footnote 76: Compare page 97. ]

[Transcriber's note: Footnote 76 refers to the paragraph beginning online 2948. ]

[Footnote 77: It is because of this classification of experience that inmany languages the verb forms which are proper, say, to a mythicalnarration differ from those commonly used in daily intercourse. We leavethese shades to the context or content ourselves with a more explicitand roundabout mode of expression, e. G. , "He is dead, as I happen toknow, " "They say he is dead, " "He must be dead by the looks of things. "]

[Footnote 78: We say "_I_ sleep" and "_I_ go, " as well as "_I_ killhim, " but "he kills _me_. " Yet _me_ of the last example is at least asclose psychologically to _I_ of "I sleep" as is the latter to _I_ of "Ikill him. " It is only by form that we can classify the "I" notion of "Isleep" as that of an acting subject. Properly speaking, I am handled byforces beyond my control when I sleep just as truly as when some one iskilling me. Numerous languages differentiate clearly between activesubject and static subject (_I go_ and _I kill him_ as distinct from _Isleep_, _I am good_, _I am killed_) or between transitive subject andintransitive subject (_I kill him_ as distinct from _I sleep_, _I amgood_, _I am killed_, _I go_). The intransitive or static subjects mayor may not be identical with the object of the transitive verb. ]

In dealing with words and their varying forms we have had to anticipatemuch that concerns the sentence as a whole. Every language has itsspecial method or methods of binding words into a larger unity. Theimportance of these methods is apt to vary with the complexity of theindividual word. The more synthetic the language, in other words, themore clearly the status of each word in the sentence is indicated by itsown resources, the less need is there for looking beyond the word to thesentence as a whole. The Latin _agit_ "(he) acts" needs no outside helpto establish its place in a proposition. Whether I say _agit dominus_"the master acts" or _sic femina agit_ "thus the woman acts, " the netresult as to the syntactic feel of the _agit_ is practically the same. It can only be a verb, the predicate of a proposition, and it can onlybe conceived as a statement of activity carried out by a person (orthing) other than you or me. It is not so with such a word as theEnglish _act_. _Act_ is a syntactic waif until we have defined itsstatus in a proposition--one thing in "they act abominably, " quiteanother in "that was a kindly act. " The Latin sentence speaks with theassurance of its individual members, the English word needs theprompting of its fellows. Roughly speaking, to be sure. And yet to saythat a sufficiently elaborate word-structure compensates for externalsyntactic methods is perilously close to begging the question. Theelements of the word are related to each other in a specific way andfollow each other in a rigorously determined sequence. This istantamount to saying that a word which consists of more than a radicalelement is a crystallization of a sentence or of some portion of asentence, that a form like _agit_ is roughly the psychological[79]equivalent of a form like _age is_ "act he. " Breaking down, then, thewall that separates word and sentence, we may ask: What, at lastanalysis, are the fundamental methods of relating word to word andelement to element, in short, of passing from the isolated notionssymbolized by each word and by each element to the unified propositionthat corresponds to a thought?

[Footnote 79: Ultimately, also historical--say, _age to_ "act that(one). "]

The answer is simple and is implied in the preceding remarks. The mostfundamental and the most powerful of all relating methods is the methodof order. Let us think of some more or less concrete idea, say a color, and set down its symbol--_red_; of another concrete idea, say a personor object, setting down its symbol--_dog_; finally, of a third concreteidea, say an action, setting down its symbol--_run_. It is hardlypossible to set down these three symbols--_red dog run_--withoutrelating them in some way, for example _(the) red dog run(s)_. I am farfrom wishing to state that the proposition has always grown up in thisanalytic manner, merely that the very process of juxtaposing concept toconcept, symbol to symbol, forces some kind of relational "feeling, " ifnothing else, upon us. To certain syntactic adhesions we are verysensitive, for example, to the attributive relation of quality (_reddog_) or the subjective relation (_dog run_) or the objective relation(_kill dog_), to others we are more indifferent, for example, to theattributive relation of circumstance (_to-day red dog run_ or _red dogto-day run_ or _red dog run to-day_, all of which are equivalentpropositions or propositions in embryo). Words and elements, then, oncethey are listed in a certain order, tend not only to establish some kindof relation among themselves but are attracted to each other in greateror in less degree. It is presumably this very greater or less thatultimately leads to those firmly solidified groups of elements (radicalelement or elements plus one or more grammatical elements) that we havestudied as complex words. They are in all likelihood nothing butsequences that have shrunk together and away from other sequences orisolated elements in the flow of speech. While they are fully alive, inother words, while they are functional at every point, they can keepthemselves at a psychological distance from their neighbors. As theygradually lose much of their life, they fall back into the embrace ofthe sentence as a whole and the sequence of independent words regainsthe importance it had in part transferred to the crystallized groups ofelements. Speech is thus constantly tightening and loosening itssequences. In its highly integrated forms (Latin, Eskimo) the "energy"of sequence is largely locked up in complex word formations, it becomestransformed into a kind of potential energy that may not be released formillennia. In its more analytic forms (Chinese, English) this energy ismobile, ready to hand for such service as we demand of it. 

There can be little doubt that stress has frequently played acontrolling influence in the formation of element-groups or complexwords out of certain sequences in the sentence. Such an English word as_withstand_ is merely an old sequence _with stand_, i. E. , "against[80]stand, " in which the unstressed adverb was permanently drawn to thefollowing verb and lost its independence as a significant element. Inthe same way French futures of the type _irai_ "(I) shall go" are butthe resultants of a coalescence of originally independent words: _ir[81]a'i_ "to-go I-have, " under the influence of a unifying accent. Butstress has done more than articulate or unify sequences that in theirown right imply a syntactic relation. Stress is the most natural meansat our disposal to emphasize a linguistic contrast, to indicate themajor element in a sequence. Hence we need not be surprised to find thataccent too, no less than sequence, may serve as the unaided symbol ofcertain relations. Such a contrast as that of _go' between_ ("one whogoes between") and _to go between'_ may be of quite secondary origin inEnglish, but there is every reason to believe that analogousdistinctions have prevailed at all times in linguistic history. Asequence like _see' man_ might imply some type of relation in which_see_ qualifies the following word, hence "a seeing man" or "a seen (orvisible) man, " or is its predication, hence "the man sees" or "the manis seen, " while a sequence like _see man'_ might indicate that theaccented word in some way limits the application of the first, say asdirect object, hence "to see a man" or "(he) sees the man. " Suchalternations of relation, as symbolized by varying stresses, areimportant and frequent in a number of languages. [82]

[Footnote 80: For _with_ in the sense of "against, " compare German_wider_ "against. "]

[Footnote 81: Cf. Latin _ire_ "to go"; also our English idiom "I have togo, " i. E. , "must go. "]

[Footnote 82: In Chinese no less than in English. ]

It is a somewhat venturesome and yet not an altogether unreasonablespeculation that sees in word order and stress the primary methods forthe expression of all syntactic relations and looks upon the presentrelational value of specific words and elements as but a secondarycondition due to a transfer of values. Thus, we may surmise that theLatin _-m_ of words like _feminam_, _dominum_, and _civem_ did notoriginally[83] denote that "woman, " "master, " and "citizen" wereobjectively related to the verb of the proposition but indicatedsomething far more concrete, [84] that the objective relation was merelyimplied by the position or accent of the word (radical element)immediately preceding the _-m_, and that gradually, as its more concretesignificance faded away, it took over a syntactic function that did notoriginally belong to it. This sort of evolution by transfer is traceablein many instances. Thus, the _of_ in an English phrase like "the law ofthe land" is now as colorless in content, as purely a relationalindicator as the "genitive" suffix _-is_ in the Latin _lex urbis_ "thelaw of the city. " We know, however, that it was originally an adverb ofconsiderable concreteness of meaning, [85] "away, moving from, " and thatthe syntactic relation was originally expressed by the case form[86] ofthe second noun. As the case form lost its vitality, the adverb tookover its function. If we are actually justified in assuming that theexpression of all syntactic relations is ultimately traceable to thesetwo unavoidable, dynamic features of speech--sequence and stress[87]--aninteresting thesis results:--All of the actual content of speech, itsclusters of vocalic and consonantal sounds, is in origin limited to theconcrete; relations were originally not expressed in outward form butwere merely implied and articulated with the help of order and rhythm. In other words, relations were intuitively felt and could only "leakout" with the help of dynamic factors that themselves move on anintuitional plane. 

[Footnote 83: By "originally" I mean, of course, some time antedatingthe earliest period of the Indo-European languages that we can get at bycomparative evidence. ]

[Footnote 84: Perhaps it was a noun-classifying element of some sort. ]

[Footnote 85: Compare its close historical parallel _off_. ]

[Footnote 86: "Ablative" at last analysis. ]

[Footnote 87: Very likely pitch should be understood along with stress. ]

There is a special method for the expression of relations that has beenso often evolved in the history of language that we must glance at itfor a moment. This is the method of "concord" or of like signaling. Itis based on the same principle as the password or label. All persons orobjects that answer to the same counter-sign or that bear the sameimprint are thereby stamped as somehow related. It makes littledifference, once they are so stamped, where they are to be found or howthey behave themselves. They are known to belong together. We arefamiliar with the principle of concord in Latin and Greek. Many of ushave been struck by such relentless rhymes as _vidi ilium bonum dominum_"I saw that good master" or _quarum dearum saevarum_ "of which sterngoddesses. " Not that sound-echo, whether in the form of rhyme or ofalliteration[88] is necessary to concord, though in its most typical andoriginal forms concord is nearly always accompanied by sound repetition. The essence of the principle is simply this, that words (elements) thatbelong together, particularly if they are syntactic equivalents or arerelated in like fashion to another word or element, are outwardly markedby the same or functionally equivalent affixes. The application of theprinciple varies considerably according to the genius of the particularlanguage. In Latin and Greek, for instance, there is concord betweennoun and qualifying word (adjective or demonstrative) as regards gender, number, and case, between verb and subject only as regards number, andno concord between verb and object. 

[Footnote 88: As in Bantu or Chinook. ]

In Chinook there is a more far-reaching concord between noun, whethersubject or object, and verb. Every noun is classified according to fivecategories--masculine, feminine, neuter, [89] dual, and plural. "Woman"is feminine, "sand" is neuter, "table" is masculine. If, therefore, Iwish to say "The woman put the sand on the table, " I must place in theverb certain class or gender prefixes that accord with correspondingnoun prefixes. The sentence reads then, "The (fem. )-woman she (fem. )-it(neut. )-it (masc. )-on-put the (neut. )-sand the (masc. )-table. " If "sand"is qualified as "much" and "table" as "large, " these new ideas areexpressed as abstract nouns, each with its inherent class-prefix ("much"is neuter or feminine, "large" is masculine) and with a possessiveprefix referring to the qualified noun. Adjective thus calls to noun, noun to verb. "The woman put much sand on the large table, " therefore, takes the form: "The (fem. )-woman she (fem. )-it (neut. )-it(masc. )-on-put the (fem. )-thereof (neut. )-quantity the (neut. )-sand the(masc. )-thereof (masc. )-largeness the (masc. )-table. " The classificationof "table" as masculine is thus three times insisted on--in the noun, inthe adjective, and in the verb. In the Bantu languages, [90] theprinciple of concord works very much as in Chinook. In them also nounsare classified into a number of categories and are brought into relationwith adjectives, demonstratives, relative pronouns, and verbs by meansof prefixed elements that call off the class and make up a complexsystem of concordances. In such a sentence as "That fierce lion who camehere is dead, " the class of "lion, " which we may call the animal class, would be referred to by concording prefixes no less than sixtimes, --with the demonstrative ("that"), the qualifying adjective, thenoun itself, the relative pronoun, the subjective prefix to the verb ofthe relative clause, and the subjective prefix to the verb of the mainclause ("is dead"). We recognize in this insistence on external clarityof reference the same spirit as moves in the more familiar _illum bonumdominum_. 

[Footnote 89: Perhaps better "general. " The Chinook "neuter" may referto persons as well as things and may also be used as a plural. "Masculine" and "feminine, " as in German and French, include a greatnumber of inanimate nouns. ]

[Footnote 90: Spoken in the greater part of the southern half of Africa. Chinook is spoken in a number of dialects in the lower Columbia Rivervalley. It is impressive to observe how the human mind has arrived atthe same form of expression in two such historically unconnectedregions. ]

Psychologically the methods of sequence and accent lie at the oppositepole to that of concord. Where they are all for implication, forsubtlety of feeling, concord is impatient of the least ambiguity butmust have its well-certificated tags at every turn. Concord tends todispense with order. In Latin and Chinook the independent words are freein position, less so in Bantu. In both Chinook and Bantu, however, themethods of concord and order are equally important for thedifferentiation of subject and object, as the classifying verb prefixesrefer to subject, object, or indirect object according to the relativeposition they occupy. These examples again bring home to us thesignificant fact that at some point or other order asserts itself inevery language as the most fundamental of relating principles. 

The observant reader has probably been surprised that all this time wehave had so little to say of the time-honored "parts of speech. " Thereason for this is not far to seek. Our conventional classification ofwords into parts of speech is only a vague, wavering approximation to aconsistently worked out inventory of experience. We imagine, to beginwith, that all "verbs" are inherently concerned with action as such, that a "noun" is the name of some definite object or personality thatcan be pictured by the mind, that all qualities are necessarilyexpressed by a definite group of words to which we may appropriatelyapply the term "adjective. " As soon as we test our vocabulary, wediscover that the parts of speech are far from corresponding to sosimple an analysis of reality. We say "it is red" and define "red" as aquality-word or adjective. We should consider it strange to think of anequivalent of "is red" in which the whole predication (adjective andverb of being) is conceived of as a verb in precisely the same way inwhich we think of "extends" or "lies" or "sleeps" as a verb. Yet as soonas we give the "durative" notion of being red an inceptive ortransitional turn, we can avoid the parallel form "it becomes red, itturns red" and say "it reddens. " No one denies that "reddens" is as gooda verb as "sleeps" or even "walks. " Yet "it is red" is related to "itreddens" very much as is "he stands" to "he stands up" or "he rises. " Itis merely a matter of English or of general Indo-European idiom that wecannot say "it reds" in the sense of "it is red. " There are hundreds oflanguages that can. Indeed there are many that can express what weshould call an adjective only by making a participle out of a verb. "Red" in such languages is merely a derivative "being red, " as our"sleeping" or "walking" are derivatives of primary verbs. 

Just as we can verbify the idea of a quality in such cases as "reddens, "so we can represent a quality or an action to ourselves as a thing. Wespeak of "the height of a building" or "the fall of an apple" quite asthough these ideas were parallel to "the roof of a building" or "theskin of an apple, " forgetting that the nouns (_height_, _fall_) have notceased to indicate a quality and an act when we have made them speakwith the accent of mere objects. And just as there are languages thatmake verbs of the great mass of adjectives, so there are others thatmake nouns of them. In Chinook, as we have seen, "the big table" is"the-table its-bigness"; in Tibetan the same idea may be expressed by"the table of bigness, " very much as we may say "a man of wealth"instead of "a rich man. "

But are there not certain ideas that it is impossible to render exceptby way of such and such parts of speech? What can be done with the "to"of "he came to the house"? Well, we can say "he reached the house" anddodge the preposition altogether, giving the verb a nuance that absorbsthe idea of local relation carried by the "to. " But let us insist ongiving independence to this idea of local relation. Must we not thenhold to the preposition? No, we can make a noun of it. We can saysomething like "he reached the proximity of the house" or "he reachedthe house-locality. " Instead of saying "he looked into the glass" we maysay "he scrutinized the glass-interior. " Such expressions are stilted inEnglish because they do not easily fit into our formal grooves, but inlanguage after language we find that local relations are expressed injust this way. The local relation is nominalized. And so we might go onexamining the various parts of speech and showing how they not merelygrade into each other but are to an astonishing degree actuallyconvertible into each other. The upshot of such an examination would beto feel convinced that the "part of speech" reflects not so much ourintuitive analysis of reality as our ability to compose that realityinto a variety of formal patterns. A part of speech outside of thelimitations of syntactic form is but a will o' the wisp. For this reasonno logical scheme of the parts of speech--their number, nature, andnecessary confines--is of the slightest interest to the linguist. Eachlanguage has its own scheme. Everything depends on the formaldemarcations which it recognizes. 

Yet we must not be too destructive. It is well to remember that speechconsists of a series of propositions. There must be something to talkabout and something must be said about this subject of discourse once itis selected. This distinction is of such fundamental importance that thevast majority of languages have emphasized it by creating some sort offormal barrier between the two terms of the proposition. The subject ofdiscourse is a noun. As the most common subject of discourse is either aperson or a thing, the noun clusters about concrete concepts of thatorder. As the thing predicated of a subject is generally an activity inthe widest sense of the word, a passage from one moment of existence toanother, the form which has been set aside for the business ofpredicating, in other words, the verb, clusters about concepts ofactivity. No language wholly fails to distinguish noun and verb, thoughin particular cases the nature of the distinction may be an elusive one. It is different with the other parts of speech. Not one of them isimperatively required for the life of language. [91]

[Footnote 91: In Yana the noun and the verb are well distinct, thoughthere are certain features that they hold in common which tend to drawthem nearer to each other than we feel to be possible. But there are, strictly speaking, no other parts of speech. The adjective is a verb. Soare the numeral, the interrogative pronoun (e. G. , "to be what?"), andcertain "conjunctions" and adverbs (e. G. , "to be and" and "to be not";one says "and-past-I go, " i. E. , "and I went"). Adverbs and prepositionsare either nouns or merely derivative affixes in the verb. ]

VI

TYPES OF LINGUISTIC STRUCTURE

So far, in dealing with linguistic form, we have been concerned onlywith single words and with the relations of words in sentences. We havenot envisaged whole languages as conforming to this or that generaltype. Incidentally we have observed that one language runs to tight-knitsynthesis where another contents itself with a more analytic, piece-mealhandling of its elements, or that in one language syntactic relationsappear pure which in another are combined with certain other notionsthat have something concrete about them, however abstract they may befelt to be in practice. In this way we may have obtained some inkling ofwhat is meant when we speak of the general form of a language. For itmust be obvious to any one who has thought about the question at all orwho has felt something of the spirit of a foreign language that there issuch a thing as a basic plan, a certain cut, to each language. This typeor plan or structural "genius" of the language is something much morefundamental, much more pervasive, than any single feature of it that wecan mention, nor can we gain an adequate idea of its nature by a mererecital of the sundry facts that make up the grammar of the language. When we pass from Latin to Russian, we feel that it is approximately thesame horizon that bounds our view, even though the near, familiarlandmarks have changed. When we come to English, we seem to notice thatthe hills have dipped down a little, yet we recognize the general layof the land. And when we have arrived at Chinese, it is an utterlydifferent sky that is looking down upon us. We can translate thesemetaphors and say that all languages differ from one another but thatcertain ones differ far more than others. This is tantamount to sayingthat it is possible to group them into morphological types. 

Strictly speaking, we know in advance that it is impossible to set up alimited number of types that would do full justice to the peculiaritiesof the thousands of languages and dialects spoken on the surface of theearth. Like all human institutions, speech is too variable and tooelusive to be quite safely ticketed. Even if we operate with a minutelysubdivided scale of types, we may be quite certain that many of ourlanguages will need trimming before they fit. To get them into thescheme at all it will be necessary to overestimate the significance ofthis or that feature or to ignore, for the time being, certaincontradictions in their mechanism. Does the difficulty of classificationprove the uselessness of the task? I do not think so. It would be tooeasy to relieve ourselves of the burden of constructive thinking and totake the standpoint that each language has its unique history, thereforeits unique structure. Such a standpoint expresses only a half truth. Just as similar social, economic, and religious institutions have grownup in different parts of the world from distinct historical antecedents, so also languages, traveling along different roads, have tended toconverge toward similar forms. Moreover, the historical study oflanguage has proven to us beyond all doubt that a language changes notonly gradually but consistently, that it moves unconsciously from onetype towards another, and that analogous trends are observable inremote quarters of the globe. From this it follows that broadly similarmorphologies must have been reached by unrelated languages, independently and frequently. In assuming the existence of comparabletypes, therefore, we are not gainsaying the individuality of allhistorical processes; we are merely affirming that back of the face ofhistory are powerful drifts that move language, like other socialproducts, to balanced patterns, in other words, to types. As linguistswe shall be content to realize that there are these types and thatcertain processes in the life of language tend to modify them. Whysimilar types should be formed, just what is the nature of the forcesthat make them and dissolve them--these questions are more easily askedthan answered. Perhaps the psychologists of the future will be able togive us the ultimate reasons for the formation of linguistic types. 

When it comes to the actual task of classification, we find that we haveno easy road to travel. Various classifications have been suggested, andthey all contain elements of value. Yet none proves satisfactory. Theydo not so much enfold the known languages in their embrace as force themdown into narrow, straight-backed seats. The difficulties have been ofvarious kinds. First and foremost, it has been difficult to choose apoint of view. On what basis shall we classify? A language shows us somany facets that we may well be puzzled. And is one point of viewsufficient? Secondly, it is dangerous to generalize from a small numberof selected languages. To take, as the sum total of our material, Latin, Arabic, Turkish, Chinese, and perhaps Eskimo or Sioux as anafterthought, is to court disaster. We have no right to assume that asprinkling of exotic types will do to supplement the few languagesnearer home that we are more immediately interested in. Thirdly, thestrong craving for a simple formula[92] has been the undoing oflinguists. There is something irresistible about a method ofclassification that starts with two poles, exemplified, say, by Chineseand Latin, clusters what it conveniently can about these poles, andthrows everything else into a "transitional type. " Hence has arisen thestill popular classification of languages into an "isolating" group, an"agglutinative" group, and an "inflective" group. Sometimes thelanguages of the American Indians are made to straggle along as anuncomfortable "polysynthetic" rear-guard to the agglutinative languages. There is justification for the use of all of these terms, though notperhaps in quite the spirit in which they are commonly employed. In anycase it is very difficult to assign all known languages to one or otherof these groups, the more so as they are not mutually exclusive. Alanguage may be both agglutinative and inflective, or inflective andpolysynthetic, or even polysynthetic and isolating, as we shall see alittle later on. 

[Footnote 92: If possible, a triune formula. ]

There is a fourth reason why the classification of languages hasgenerally proved a fruitless undertaking. It is probably the mostpowerful deterrent of all to clear thinking. This is the evolutionaryprejudice which instilled itself into the social sciences towards themiddle of the last century and which is only now beginning to abate itstyrannical hold on our mind. Intermingled with this scientific prejudiceand largely anticipating it was another, a more human one. The vastmajority of linguistic theorists themselves spoke languages of a certaintype, of which the most fully developed varieties were the Latin andGreek that they had learned in their childhood. It was not difficultfor them to be persuaded that these familiar languages represented the"highest" development that speech had yet attained and that all othertypes were but steps on the way to this beloved "inflective" type. Whatever conformed to the pattern of Sanskrit and Greek and Latin andGerman was accepted as expressive of the "highest, " whatever departedfrom it was frowned upon as a shortcoming or was at best an interestingaberration. [93] Now any classification that starts with preconceivedvalues or that works up to sentimental satisfactions is self-condemnedas unscientific. A linguist that insists on talking about the Latin typeof morphology as though it were necessarily the high-water mark oflinguistic development is like the zoölogist that sees in the organicworld a huge conspiracy to evolve the race-horse or the Jersey cow. Language in its fundamental forms is the symbolic expression of humanintuitions. These may shape themselves in a hundred ways, regardless ofthe material advancement or backwardness of the people that handle theforms, of which, it need hardly be said, they are in the mainunconscious. If, therefore, we wish to understand language in its trueinwardness we must disabuse our minds of preferred "values"[94] andaccustom ourselves to look upon English and Hottentot with the samecool, yet interested, detachment. 

[Footnote 93: One celebrated American writer on culture and languagedelivered himself of the dictum that, estimable as the speakers ofagglutinative languages might be, it was nevertheless a crime for aninflecting woman to marry an agglutinating man. Tremendous spiritualvalues were evidently at stake. Champions of the "inflective" languagesare wont to glory in the very irrationalities of Latin and Greek, exceptwhen it suits them to emphasize their profoundly "logical" character. Yet the sober logic of Turkish or Chinese leaves them cold. The gloriousirrationalities and formal complexities of many "savage" languages theyhave no stomach for. Sentimentalists are difficult people. ]

[Footnote 94: I have in mind valuations of form as such. Whether or nota language has a large and useful vocabulary is another matter. Theactual size of a vocabulary at a given time is not a thing of realinterest to the linguist, as all languages have the resources at theirdisposal for the creation of new words, should need for them arise. Furthermore, we are not in the least concerned with whether or not alanguage is of great practical value or is the medium of a greatculture. All these considerations, important from other standpoints, have nothing to do with form value. ]

We come back to our first difficulty. What point of view shall we adoptfor our classification? After all that we have said about grammaticalform in the preceding chapter, it is clear that we cannot now make thedistinction between form languages and formless languages that used toappeal to some of the older writers. Every language can and must expressthe fundamental syntactic relations even though there is not a singleaffix to be found in its vocabulary. We conclude that every language isa form language. Aside from the expression of pure relation a languagemay, of course, be "formless"--formless, that is, in the mechanical andrather superficial sense that it is not encumbered by the use ofnon-radical elements. The attempt has sometimes been made to formulate adistinction on the basis of "inner form. " Chinese, for instance, has noformal elements pure and simple, no "outer form, " but it evidences akeen sense of relations, of the difference between subject and object, attribute and predicate, and so on. In other words, it has an "innerform" in the same sense in which Latin possesses it, though it isoutwardly "formless" where Latin is outwardly "formal. " On the otherhand, there are supposed to be languages[95] which have no true grasp ofthe fundamental relations but content themselves with the more or lessminute expression of material ideas, sometimes with an exuberantdisplay of "outer form, " leaving the pure relations to be merelyinferred from the context. I am strongly inclined to believe that thissupposed "inner formlessness" of certain languages is an illusion. Itmay well be that in these languages the relations are not expressed inas immaterial a way as in Chinese or even as in Latin, [96] or that theprinciple of order is subject to greater fluctuations than in Chinese, or that a tendency to complex derivations relieves the language of thenecessity of expressing certain relations as explicitly as a moreanalytic language would have them expressed. [97] All this does not meanthat the languages in question have not a true feeling for thefundamental relations. We shall therefore not be able to use the notionof "inner formlessness, " except in the greatly modified sense thatsyntactic relations may be fused with notions of another order. To thiscriterion of classification we shall have to return a little later. 

[Footnote 95: E. G. , Malay, Polynesian. ]

[Footnote 96: Where, as we have seen, the syntactic relations are by nomeans free from an alloy of the concrete. ]

[Footnote 97: Very much as an English _cod-liver oil_ dodges to someextent the task of explicitly defining the relations of the three nouns. Contrast French _huile de foie de morue_ "oil of liver of cod. "]

More justifiable would be a classification according to the formalprocesses[98] most typically developed in the language. Those languagesthat always identify the word with the radical element would be set offas an "isolating" group against such as either affix modifying elements(affixing languages) or possess the power to change the significance ofthe radical element by internal changes (reduplication; vocalic andconsonantal change; changes in quantity, stress, and pitch). The lattertype might be not inaptly termed "symbolic" languages. [99] The affixinglanguages would naturally subdivide themselves into such as areprevailingly prefixing, like Bantu or Tlingit, and such as are mainly orentirely suffixing, like Eskimo or Algonkin or Latin. There are twoserious difficulties with this fourfold classification (isolating, prefixing, suffixing, symbolic). In the first place, most languages fallinto more than one of these groups. The Semitic languages, for instance, are prefixing, suffixing, and symbolic at one and the same time. In thesecond place, the classification in its bare form is superficial. Itwould throw together languages that differ utterly in spirit merelybecause of a certain external formal resemblance. There is clearly aworld of difference between a prefixing language like Cambodgian, whichlimits itself, so far as its prefixes (and infixes) are concerned, tothe expression of derivational concepts, and the Bantu languages, inwhich the prefixed elements have a far-reaching significance as symbolsof syntactic relations. The classification has much greater value if itis taken to refer to the expression of relational concepts[100] alone. In this modified form we shall return to it as a subsidiary criterion. We shall find that the terms "isolating, " "affixing, " and "symbolic"have a real value. But instead of distinguishing between prefixing andsuffixing languages, we shall find that it is of superior interest tomake another distinction, one that is based on the relative firmnesswith which the affixed elements are united with the core of theword. [101]

[Footnote 98: See Chapter IV. ]

[Footnote 99: There is probably a real psychological connection betweensymbolism and such significant alternations as _drink_, _drank_, _drunk_or Chinese _mai_ (with rising tone) "to buy" and _mai_ (with fallingtone) "to sell. " The unconscious tendency toward symbolism is justlyemphasized by recent psychological literature. Personally I feel thatthe passage from _sing_ to _sang_ has very much the same feeling as thealternation of symbolic colors--e. G. , green for safe, red for danger. But we probably differ greatly as to the intensity with which we feelsymbolism in linguistic changes of this type. ]

[Footnote 100: Pure or "concrete relational. " See Chapter V. ]

[Footnote 101: In spite of my reluctance to emphasize the differencebetween a prefixing and a suffixing language, I feel that there is moreinvolved in this difference than linguists have generally recognized. Itseems to me that there is a rather important psychological distinctionbetween a language that settles the formal status of a radical elementbefore announcing it--and this, in effect, is what such languages asTlingit and Chinook and Bantu are in the habit of doing--and one thatbegins with the concrete nucleus of a word and defines the status ofthis nucleus by successive limitations, each curtailing in some degreethe generality of all that precedes. The spirit of the former method hassomething diagrammatic or architectural about it, the latter is a methodof pruning afterthoughts. In the more highly wrought prefixing languagesthe word is apt to affect us as a crystallization of floating elements, the words of the typical suffixing languages (Turkish, Eskimo, Nootka)are "determinative" formations, each added element determining the formof the whole anew. It is so difficult in practice to apply theseelusive, yet important, distinctions that an elementary study has norecourse but to ignore them. ]

There is another very useful set of distinctions that can be made, butthese too must not be applied exclusively, or our classification willagain be superficial. I refer to the notions of "analytic, " "synthetic, "and "polysynthetic. " The terms explain themselves. An analytic languageis one that either does not combine concepts into single words at all(Chinese) or does so economically (English, French). In an analyticlanguage the sentence is always of prime importance, the word is ofminor interest. In a synthetic language (Latin, Arabic, Finnish) theconcepts cluster more thickly, the words are more richly chambered, butthere is a tendency, on the whole, to keep the range of concretesignificance in the single word down to a moderate compass. Apolysynthetic language, as its name implies, is more than ordinarilysynthetic. The elaboration of the word is extreme. Concepts which weshould never dream of treating in a subordinate fashion are symbolizedby derivational affixes or "symbolic" changes in the radical element, while the more abstract notions, including the syntactic relations, mayalso be conveyed by the word. A polysynthetic language illustrates noprinciples that are not already exemplified in the more familiarsynthetic languages. It is related to them very much as a syntheticlanguage is related to our own analytic English. [102] The three termsare purely quantitative--and relative, that is, a language may be"analytic" from one standpoint, "synthetic" from another. I believe theterms are more useful in defining certain drifts than as absolutecounters. It is often illuminating to point out that a language has beenbecoming more and more analytic in the course of its history or that itshows signs of having crystallized from a simple analytic base into ahighly synthetic form. [103]

[Footnote 102: English, however, is only analytic in tendency. Relatively to French, it is still fairly synthetic, at least in certainaspects. ]

[Footnote 103: The former process is demonstrable for English, French, Danish, Tibetan, Chinese, and a host of other languages. The lattertendency may be proven, I believe, for a number of American Indianlanguages, e. G. , Chinook, Navaho. Underneath their present moderatelypolysynthetic form is discernible an analytic base that in the one casemay be roughly described as English-like, in the other, Tibetan-like. ]

We now come to the difference between an "inflective" and an"agglutinative" language. As I have already remarked, the distinction isa useful, even a necessary, one, but it has been generally obscured by anumber of irrelevancies and by the unavailing effort to make the termscover all languages that are not, like Chinese, of a definitelyisolating cast. The meaning that we had best assign to the term"inflective" can be gained by considering very briefly what are some ofthe basic features of Latin and Greek that have been looked upon aspeculiar to the inflective languages. First of all, they are syntheticrather than analytic. This does not help us much. Relatively to manyanother language that resembles them in broad structural respects, Latinand Greek are not notably synthetic; on the other hand, their moderndescendants, Italian and Modern Greek, while far more analytic[104] thanthey, have not departed so widely in structural outlines as to warranttheir being put in a distinct major group. An inflective language, wemust insist, may be analytic, synthetic, or polysynthetic. 

[Footnote 104: This applies more particularly to the Romance group:Italian, Spanish, Portuguese, French, Roumanian. Modern Greek is not soclearly analytic. ]

Latin and Greek are mainly affixing in their method, with the emphasisheavily on suffixing. The agglutinative languages are just as typicallyaffixing as they, some among them favoring prefixes, others running tothe use of suffixes. Affixing alone does not define inflection. Possiblyeverything depends on just what kind of affixing we have to deal with. If we compare our English words _farmer_ and _goodness_ with such wordsas _height_ and _depth_, we cannot fail to be struck by a notabledifference in the affixing technique of the two sets. The _-er_ and_-ness_ are affixed quite mechanically to radical elements which are atthe same time independent words (_farm_, _good_). They are in no senseindependently significant elements, but they convey their meaning(agentive, abstract quality) with unfailing directness. Their use issimple and regular and we should have no difficulty in appending them toany verb or to any adjective, however recent in origin. From a verb _tocamouflage_ we may form the noun _camouflager_ "one who camouflages, "from an adjective _jazzy_ proceeds with perfect ease the noun_jazziness_. It is different with _height_ and _depth_. Functionallythey are related to _high_ and _deep_ precisely as is _goodness_ to_good_, but the degree of coalescence between radical element and affixis greater. Radical element and affix, while measurably distinct, cannotbe torn apart quite so readily as could the _good_ and _-ness_ of_goodness_. The _-t_ of _height_ is not the typical form of the affix(compare _strength_, _length_, _filth_, _breadth_, _youth_), while_dep-_ is not identical with _deep_. We may designate the two types ofaffixing as "fusing" and "juxtaposing. " The juxtaposing technique we maycall an "agglutinative" one, if we like. 

Is the fusing technique thereby set off as the essence of inflection? Iam afraid that we have not yet reached our goal. If our language werecrammed full of coalescences of the type of _depth_, but if, on theother hand, it used the plural independently of verb concord (e. G. , _thebooks falls_ like _the book falls_, or _the book fall_ like _the booksfall_), the personal endings independently of tense (e. G. , _the bookfells_ like _the book falls_, or _the book fall_ like _the book fell_), and the pronouns independently of case (e. G. , _I see he_ like _he seesme_, or _him see the man_ like _the man sees him_), we should hesitateto describe it as inflective. The mere fact of fusion does not seem tosatisfy us as a clear indication of the inflective process. There are, indeed, a large number of languages that fuse radical element and affixin as complete and intricate a fashion as one could hope to findanywhere without thereby giving signs of that particular kind offormalism that marks off such languages as Latin and Greek asinflective. 

What is true of fusion is equally true of the "symbolic" processes. [105]There are linguists that speak of alternations like _drink_ and _drank_as though they represented the high-water mark of inflection, a kind ofspiritualized essence of pure inflective form. In such Greek forms, nevertheless, as _pepomph-a_ "I have sent, " as contrasted with _pemp-o_"I send, " with its trebly symbolic change of the radical element(reduplicating _pe-_, change of _e_ to _o_, change of _p_ to _ph_), itis rather the peculiar alternation of the first person singular _-a_ ofthe perfect with the _-o_ of the present that gives them theirinflective cast. Nothing could be more erroneous than to imagine thatsymbolic changes of the radical element, even for the expression of suchabstract concepts as those of number and tense, is always associatedwith the syntactic peculiarities of an inflective language. If by an"agglutinative" language we mean one that affixes according to thejuxtaposing technique, then we can only say that there are hundreds offusing and symbolic languages--non-agglutinative by definition--thatare, for all that, quite alien in spirit to the inflective type of Latinand Greek. We can call such languages inflective, if we like, but wemust then be prepared to revise radically our notion of inflective form. 

[Footnote 105: See pages 133, 134. ]

[Transcriber's note: Footnote 105 refers to the paragraph beginning online 4081. ]

It is necessary to understand that fusion of the radical element and theaffix may be taken in a broader psychological sense than I have yetindicated. If every noun plural in English were of the type of _book_:_books_, if there were not such conflicting patterns as _deer_: _deer_, _ox_: _oxen_, _goose_: _geese_ to complicate the general form picture ofplurality, there is little doubt that the fusion of the elements _book_and _-s_ into the unified word _books_ would be felt as a little lesscomplete than it actually is. One reasons, or feels, unconsciously aboutthe matter somewhat as follows:--If the form pattern represented by theword _books_ is identical, as far as use is concerned, with that of theword _oxen_, the pluralizing elements _-s_ and _-en_ cannot have quiteso definite, quite so autonomous, a value as we might at first beinclined to suppose. They are plural elements only in so far asplurality is predicated of certain selected concepts. The words _books_and _oxen_ are therefore a little other than mechanical combinations ofthe symbol of a thing (_book_, _ox_) and a clear symbol of plurality. There is a slight psychological uncertainty or haze about the juncturein _book-s_ and _ox-en_. A little of the force of _-s_ and _-en_ isanticipated by, or appropriated by, the words _book_ and _ox_themselves, just as the conceptual force of _-th_ in _dep-th_ isappreciably weaker than that of _-ness_ in _good-ness_ in spite of thefunctional parallelism between _depth_ and _goodness_. Where there isuncertainty about the juncture, where the affixed element cannot rightlyclaim to possess its full share of significance, the unity of thecomplete word is more strongly emphasized. The mind must rest onsomething. If it cannot linger on the constituent elements, it hastensall the more eagerly to the acceptance of the word as a whole. A wordlike _goodness_ illustrates "agglutination, " _books_ "regular fusion, "_depth_ "irregular fusion, " _geese_ "symbolic fusion" or"symbolism. "[106]

[Footnote 106: The following formulae may prove useful to those that aremathematically inclined. Agglutination: c = a + b; regular fusion:c = a + (b - x) + x; irregular fusion: c = (a - x) + (b - y) + (x + y);symbolism: c = (a - x) + x. I do not wish to imply that there is anymystic value in the process of fusion. It is quite likely to havedeveloped as a purely mechanical product of phonetic forces that broughtabout irregularities of various sorts. ]

The psychological distinctness of the affixed elements in anagglutinative term may be even more marked than in the _-ness_ of_goodness_. To be strictly accurate, the significance of the _-ness_ isnot quite as inherently determined, as autonomous, as it might be. Itis at the mercy of the preceding radical element to this extent, that itrequires to be preceded by a particular type of such element, anadjective. Its own power is thus, in a manner, checked in advance. Thefusion here, however, is so vague and elementary, so much a matter ofcourse in the great majority of all cases of affixing, that it isnatural to overlook its reality and to emphasize rather the juxtaposingor agglutinative nature of the affixing process. If the _-ness_ could beaffixed as an abstractive element to each and every type of radicalelement, if we could say _fightness_ ("the act or quality of fighting")or _waterness_ ("the quality or state of water") or _awayness_ ("thestate of being away") as we can say _goodness_ ("the state of beinggood"), we should have moved appreciably nearer the agglutinative pole. A language that runs to synthesis of this loose-jointed sort may belooked upon as an example of the ideal agglutinative type, particularlyif the concepts expressed by the agglutinated elements are relationalor, at the least, belong to the abstracter class of derivational ideas. 

Instructive forms may be cited from Nootka. We shall return to our "firein the house. "[107] The Nootka word _inikw-ihl_ "fire in the house" isnot as definitely formalized a word as its translation, suggests. Theradical element _inikw-_ "fire" is really as much of a verbal as of anominal term; it may be rendered now by "fire, " now by "burn, " accordingto the syntactic exigencies of the sentence. The derivational element_-ihl_ "in the house" does not mitigate this vagueness or generality;_inikw-ihl_ is still "fire in the house" or "burn in the house. " It maybe definitely nominalized or verbalized by the affixing of elements thatare exclusively nominal or verbal in force. For example, _inikw-ihl-'i_, with its suffixed article, is a clear-cut nominal form:"the burning in the house, the fire in the house"; _inikw-ihl-ma_, withits indicative suffix, is just as clearly verbal: "it burns in thehouse. " How weak must be the degree of fusion between "fire in thehouse" and the nominalizing or verbalizing suffix is apparent from thefact that the formally indifferent _inikwihl_ is not an abstractiongained by analysis but a full-fledged word, ready for use in thesentence. The nominalizing _-'i_ and the indicative _-ma_ are not fusedform-affixes, they are simply additions of formal import. But we cancontinue to hold the verbal or nominal nature of _inikwihl_ in abeyancelong before we reach the _-'i_ or _-ma_. We can pluralize it:_inikw-ihl-'minih_; it is still either "fires in the house" or "burnplurally in the house. " We can diminutivize this plural:_inikw-ihl-'minih-'is_, "little fires in the house" or "burn plurallyand slightly in the house. " What if we add the preterit tense suffix_-it_? Is not _inikw-ihl-'minih-'is-it_ necessarily a verb: "severalsmall fires were burning in the house"? It is not. It may still benominalized; _inikwihl'minih'isit-'i_ means "the former small fires inthe house, the little fires that were once burning in the house. " It isnot an unambiguous verb until it is given a form that excludes everyother possibility, as in the indicative _inikwihl-minih'isit-a_ "severalsmall fires were burning in the house. " We recognize at once that theelements _-ihl_, _-'minih_, _-'is_, and _-it_, quite aside from therelatively concrete or abstract nature of their content and aside, further, from the degree of their outer (phonetic) cohesion with theelements that precede them, have a psychological independence that ourown affixes never have. They are typically agglutinated elements, thoughthey have no greater external independence, are no more capable ofliving apart from the radical element to which they are suffixed, thanthe _-ness_ and _goodness_ or the _-s_ of _books_. It does not followthat an agglutinative language may not make use of the principle offusion, both external and psychological, or even of symbolism to aconsiderable extent. It is a question of tendency. Is the formativeslant clearly towards the agglutinative method? Then the language is"agglutinative. " As such, it may be prefixing or suffixing, analytic, synthetic, or polysynthetic. 

[Footnote 107: See page 110. ]

[Transcriber's note: Footnote 107 refers to the paragraph beginning online 3331. ]

To return to inflection. An inflective language like Latin or Greek usesthe method of fusion, and this fusion has an inner psychological as wellas an outer phonetic meaning. But it is not enough that the fusionoperate merely in the sphere of derivational concepts (group II), [108]it must involve the syntactic relations, which may either be expressedin unalloyed form (group IV) or, as in Latin and Greek, as "concreterelational concepts" (group III). [109] As far as Latin and Greek areconcerned, their inflection consists essentially of the fusing ofelements that express logically impure relational concepts with radicalelements and with elements expressing derivational concepts. Both fusionas a general method and the expression of relational concepts in theword are necessary to the notion of "inflection. "

[Footnote 108: See Chapter V. ]

[Footnote 109: If we deny the application of the term "inflective" tofusing languages that express the syntactic relations in pure form, thatis, without the admixture of such concepts as number, gender, and tense, merely because such admixture is familiar to us in Latin and Greek, wemake of "inflection" an even more arbitrary concept than it need be. Atthe same time it is true that the method of fusion itself tends to breakdown the wall between our conceptual groups II and IV, to create groupIII. Yet the possibility of such "inflective" languages should not bedenied. In modern Tibetan, for instance, in which concepts of group IIare but weakly expressed, if at all, and in which the relationalconcepts (e. G. , the genitive, the agentive or instrumental) areexpressed without alloy of the material, we get many interestingexamples of fusion, even of symbolism. _Mi di_, e. G. , "man this, theman" is an absolutive form which may be used as the subject of anintransitive verb. When the verb is transitive (really passive), the(logical) subject has to take the agentive form. _Mi di_ then becomes_mi di_ "by the man, " the vowel of the demonstrative pronoun (orarticle) being merely lengthened. (There is probably also a change inthe tone of the syllable. ) This, of course, is of the very essence ofinflection. It is an amusing commentary on the insufficiency of ourcurrent linguistic classification, which considers "inflective" and"isolating" as worlds asunder, that modern Tibetan may be not inaptlydescribed as an isolating language, aside from such examples of fusionand symbolism as the foregoing. ]

But to have thus defined inflection is to doubt the value of the term asdescriptive of a major class. Why emphasize both a technique and aparticular content at one and the same time? Surely we should be clearin our minds as to whether we set more store by one or the other. "Fusional" and "symbolic" contrast with "agglutinative, " which is not ona par with "inflective" at all. What are we to do with the fusional andsymbolic languages that do not express relational concepts in the wordbut leave them to the sentence? And are we not to distinguish betweenagglutinative languages that express these same concepts in the word--inso far inflective-like--and those that do not? We dismissed the scale:analytic, synthetic, polysynthetic, as too merely quantitative for ourpurpose. Isolating, affixing, symbolic--this also seemed insufficientfor the reason that it laid too much stress on technical externals. Isolating, agglutinative, fusional, and symbolic is a preferable scheme, but still skirts the external. We shall do best, it seems to me, to holdto "inflective" as a valuable suggestion for a broader and moreconsistently developed scheme, as a hint for a classification based onthe nature of the concepts expressed by the language. The other twoclassifications, the first based on degree of synthesis, the second ondegree of fusion, may be retained as intercrossing schemes that give usthe opportunity to subdivide our main conceptual types. 

It is well to recall that all languages must needs express radicalconcepts (group I) and relational ideas (group IV). Of the two otherlarge groups of concepts--derivational (group II) and mixed relational(group III)--both may be absent, both present, or only one present. Thisgives us at once a simple, incisive, and absolutely inclusive method ofclassifying all known languages. They are:

A. Such as express only concepts of groups I and IV; in other words, languages that keep the syntactic relations pure and that do not possessthe power to modify the significance of their radical elements by meansof affixes or internal changes. [110] We may call these _Pure-relationalnon-deriving languages_ or, more tersely, _Simple Pure-relationallanguages_. These are the languages that cut most to the bone oflinguistic expression. 

B. Such as express concepts of groups I, II, and IV; in other words, languages that keep the syntactic relations pure and that also possessthe power to modify the significance of their radical elements by meansof affixes or internal changes. These are the _Pure-relational derivinglanguages_ or _Complex Pure-relational languages_. 

C. Such as express concepts of groups I and III;[111] in other words, languages in which the syntactic relations are expressed in necessaryconnection with concepts that are not utterly devoid of concretesignificance but that do not, apart from such mixture, possess the powerto modify the significance of their radical elements by means of affixesor internal changes. [112] These are the _Mixed-relational non-derivinglanguages_ or _Simple Mixed-relational languages_. 

D. Such as express concepts of groups I, II, and III; in other words, languages in which the syntactic relations are expressed in mixed form, as in C, and that also possess the power to modify the significance oftheir radical elements by means of affixes or internal changes. Theseare the _Mixed-relational deriving languages_ or _ComplexMixed-relational languages_. Here belong the "inflective" languages thatwe are most familiar with as well as a great many "agglutinative"languages, some "polysynthetic, " others merely synthetic. 

[Footnote 110: I am eliminating entirely the possibility of compoundingtwo or more radical elements into single words or word-like phrases (seepages 67-70). To expressly consider compounding in the present survey oftypes would be to complicate our problem unduly. Most languages thatpossess no derivational affixes of any sort may nevertheless freelycompound radical elements (independent words). Such compounds often havea fixity that simulates the unity of single words. ]

[Transcriber's note: Footnote 110 refers to the three paragraphsbeginning on line 2066. ]

[Footnote 111: We may assume that in these languages and in those oftype D all or most of the relational concepts are expressed in "mixed"form, that such a concept as that of subjectivity, for instance, cannotbe expressed without simultaneously involving number or gender or thatan active verb form must be possessed of a definite tense. Hence groupIII will be understood to include, or rather absorb, group IV. Theoretically, of course, certain relational concepts may be expressedpure, others mixed, but in practice it will not be found easy to makethe distinction. ]

[Footnote 112: The line between types C and D cannot be very sharplydrawn. It is a matter largely of degree. A language of markedlymixed-relational type, but of little power of derivation pure andsimple, such as Bantu or French, may be conveniently put into type C, even though it is not devoid of a number of derivational affixes. Roughly speaking, languages of type C may be considered as highlyanalytic ("purified") forms of type D. ]

This conceptual classification of languages, I must repeat, does notattempt to take account of the technical externals of language. Itanswers, in effect, two fundamental questions concerning thetranslation of concepts into linguistic symbols. Does the language, inthe first place, keep its radical concepts pure or does it build up itsconcrete ideas by an aggregation of inseparable elements (types A and C_versus_ types B and D)? And, in the second place, does it keep thebasic relational concepts, such as are absolutely unavoidable in theordering of a proposition, free of an admixture of the concrete or not(types A and B _versus_ types C and D)? The second question, it seems tome, is the more fundamental of the two. We can therefore simplify ourclassification and present it in the following form: _ I. Pure-relational _/ A. Simple Languages \_ B. Complex _II. Mixed-relational _/ C. Simple Languages \_ D. Complex

The classification is too sweeping and too broad for an easy, descriptive survey of the many varieties of human speech. It needs to beamplified. Each of the types A, B, C, D may be subdivided into anagglutinative, a fusional, and a symbolic sub-type, according to theprevailing method of modification of the radical element. In type A wedistinguish in addition an isolating sub-type, characterized by theabsence of all affixes and modifications of the radical element. In theisolating languages the syntactic relations are expressed by theposition of the words in the sentence. This is also true of manylanguages of type B, the terms "agglutinative, " "fusional, " and"symbolic" applying in their case merely to the treatment of thederivational, not the relational, concepts. Such languages could betermed "agglutinative-isolating, " "fusional-isolating" and"symbolic-isolating. "

This brings up the important general consideration that the method ofhandling one group of concepts need not in the least be identical withthat used for another. Compound terms could be used to indicate thisdifference, if desired, the first element of the compound referring tothe treatment of the concepts of group II, the second to that of theconcepts of groups III and IV. An "agglutinative" language wouldnormally be taken to mean one that agglutinates all of its affixedelements or that does so to a preponderating extent. In an"agglutinative-fusional" language the derivational elements areagglutinated, perhaps in the form of prefixes, while the relationalelements (pure or mixed) are fused with the radical element, possibly asanother set of prefixes following the first set or in theform of suffixes or as part prefixes and part suffixes. By a"fusional-agglutinative" language we would understand one that fuses itsderivational elements but allows a greater independence to those thatindicate relations. All these and similar distinctions are not merelytheoretical possibilities, they can be abundantly illustrated from thedescriptive facts of linguistic morphology. Further, should it provedesirable to insist on the degree of elaboration of the word, the terms"analytic, " "synthetic, " and "polysynthetic" can be added as descriptiveterms. It goes without saying that languages of type A are necessarilyanalytic and that languages of type C also are prevailingly analytic andare not likely to develop beyond the synthetic stage. 

But we must not make too much of terminology. Much depends on therelative emphasis laid on this or that feature or point of view. Themethod of classifying languages here developed has this greatadvantage, that it can be refined or simplified according to the needsof a particular discussion. The degree of synthesis may be entirelyignored; "fusion" and "symbolism" may often be combined with advantageunder the head of "fusion"; even the difference between agglutinationand fusion may, if desired, be set aside as either too difficult to drawor as irrelevant to the issue. Languages, after all, are exceedinglycomplex historical structures. It is of less importance to put eachlanguage in a neat pigeon-hole than to have evolved a flexible methodwhich enables us to place it, from two or three independent standpoints, relatively to another language. All this is not to deny that certainlinguistic types are more stable and frequently represented than othersthat are just as possible from a theoretical standpoint. But we are tooill-informed as yet of the structural spirit of great numbers oflanguages to have the right to frame a classification that is other thanflexible and experimental. 

The reader will gain a somewhat livelier idea of the possibilities oflinguistic morphology by glancing down the subjoined analytical table ofselected types. The columns II, III, IV refer to the groups of conceptsso numbered in the preceding chapter. The letters _a_, _b_, _c_, _d_refer respectively to the processes of isolation (position in thesentence), agglutination, fusion, and symbolism. Where more than onetechnique is employed, they are put in the order of theirimportance. [113]

[Footnote 113: In defining the type to which a language belongs one mustbe careful not to be misled by structural features which are meresurvivals of an older stage, which have no productive life and do notenter into the unconscious patterning of the language. All languages arelittered with such petrified bodies. The English _-ster_ of _spinster_and _Webster_ is an old agentive suffix, but, as far as the feeling ofthe present English-speaking generation is concerned, it cannot be saidto really exist at all; _spinster_ and _Webster_ have been completelydisconnected from the etymological group of _spin_ and of _weave (web)_. Similarly, there are hosts of related words in Chinese which differ inthe initial consonant, the vowel, the tone, or in the presence orabsence of a final consonant. Even where the Chinaman feels theetymological relationship, as in certain cases he can hardly help doing, he can assign no particular function to the phonetic variation as such. Hence it forms no live feature of the language-mechanism and must beignored in defining the general form of the language. The caution is allthe more necessary, as it is precisely the foreigner, who approaches anew language with a certain prying inquisitiveness, that is most apt tosee life in vestigial features which the native is either completelyunaware of or feels merely as dead form. ]

Note. --Parentheses indicate a weak development of the process inquestion. 

+----------------+---+----+---+--------------+----------+--------------+|Fundamental Type"II |III |IV |Technique "Synthesis "Examples |+----------------+---+----+---+--------------+----------+--------------+| A " | | | " " ||(Simple Pure- "-- |-- |a |Isolating "Analytic "Chinese; || relational) " | | | " "Annamite || " | | | " " || "(d)|-- |a, b|Isolating "Analytic "Ewe || " | | |(weakly " "(Guinea Coast)|| " | | |agglutinative)" " || " | | | " " || "(b)|-- |a, |Agglutinative "Analytic "Modern Tibetan|| " | |b, c|(mildly " " || " | | |agglutinative-" " || " | | |fusional) " " || " | | | " " || B " | | | " " ||(Complex Pure- "b, |-- |a |Agglutinative-"Analytic "Polynesian || relational) "(d)| | |isolating " " || " | | | " " || "b |-- |a, |Agglutinative-"Polysyn- "Haida || " | |(b)|isolating "thetic " || " | | | " " || "c |-- |a |Fusional- "Analytic "Cambodgian || " | | |isolating " " || " | | | " " || "b |-- |b |Agglutinative "Synthetic "Turkish || " | | | " " || "b, d|(b) |b |Agglutinative "Polysyn- "Yana (N. || " | | |(symbolic "thetic "California) || " | | |tinge) " " || " | | | " " || "c, |-- |a, b|Fusional- "Synthetic "Classical || "d, | | |agglutinative "(mildly) "Tibetan || "(b)| | |(symbolic " " || " | | |tinge) " " || " | | | " " || "b |-- |c |Agglutinative-"Synthetic "Sioux || " | | |fusional "(mildly " || " | | | "polysyn- " || " | | | "thetic) " || " | | | " " || "c |-- |c |Fusional "Synthetic "Salinan (S. W. || " | | | " "California) || " | | | " " || "d, c|(d) |d, |Symbolic "Analytic "Shilluk || " | |c, a| " "(Upper Nile) || " | | | " " || C " | | | " " ||(Simple Mixed- "(b)|b |-- |Agglutinative "Synthetic "Bantu || relational) " | | | " " || "(c)|c, |a |Fusional "Analytic "French[114] || " |(d) | | "(mildly " || " | | | "synthetic)" || " | | | " " || D " | | | " " ||(Complex Mixed- "b, |b |b |Agglutinative "Polysyn- "Nootka || relational) "c, d| | | "thetic "(Vancouver || " | | | "(symbolic "Island)[115] || " | | | "tinge) " || " | | | " " || "c, |b |-- |Fusional- "Polysyn- "Chinook (lower|| "(d)| | |agglutinative "thetic "Columbia R. ) || " | | | "(mildly) " || " | | | " " || "c, |c, |-- |Fusional "Polysyn- "Algonkin || "(d)|(d), | | "thetic " || " |(b) | | " " || " | | | " " || "c |c, d |a |Fusional "Analytic "English || " | | | " " || "c, d|c, d |-- |Fusional "Synthetic "Latin, Greek, || " | | |(symbolic " "Sanskrit || " | | |tinge) " " || " | | | " " || "c, |c, d |(a)|Fusional "Synthetic "Takelma || "b, d| | |(strongly " "(S. W. Oregon) || " | | |symbolic) " " || " | | | " " || "d, c|c, d |(a)|Symbolic- "Synthetic "Semitic || " | | |fusional " "(Arabic, || " | | | " "Hebrew) |+----------------+---+----+---+--------------+----------+--------------+

[Footnote 114: Might nearly as well have come under D. ]

[Footnote 115: Very nearly complex pure-relational. ]

I need hardly point out that these examples are far from exhausting thepossibilities of linguistic structure. Nor that the fact that twolanguages are similarly classified does not necessarily mean that theypresent a great similarity on the surface. We are here concerned withthe most fundamental and generalized features of the spirit, thetechnique, and the degree of elaboration of a given language. Nevertheless, in numerous instances we may observe this highlysuggestive and remarkable fact, that languages that fall into the sameclass have a way of paralleling each other in many details or instructural features not envisaged by the scheme of classification. Thus, a most interesting parallel could be drawn on structural lines betweenTakelma and Greek, [116] languages that are as geographically remote fromeach other and as unconnected in a historical sense as two languagesselected at random can well be. Their similarity goes beyond thegeneralized facts registered in the table. It would almost seem thatlinguistic features that are easily thinkable apart from each other, that seem to have no necessary connection in theory, have nevertheless atendency to cluster or to follow together in the wake of some deep, controlling impulse to form that dominates their drift. If, therefore, we can only be sure of the intuitive similarity of two given languages, of their possession of the same submerged form-feeling, we need not betoo much surprised to find that they seek and avoid certain linguisticdevelopments in common. We are at present very far from able to definejust what these fundamental form intuitions are. We can only feel themrather vaguely at best and must content ourselves for the most part withnoting their symptoms. These symptoms are being garnered in ourdescriptive and historical grammars of diverse languages. Some day, itmay be, we shall be able to read from them the great underlyingground-plans. 

[Footnote 116: Not Greek specifically, of course, but as a typicalrepresentative of Indo-European. ]

Such a purely technical classification of languages as the current oneinto "isolating, " "agglutinative, " and "inflective" (read "fusional")cannot claim to have great value as an entering wedge into the discoveryof the intuitional forms of language. I do not know whether thesuggested classification into four conceptual groups is likely to drivedeeper or not. My own feeling is that it does, but classifications, neatconstructions of the speculative mind, are slippery things. They have tobe tested at every possible opportunity before they have the right tocry for acceptance. Meanwhile we may take some encouragement from theapplication of a rather curious, yet simple, historical test. Languagesare in constant process of change, but it is only reasonable to supposethat they tend to preserve longest what is most fundamental in theirstructure. Now if we take great groups of genetically relatedlanguages, [117] we find that as we pass from one to another or trace thecourse of their development we frequently encounter a gradual change ofmorphological type. This is not surprising, for there is no reason why alanguage should remain permanently true to its original form. It isinteresting, however, to note that of the three intercrossingclassifications represented in our table (conceptual type, technique, and degree of synthesis), it is the degree of synthesis that seems tochange most readily, that the technique is modifiable but far lessreadily so, and that the conceptual type tends to persist the longest ofall. 

[Footnote 117: Such, in other words, as can be shown by documentary orcomparative evidence to have been derived from a common source. SeeChapter VII. ]

The illustrative material gathered in the table is far too scanty toserve as a real basis of proof, but it is highly suggestive as far as itgoes. The only changes of conceptual type within groups of relatedlanguages that are to be gleaned from the table are of B to A (Shillukas contrasted with Ewe;[118] Classical Tibetan as contrasted with ModernTibetan and Chinese) and of D to C (French as contrasted withLatin[119]). But types A : B and C : D are respectively related to eachother as a simple and a complex form of a still more fundamental type(pure-relational, mixed-relational). Of a passage from a pure-relationalto a mixed-relational type or _vice versa_ I can give no convincingexamples. 

[Footnote 118: These are far-eastern and far-western representatives ofthe "Soudan" group recently proposed by D. Westermann. The geneticrelationship between Ewe and Shilluk is exceedingly remote at best. ]

[Footnote 119: This case is doubtful at that. I have put French in Crather than in D with considerable misgivings. Everything depends on howone evaluates elements like _-al_ in _national_, _-té_ in _bonté_, or_re-_ in _retourner_. They are common enough, but are they as alive, aslittle petrified or bookish, as our English _-ness_ and _-ful_ and_un-_?]

The table shows clearly enough how little relative permanence there isin the technical features of language. That highly synthetic languages(Latin; Sanskrit) have frequently broken down into analytic forms(French; Bengali) or that agglutinative languages (Finnish) have inmany instances gradually taken on "inflective" features are well-knownfacts, but the natural inference does not seem to have been often drawnthat possibly the contrast between synthetic and analytic oragglutinative and "inflective" (fusional) is not so fundamental afterall. Turning to the Indo-Chinese languages, we find that Chinese is asnear to being a perfectly isolating language as any example we arelikely to find, while Classical Tibetan has not only fusional but strongsymbolic features (e. G. , _g-tong-ba_ "to give, " past _b-tang_, future_gtang_, imperative _thong_); but both are pure-relational languages. Ewe is either isolating or only barely agglutinative, while Shilluk, though soberly analytic, is one of the most definitely symboliclanguages I know; both of these Soudanese languages are pure-relational. The relationship between Polynesian and Cambodgian is remote, thoughpractically certain; while the latter has more markedly fusionalfeatures than the former, [120] both conform to the complexpure-relational type. Yana and Salinan are superficially very dissimilarlanguages. Yana is highly polysynthetic and quite typicallyagglutinative, Salinan is no more synthetic than and as irregularly andcompactly fusional ("inflective") as Latin; both are pure-relational, Chinook and Takelma, remotely related languages of Oregon, have divergedvery far from each other, not only as regards technique and synthesis ingeneral but in almost all the details of their structure; both arecomplex mixed-relational languages, though in very different ways. Factssuch as these seem to lend color to the suspicion that in the contrastof pure-relational and mixed-relational (or concrete-relational) we areconfronted by something deeper, more far-reaching, than the contrast ofisolating, agglutinative, and fusional. [121]

[Footnote 120: In spite of its more isolating cast. ]

[Footnote 121: In a book of this sort it is naturally impossible to givean adequate idea of linguistic structure in its varying forms. Only afew schematic indications are possible. A separate volume would beneeded to breathe life into the scheme. Such a volume would point outthe salient structural characteristics of a number of languages, soselected as to give the reader an insight into the formal economy ofstrikingly divergent types. ]

VII

LANGUAGE AS A HISTORICAL PRODUCT: DRIFT

Every one knows that language is variable. Two individuals of the samegeneration and locality, speaking precisely the same dialect and movingin the same social circles, are never absolutely at one in their speechhabits. A minute investigation of the speech of each individual wouldreveal countless differences of detail--in choice of words, in sentencestructure, in the relative frequency with which particular forms orcombinations of words are used, in the pronunciation of particularvowels and consonants and of combinations of vowels and consonants, inall those features, such as speed, stress, and tone, that give life tospoken language. In a sense they speak slightly divergent dialects ofthe same language rather than identically the same language. 

There is an important difference, however, between individual anddialectic variations. If we take two closely related dialects, sayEnglish as spoken by the "middle classes" of London and English asspoken by the average New Yorker, we observe that, however much theindividual speakers in each city differ from each other, the body ofLondoners forms a compact, relatively unified group in contrast to thebody of New Yorkers. The individual variations are swamped in orabsorbed by certain major agreements--say of pronunciation andvocabulary--which stand out very strongly when the language of thegroup as a whole is contrasted with that of the other group. This meansthat there is something like an ideal linguistic entity dominating thespeech habits of the members of each group, that the sense of almostunlimited freedom which each individual feels in the use of his languageis held in leash by a tacitly directing norm. One individual plays onthe norm in a way peculiar to himself, the next individual is nearer thedead average in that particular respect in which the first speaker mostcharacteristically departs from it but in turn diverges from the averagein a way peculiar to himself, and so on. What keeps the individual'svariations from rising to dialectic importance is not merely the factthat they are in any event of small moment--there are well-markeddialectic variations that are of no greater magnitude than individualvariations within a dialect--it is chiefly that they are silently"corrected" or canceled by the consensus of usage. If all the speakersof a given dialect were arranged in order in accordance with the degreeof their conformity to average usage, there is little doubt that theywould constitute a very finely intergrading series clustered about awell-defined center or norm. The differences between any two neighboringspeakers of the series[122] would be negligible for any but the mostmicroscopic linguistic research. The differences between the outer-mostmembers of the series are sure to be considerable, in all likelihoodconsiderable enough to measure up to a true dialectic variation. Whatprevents us from saying that these untypical individuals speak distinctdialects is that their peculiarities, as a unified whole, are notreferable to another norm than the norm of their own series. 

[Footnote 122: In so far as they do not fall out of the normal speechgroup by reason of a marked speech defect or because they are isolatedforeigners that have acquired the language late in life. ]

If the speech of any member of the series could actually be made to fitinto another dialect series, [123] we should have no true barriersbetween dialects (and languages) at all. We should merely have acontinuous series of individual variations extending over the wholerange of a historically unified linguistic area, and the cutting up ofthis large area (in some cases embracing parts of several continents)into distinct dialects and languages would be an essentially arbitraryproceeding with no warrant save that of practical convenience. But sucha conception of the nature of dialectic variation does not correspond tothe facts as we know them. Isolated individuals may be found who speak acompromise between two dialects of a language, and if their number andimportance increases they may even end by creating a new dialectic normof their own, a dialect in which the extreme peculiarities of the parentdialects are ironed out. In course of time the compromise dialect mayabsorb the parents, though more frequently these will tend to lingerindefinitely as marginal forms of the enlarged dialect area. But suchphenomena--and they are common enough in the history of language--areevidently quite secondary. They are closely linked with such socialdevelopments as the rise of nationality, the formation of literaturesthat aim to have more than a local appeal, the movement of ruralpopulations into the cities, and all those other tendencies that breakup the intense localism that unsophisticated man has always foundnatural. 

[Footnote 123: Observe that we are speaking of an individual's speech asa whole. It is not a question of isolating some particular peculiarityof pronunciation or usage and noting its resemblance to or identity witha feature in another dialect. ]

The explanation of primary dialectic differences is still to seek. Itis evidently not enough to say that if a dialect or language is spokenin two distinct localities or by two distinct social strata it naturallytakes on distinctive forms, which in time come to be divergent enough todeserve the name of dialects. This is certainly true as far as it goes. Dialects do belong, in the first instance, to very definitelycircumscribed social groups, homogeneous enough to secure the commonfeeling and purpose needed to create a norm. But the embarrassingquestion immediately arises, If all the individual variations within adialect are being constantly leveled out to the dialectic norm, if thereis no appreciable tendency for the individual's peculiarities toinitiate a dialectic schism, why should we have dialectic variations atall? Ought not the norm, wherever and whenever threatened, automaticallyto reassert itself? Ought not the individual variations of eachlocality, even in the absence of intercourse between them, to cancel outto the same accepted speech average?

If individual variations "on a flat" were the only kind of variabilityin language, I believe we should be at a loss to explain why and howdialects arise, why it is that a linguistic prototype gradually breaksup into a number of mutually unintelligible languages. But language isnot merely something that is spread out in space, as it were--a seriesof reflections in individual minds of one and the same timeless picture. Language moves down time in a current of its own making. It has a drift. If there were no breaking up of a language into dialects, if eachlanguage continued as a firm, self-contained unity, it would still beconstantly moving away from any assignable norm, developing new featuresunceasingly and gradually transforming itself into a language sodifferent from its starting point as to be in effect a new language. Nowdialects arise not because of the mere fact of individual variation butbecause two or more groups of individuals have become sufficientlydisconnected to drift apart, or independently, instead of together. Solong as they keep strictly together, no amount of individual variationwould lead to the formation of dialects. In practice, of course, nolanguage can be spread over a vast territory or even over a considerablearea without showing dialectic variations, for it is impossible to keepa large population from segregating itself into local groups, thelanguage of each of which tends to drift independently. Under culturalconditions such as apparently prevail to-day, conditions that fightlocalism at every turn, the tendency to dialectic cleavage is beingconstantly counteracted and in part "corrected" by the uniformizingfactors already referred to. Yet even in so young a country as Americathe dialectic differences are not inconsiderable. 

Under primitive conditions the political groups are small, the tendencyto localism exceedingly strong. It is natural, therefore, that thelanguages of primitive folk or of non-urban populations in general aredifferentiated into a great number of dialects. There are parts of theglobe where almost every village has its own dialect. The life of thegeographically limited community is narrow and intense; its speech iscorrespondingly peculiar to itself. It is exceedingly doubtful if alanguage will ever be spoken over a wide area without multiplying itselfdialectically. No sooner are the old dialects ironed out by compromisesor ousted by the spread and influence of the one dialect which isculturally predominant when a new crop of dialects arises to undo theleveling work of the past. This is precisely what happened in Greece, for instance. In classical antiquity there were spoken a large number oflocal dialects, several of which are represented in the literature. Asthe cultural supremacy of Athens grew, its dialect, the Attic, spread atthe expense of the rest, until, in the so-called Hellenistic periodfollowing the Macedonian conquest, the Attic dialect, in the vulgarizedform known as the "Koine, " became the standard speech of all Greece. Butthis linguistic uniformity[124] did not long continue. During the twomillennia that separate the Greek of to-day from its classical prototypethe Koine gradually split up into a number of dialects. Now Greece is asrichly diversified in speech as in the time of Homer, though the presentlocal dialects, aside from those of Attica itself, are not the linealdescendants of the old dialects of pre-Alexandrian days. [125] Theexperience of Greece is not exceptional. Old dialects are beingcontinually wiped out only to make room for new ones. Languages canchange at so many points of phonetics, morphology, and vocabulary thatit is not surprising that once the linguistic community is broken itshould slip off in different directions. It would be too much to expecta locally diversified language to develop along strictly parallel lines. If once the speech of a locality has begun to drift on its own account, it is practically certain to move further and further away from itslinguistic fellows. Failing the retarding effect of dialecticinterinfluences, which I have already touched upon, a group of dialectsis bound to diverge on the whole, each from all of the others. 

[Footnote 124: It is doubtful if we have the right to speak oflinguistic uniformity even during the predominance of the Koine. It ishardly conceivable that when the various groups of non-Attic Greeks tookon the Koine they did not at once tinge it with dialectic peculiaritiesinduced by their previous speech habits. ]

[Footnote 125: The Zaconic dialect of Lacedaemon is the sole exception. It is not derived from the Koine, but stems directly from the Doricdialect of Sparta. ]

In course of time each dialect itself splits up into sub-dialects, whichgradually take on the dignity of dialects proper while the primarydialects develop into mutually unintelligible languages. And so thebudding process continues, until the divergences become so great thatnone but a linguistic student, armed with his documentary evidence andwith his comparative or reconstructive method, would infer that thelanguages in question were genealogically related, representedindependent lines of development, in other words, from a remote andcommon starting point. Yet it is as certain as any historical fact canbe that languages so little resembling each other as Modern Irish, English, Italian, Greek, Russian, Armenian, Persian, and Bengali are butend-points in the present of drifts that converge to a meeting-point inthe dim past. There is naturally no reason to believe that this earliest"Indo-European" (or "Aryan") prototype which we can in part reconstruct, in part but dimly guess at, is itself other than a single "dialect" of agroup that has either become largely extinct or is now furtherrepresented by languages too divergent for us, with our limited means, to recognize as clear kin. [126]

[Footnote 126: Though indications are not lacking of what these remoterkin of the Indo-European languages may be. This is disputed ground, however, and hardly fit subject for a purely general study of speech. ]

All languages that are known to be genetically related, i. E. , to bedivergent forms of a single prototype, may be considered as constitutinga "linguistic stock. " There is nothing final about a linguistic stock. When we set it up, we merely say, in effect, that thus far we can goand no farther. At any point in the progress of our researches anunexpected ray of light may reveal the "stock" as but a "dialect" of alarger group. The terms dialect, language, branch, stock--it goeswithout saying--are purely relative terms. They are convertible as ourperspective widens or contracts. [127] It would be vain to speculate asto whether or not we shall ever be able to demonstrate that alllanguages stem from a common source. Of late years linguists have beenable to make larger historical syntheses than were at one time deemedfeasible, just as students of culture have been able to show historicalconnections between culture areas or institutions that were at one timebelieved to be totally isolated from each other. The human world iscontracting not only prospectively but to the backward-probing eye ofculture-history. Nevertheless we are as yet far from able to reduce theriot of spoken languages to a small number of "stocks. " We must stilloperate with a quite considerable number of these stocks. Some of them, like Indo-European or Indo-Chinese, are spoken over tremendous reaches;others, like Basque, [128] have a curiously restricted range and are inall likelihood but dwindling remnants of groups that were at one timemore widely distributed. As for the single or multiple origin of speech, it is likely enough that language as a human institution (or, if oneprefers, as a human "faculty") developed but once in the history of therace, that all the complex history of language is a unique culturalevent. Such a theory constructed "on general principles" is of no realinterest, however, to linguistic science. What lies beyond thedemonstrable must be left to the philosopher or the romancer. 

[Footnote 127: "Dialect" in contrast to an accepted literary norm is ause of the term that we are not considering. ]

[Footnote 128: Spoken in France and Spain in the region of thePyrenees. ]

We must return to the conception of "drift" in language. If thehistorical changes that take place in a language, if the vastaccumulation of minute modifications which in time results in thecomplete remodeling of the language, are not in essence identical withthe individual variations that we note on every hand about us, if thesevariations are born only to die without a trace, while the equallyminute, or even minuter, changes that make up the drift are foreverimprinted on the history of the language, are we not imputing to thishistory a certain mystical quality? Are we not giving language a powerto change of its own accord over and above the involuntary tendency ofindividuals to vary the norm? And if this drift of language is notmerely the familiar set of individual variations seen in verticalperspective, that is historically, instead of horizontally, that is indaily experience, what is it? Language exists only in so far as it isactually used--spoken and heard, written and read. What significantchanges take place in it must exist, to begin with, as individualvariations. This is perfectly true, and yet it by no means follows thatthe general drift of language can be understood[129] from an exhaustivedescriptive study of these variations alone. They themselves are randomphenomena, [130] like the waves of the sea, moving backward and forwardin purposeless flux. The linguistic drift has direction. In other words, only those individual variations embody it or carry it which move in acertain direction, just as only certain wave movements in the bayoutline the tide. The drift of a language is constituted by theunconscious selection on the part of its speakers of those individualvariations that are cumulative in some special direction. This directionmay be inferred, in the main, from the past history of the language. Inthe long run any new feature of the drift becomes part and parcel of thecommon, accepted speech, but for a long time it may exist as a meretendency in the speech of a few, perhaps of a despised few. As we lookabout us and observe current usage, it is not likely to occur to us thatour language has a "slope, " that the changes of the next few centuriesare in a sense prefigured in certain obscure tendencies of the presentand that these changes, when consummated, will be seen to be butcontinuations of changes that have been already effected. We feel ratherthat our language is practically a fixed system and that what slightchanges are destined to take place in it are as likely to move in onedirection as another. The feeling is fallacious. Our very uncertainty asto the impending details of change makes the eventual consistency oftheir direction all the more impressive. 

[Footnote 129: Or rather apprehended, for we do not, in sober fact, entirely understand it as yet. ]

[Footnote 130: Not ultimately random, of course, only relatively so. ]

Sometimes we can feel where the drift is taking us even while westruggle against it. Probably the majority of those who read these wordsfeel that it is quite "incorrect" to say "Who did you see?" We readersof many books are still very careful to say "Whom did you see?" but wefeel a little uncomfortable (uncomfortably proud, it may be) in theprocess. We are likely to avoid the locution altogether and to say "Whowas it you saw?" conserving literary tradition (the "whom") with thedignity of silence. [131] The folk makes no apology. "Whom did you see?"might do for an epitaph, but "Who did you see?" is the natural form foran eager inquiry. It is of course the uncontrolled speech of the folk towhich we must look for advance information as to the general linguisticmovement. It is safe to prophesy that within a couple of hundred yearsfrom to-day not even the most learned jurist will be saying "Whom didyou see?" By that time the "whom" will be as delightfully archaic as theElizabethan "his" for "its. "[132] No logical or historical argument willavail to save this hapless "whom. " The demonstration "I: me = he: him =who: whom" will be convincing in theory and will go unheeded inpractice. 

[Footnote 131: In relative clauses too we tend to avoid the objectiveform of "who. " Instead of "The man whom I saw" we are likely to say "Theman that I saw" or "The man I saw. "]

[Footnote 132: "Its" was at one time as impertinent a departure as the"who" of "Who did you see?" It forced itself into English because theold cleavage between masculine, feminine, and neuter was being slowlyand powerfully supplemented by a new one between thing-class andanimate-class. The latter classification proved too vital to allow usageto couple males and things ("his") as against females ("her"). The form"its" had to be created on the analogy of words like "man's, " to satisfythe growing form feeling. The drift was strong enough to sanction agrammatical blunder. ]

Even now we may go so far as to say that the majority of us are secretlywishing they could say "Who did you see?" It would be a weight off theirunconscious minds if some divine authority, overruling the lifted fingerof the pedagogue, gave them _carte blanche_. But we cannot too franklyanticipate the drift and maintain caste. We must affect ignoranceof whither we are going and rest content with our mentalconflict--uncomfortable conscious acceptance of the "whom, " unconsciousdesire for the "who. "[133] Meanwhile we indulge our sneaking desire forthe forbidden locution by the use of the "who" in certain twilight casesin which we can cover up our fault by a bit of unconscious specialpleading. Imagine that some one drops the remark when you are notlistening attentively, "John Smith is coming to-night. " You have notcaught the name and ask, not "Whom did you say?" but "Who did you say?"There is likely to be a little hesitation in the choice of the form, butthe precedent of usages like "Whom did you see?" will probably not seemquite strong enough to induce a "Whom did you say?" Not quite relevantenough, the grammarian may remark, for a sentence like "Who did yousay?" is not strictly analogous to "Whom did you see?" or "Whom did youmean?" It is rather an abbreviated form of some such sentence as "Who, did you say, is coming to-night?" This is the special pleading that Ihave referred to, and it has a certain logic on its side. Yet the caseis more hollow than the grammarian thinks it to be, for in reply to sucha query as "You're a good hand at bridge, John, aren't you?" John, alittle taken aback, might mutter "Did you say me?" hardly "Did you sayI?" Yet the logic for the latter ("Did you say I was a good hand atbridge?") is evident. The real point is that there is not enoughvitality in the "whom" to carry it over such little difficultiesas a "me" can compass without a thought. The proportion"I : me = he : him = who : whom" is logically and historically sound, butpsychologically shaky. "Whom did you see?" is correct, but there issomething false about its correctness. 

[Footnote 133: Psychoanalysts will recognize the mechanism. Themechanisms of "repression of impulse" and of its symptomaticsymbolization can be illustrated in the most unexpected corners ofindividual and group psychology. A more general psychology than Freud'swill eventually prove them to be as applicable to the groping forabstract form, the logical or esthetic ordering of experience, as to thelife of the fundamental instincts. ]

It is worth looking into the reason for our curious reluctance to uselocutions involving the word "whom" particularly in its interrogativesense. The only distinctively objective forms which we still possess inEnglish are _me_, _him_, _her_ (a little blurred because of its identitywith the possessive _her_), _us_, _them_, and _whom_. In all other casesthe objective has come to be identical with the subjective--that is, inouter form, for we are not now taking account of position in thesentence. We observe immediately in looking through the list ofobjective forms that _whom_ is psychologically isolated. _Me_, _him_, _her_, _us_, and _them_ form a solid, well-integrated group of objectivepersonal pronouns parallel to the subjective series _I_, _he_, _she_, _we_, _they_. The forms _who_ and _whom_ are technically "pronouns" butthey are not felt to be in the same box as the personal pronouns. _Whom_has clearly a weak position, an exposed flank, for words of a feathertend to flock together, and if one strays behind, it is likely to incurdanger of life. Now the other interrogative and relative pronouns(_which_, _what_, _that_), with which _whom_ should properly flock, donot distinguish the subjective and objective forms. It ispsychologically unsound to draw the line of form cleavage between _whom_and the personal pronouns on the one side, the remaining interrogativeand relative pronouns on the other. The form groups should besymmetrically related to, if not identical with, the function groups. Had _which_, _what_, and _that_ objective forms parallel to _whom_, theposition of this last would be more secure. As it is, there is somethingunesthetic about the word. It suggests a form pattern which is notfilled out by its fellows. The only way to remedy the irregularity ofform distribution is to abandon the _whom_ altogether for we have lostthe power to create new objective forms and cannot remodel our_which_-_what_-_that_ group so as to make it parallel with the smallergroup _who-whom_. Once this is done, _who_ joins its flock and ourunconscious desire for form symmetry is satisfied. We do not secretlychafe at "Whom did you see?" without reason. [134]

[Footnote 134: Note that it is different with _whose_. This has not thesupport of analogous possessive forms in its own functional group, butthe analogical power of the great body of possessives of nouns (_man's_, _boy's_) as well as of certain personal pronouns (_his_, _its_; aspredicated possessive also _hers_, _yours_, _theirs_) is sufficient togive it vitality. ]

But the drift away from _whom_ has still other determinants. The words_who_ and _whom_ in their interrogative sense are psychologicallyrelated not merely to the pronouns _which_ and _what_, but to a group ofinterrogative adverbs--_where_, _when_, _how_--all of which areinvariable and generally emphatic. I believe it is safe to infer thatthere is a rather strong feeling in English that the interrogativepronoun or adverb, typically an emphatic element in the sentence, shouldbe invariable. The inflective _-m_ of _whom_ is felt as a drag upon therhetorical effectiveness of the word. It needs to be eliminated if theinterrogative pronoun is to receive all its latent power. There is stilla third, and a very powerful, reason for the avoidance of _whom_. Thecontrast between the subjective and objective series of personalpronouns (_I_, _he_, _she_, _we_, _they_: _me_, _him_, _her_, _us_, _them_) is in English associated with a difference of position. We say_I see the man_ but _the man sees me_; _he told him_, never _him hetold_ or _him told he_. Such usages as the last two are distinctlypoetic and archaic; they are opposed to the present drift of thelanguage. Even in the interrogative one does not say _Him did you see?_It is only in sentences of the type _Whom did you see?_ that aninflected objective before the verb is now used at all. On the otherhand, the order in _Whom did you see?_ is imperative because of itsinterrogative form; the interrogative pronoun or adverb normally comesfirst in the sentence (_What are you doing?_ _When did he go?_ _Whereare you from?_). In the "whom" of _Whom did you see?_ there isconcealed, therefore, a conflict between the order proper to a sentencecontaining an inflected objective and the order natural to a sentencewith an interrogative pronoun or adverb. The solution _Did you seewhom?_ or _You saw whom?_[135] is too contrary to the idiomatic drift ofour language to receive acceptance. The more radical solution _Who didyou see?_ is the one the language is gradually making for. 

[Footnote 135: Aside from certain idiomatic usages, as when _You sawwhom?_ is equivalent to _You saw so and so and that so and so is who?_In such sentences _whom_ is pronounced high and lingeringly to emphasizethe fact that the person just referred to by the listener is not knownor recognized. ]

These three conflicts--on the score of form grouping, of rhetoricalemphasis, and of order--are supplemented by a fourth difficulty. Theemphatic _whom_, with its heavy build (half-long vowel followed bylabial consonant), should contrast with a lightly tripping syllableimmediately following. In _whom did_, however, we have an involuntaryretardation that makes the locution sound "clumsy. " This clumsiness is aphonetic verdict, quite apart from the dissatisfaction due to thegrammatical factors which we have analyzed. The same prosodic objectiondoes not apply to such parallel locutions as _what did_ and _when did_. The vowels of _what_ and _when_ are shorter and their final consonantsmelt easily into the following _d_, which is pronounced in the sametongue position as _t_ and _n_. Our instinct for appropriate rhythmsmakes it as difficult for us to feel content with _whom did_ as for apoet to use words like _dreamed_ and _hummed_ in a rapid line. Neithercommon feeling nor the poet's choice need be at all conscious. It may bethat not all are equally sensitive to the rhythmic flow of speech, butit is probable that rhythm is an unconscious linguistic determinant evenwith those who set little store by its artistic use. In any event thepoet's rhythms can only be a more sensitive and stylicized applicationof rhythmic tendencies that are characteristic of the daily speech ofhis people. 

We have discovered no less than four factors which enter into our subtledisinclination to say "Whom did you see?" The uneducated folk that says"Who did you see?" with no twinge of conscience has a more acute flairfor the genuine drift of the language than its students. Naturally thefour restraining factors do not operate independently. Their separateenergies, if we may make bold to use a mechanical concept, are"canalized" into a single force. This force or minute embodiment of thegeneral drift of the language is psychologically registered as a slighthesitation in using the word _whom_. The hesitation is likely to bequite unconscious, though it may be readily acknowledged when attentionis called to it. The analysis is certain to be unconscious, or ratherunknown, to the normal speaker. [136] How, then, can we be certain insuch an analysis as we have undertaken that all of the assigneddeterminants are really operative and not merely some one of them?Certainly they are not equally powerful in all cases. Their values arevariable, rising and falling according to the individual and thelocution. [137] But that they really exist, each in its own right, maysometimes be tested by the method of elimination. If one or other of thefactors is missing and we observe a slight diminution in thecorresponding psychological reaction ("hesitation" in our case), we mayconclude that the factor is in other uses genuinely positive. The secondof our four factors applies only to the interrogative use of _whom_, thefourth factor applies with more force to the interrogative than to therelative. We can therefore understand why a sentence like _Is he the manwhom you referred to?_ though not as idiomatic as _Is he the man (that)you referred to?_ (remember that it sins against counts one and three), is still not as difficult to reconcile with our innate feeling forEnglish expression as _Whom did you see?_ If we eliminate the fourthfactor from the interrogative usage, [138] say in _Whom are you lookingat?_ where the vowel following _whom_ relieves this word of its phoneticweight, we can observe, if I am not mistaken, a lesser reluctance to usethe _whom_. _Who are you looking at?_ might even sound slightlyoffensive to ears that welcome _Who did you see?_

[Footnote 136: Students of language cannot be entirely normal in theirattitude towards their own speech. Perhaps it would be better to say"naïve" than "normal. "]

[Footnote 137: It is probably this _variability of value_ in thesignificant compounds of a general linguistic drift that is responsiblefor the rise of dialectic variations. Each dialect continues the generaldrift of the common parent, but has not been able to hold fast toconstant values for each component of the drift. Deviations as to thedrift itself, at first slight, later cumulative, are thereforeunavoidable. ]

[Footnote 138: Most sentences beginning with interrogative _whom_ arelikely to be followed by _did_ or _does_, _do_. Yet not all. ]

We may set up a scale of "hesitation values" somewhat after thisfashion:

Value 1: factors 1, 3. "The man whom I referred to. "Value 2: factors 1, 3, 4. "The man whom they referred to. "Value 3: factors 1, 2, 3. "Whom are you looking at?"Value 4: factors 1, 2, 3, 4. "Whom did you see?"

We may venture to surmise that while _whom_ will ultimately disappearfrom English speech, locutions of the type _Whom did you see?_ will beobsolete when phrases like _The man whom I referred to_ are still inlingering use. It is impossible to be certain, however, for we can nevertell if we have isolated all the determinants of a drift. In ourparticular case we have ignored what may well prove to be a controllingfactor in the history of _who_ and _whom_ in the relative sense. This isthe unconscious desire to leave these words to their interrogativefunction and to concentrate on _that_ or mere word order as expressionsof the relative (e. G. , _The man that I referred to_ or _The man Ireferred to_). This drift, which does not directly concern the use of_whom_ as such (merely of _whom_ as a form of _who_), may have made therelative _who_ obsolete before the other factors affecting relative_whom_ have run their course. A consideration like this is instructivebecause it indicates that knowledge of the general drift of a languageis insufficient to enable us to see clearly what the drift is headingfor. We need to know something of the relative potencies and speeds ofthe components of the drift. 

It is hardly necessary to say that the particular drifts involved in theuse of _whom_ are of interest to us not for their own sake but assymptoms of larger tendencies at work in the language. At least threedrifts of major importance are discernible. Each of these has operatedfor centuries, each is at work in other parts of our linguisticmechanism, each is almost certain to continue for centuries, possiblymillennia. The first is the familiar tendency to level the distinctionbetween the subjective and the objective, itself but a late chapter inthe steady reduction of the old Indo-European system of syntactic cases. This system, which is at present best preserved in Lithuanian, [139] wasalready considerably reduced in the old Germanic language of whichEnglish, Dutch, German, Danish, and Swedish are modern dialectic forms. The seven Indo-European cases (nominative genitive, dative, accusative, ablative, locative, instrumental) had been already reduced to four(nominative genitive, dative, accusative). We know this from a carefulcomparison of and reconstruction based on the oldest Germanic dialectsof which we still have records (Gothic, Old Icelandic, Old High German, Anglo-Saxon). In the group of West Germanic dialects, for the study ofwhich Old High German, Anglo-Saxon, Old Frisian, and Old Saxon are ouroldest and most valuable sources, we still have these four cases, butthe phonetic form of the case syllables is already greatly reduced andin certain paradigms particular cases have coalesced. The case system ispractically intact but it is evidently moving towards furtherdisintegration. Within the Anglo-Saxon and early Middle English periodthere took place further changes in the same direction. The phoneticform of the case syllables became still further reduced and thedistinction between the accusative and the dative finally disappeared. The new "objective" is really an amalgam of old accusative and dativeforms; thus, _him_, the old dative (we still say _I give him the book_, not "abbreviated" from _I give to him_; compare Gothic _imma_, modernGerman _ihm_), took over the functions of the old accusative(Anglo-Saxon _hine_; compare Gothic _ina_, Modern German _ihn_) anddative. The distinction between the nominative and accusative wasnibbled away by phonetic processes and morphological levelings untilonly certain pronouns retained distinctive subjective and objectiveforms. 

[Footnote 139: Better, indeed, than in our oldest Latin and Greekrecords. The old Indo-Iranian languages alone (Sanskrit, Avestan) showan equally or more archaic status of the Indo-European parent tongue asregards case forms. ]

In later medieval and in modern times there have been comparatively fewapparent changes in our case system apart from the gradual replacementof _thou_--_thee_ (singular) and subjective _ye_--objective _you_(plural) by a single undifferentiated form _you_. All the while, however, the case system, such as it is (subjective-objective, reallyabsolutive, and possessive in nouns; subjective, objective, andpossessive in certain pronouns) has been steadily weakening inpsychological respects. At present it is more seriously undermined thanmost of us realize. The possessive has little vitality except in thepronoun and in animate nouns. Theoretically we can still say _the moon'sphases_ or _a newspaper's vogue_; practically we limit ourselves prettymuch to analytic locutions like _the phases of the moon_ and _the vogueof a newspaper_. The drift is clearly toward the limitation, ofpossessive forms to animate nouns. All the possessive pronominal formsexcept _its_ and, in part, _their_ and _theirs_, are also animate. It issignificant that _theirs_ is hardly ever used in reference to inanimatenouns, that there is some reluctance to so use _their_, and that _its_also is beginning to give way to _of it_. _The appearance of it_ or _thelooks of it_ is more in the current of the language than _itsappearance_. It is curiously significant that _its young_ (referring toan animal's cubs) is idiomatically preferable to _the young of it_. Theform is only ostensibly neuter, in feeling it is animate;psychologically it belongs with _his children_, not with _the pieces ofit_. Can it be that so common a word as _its_ is actually beginning tobe difficult? Is it too doomed to disappear? It would be rash to saythat it shows signs of approaching obsolescence, but that it is steadilyweakening is fairly clear. [140] In any event, it is not too much to saythat there is a strong drift towards the restriction of the inflectedpossessive forms to animate nouns and pronouns. 

[Footnote 140: Should _its_ eventually drop out, it will have had acurious history. It will have played the rôle of a stop-gap between_his_ in its non-personal use (see footnote 11, page 167) and the lateranalytic of _it_. ]

[Transcriber's note: Footnote 140 refers to Footnote 132, beginning online 5142. ]

How is it with the alternation of subjective and objective in thepronoun? Granted that _whom_ is a weak sister, that the two cases havebeen leveled in _you_ (in _it_, _that_, and _what_ they were neverdistinct, so far as we can tell[141]), and that _her_ as an objective isa trifle weak because of its formal identity with the possessive _her_, is there any reason to doubt the vitality of such alternations as _I seethe man_ and _the man sees me_? Surely the distinction betweensubjective _I_ and objective _me_, between subjective _he_ and objective_him_, and correspondingly for other personal pronouns, belongs to thevery core of the language. We can throw _whom_ to the dogs, somehow makeshift to do without an _its_, but to level _I_ and _me_ to a singlecase--would that not be to un-English our language beyond recognition?There is no drift toward such horrors as _Me see him_ or _I see he_. True, the phonetic disparity between _I_ and _me_, _he_ and _him_, _we_and _us_, has been too great for any serious possibility of formleveling. It does not follow that the case distinction as such is stillvital. One of the most insidious peculiarities of a linguistic drift isthat where it cannot destroy what lies in its way it renders itinnocuous by washing the old significance out of it. It turns its veryenemies to its own uses. This brings us to the second of the majordrifts, the tendency to fixed position in the sentence, determined bythe syntactic relation of the word. 

[Footnote 141: Except in so far as _that_ has absorbed otherfunctions than such as originally belonged to it. It was only anominative-accusative neuter to begin with. ]

We need not go into the history of this all-important drift. It isenough to know that as the inflected forms of English became scantier, as the syntactic relations were more and more inadequately expressed bythe forms of the words themselves, position in the sentence graduallytook over functions originally foreign to it. _The man_ in _the man seesthe dog_ is subjective; in _the dog sees the man_, objective. Strictlyparallel to these sentences are _he sees the dog_ and _the dog seeshim_. Are the subjective value of _he_ and the objective value of _him_entirely, or even mainly, dependent on the difference of form? I doubtit. We could hold to such a view if it were possible to say _the dogsees he_ or _him sees the dog_. It was once possible to say such things, but we have lost the power. In other words, at least part of the casefeeling in _he_ and _him_ is to be credited to their position before orafter the verb. May it not be, then, that _he_ and _him_, _we_ and _us_, are not so much subjective and objective forms as pre-verbal andpost-verbal[142] forms, very much as _my_ and _mine_ are now pre-nominaland post-nominal forms of the possessive (_my father_ but _father mine_;_it is my book_ but _the book is mine_)? That this interpretationcorresponds to the actual drift of the English language is againindicated by the language of the folk. The folk says _it is me_, not _itis I_, which is "correct" but just as falsely so as the _whom did yousee_? that we have analyzed. _I'm the one_, _it's me_; _we're the ones_, _it's us that will win out_--such are the live parallelisms in Englishto-day. There is little doubt that _it is I_ will one day be asimpossible in English as _c'est je_, for _c'est moi_, is now in French. 

[Footnote 142: Aside from the interrogative: _am I?_ _is he?_ Emphasiscounts for something. There is a strong tendency for the old "objective"forms to bear a stronger stress than the "subjective" forms. This is whythe stress in locutions like _He didn't go, did he?_ and _isn't he?_ isthrown back on the verb; it is not a matter of logical emphasis. ]

How differently our _I_: _me_ feels than in Chaucer's day is shown bythe Chaucerian _it am I_. Here the distinctively subjective aspect ofthe _I_ was enough to influence the form of the preceding verb in spiteof the introductory _it_; Chaucer's locution clearly felt more like aLatin _sum ego_ than a modern _it is I_ or colloquial _it is me_. Wehave a curious bit of further evidence to prove that the Englishpersonal pronouns have lost some share of their original syntacticforce. Were _he_ and _she_ subjective forms pure and simple, were theynot striving, so to speak, to become caseless absolutives, like _man_ orany other noun, we should not have been able to coin such compounds as_he-goat_ and _she-goat_, words that are psychologically analogous to_bull-moose_ and _mother-bear_. Again, in inquiring about a new-bornbaby, we ask _Is it a he or a she?_ quite as though _he_ and _she_ werethe equivalents of _male_ and _female_ or _boy_ and _girl_. All in all, we may conclude that our English case system is weaker than it looks andthat, in one way or another, it is destined to get itself reduced to anabsolutive (caseless) form for all nouns and pronouns but those that areanimate. Animate nouns and pronouns are sure to have distinctivepossessive forms for an indefinitely long period. 

Meanwhile observe that the old alignment of case forms is being invadedby two new categories--a positional category (pre-verbal, post-verbal)and a classificatory category (animate, inanimate). The facts that inthe possessive animate nouns and pronouns are destined to be more andmore sharply distinguished from inanimate nouns and pronouns (_theman's_, but _of the house_; _his_, but _of it_) and that, on the whole, it is only animate pronouns that distinguish pre-verbal and post-verbalforms[143] are of the greatest theoretical interest. They show that, however the language strive for a more and more analytic form, it is byno means manifesting a drift toward the expression of "pure" relationalconcepts in the Indo-Chinese manner. [144] The insistence on theconcreteness of the relational concepts is clearly stronger than thedestructive power of the most sweeping and persistent drifts that weknow of in the history and prehistory of our language. 

[Footnote 143: _They_: _them_ as an inanimate group may be looked uponas a kind of borrowing from the animate, to which, in feeling, it moreproperly belongs. ]

[Footnote 144: See page 155. ]

[Transcriber's note: Footnote 144 refers to the paragraph beginning online 4795. ]

The drift toward the abolition of most case distinctions and thecorrelative drift toward position as an all-important grammatical methodare accompanied, in a sense dominated, by the last of the three majordrifts that I have referred to. This is the drift toward the invariableword. In analyzing the "whom" sentence I pointed out that the rhetoricalemphasis natural to an interrogative pronoun lost something by its formvariability (_who_, _whose_, _whom_). This striving for a simple, unnuanced correspondence between idea and word, as invariable as may be, is very strong in English. It accounts for a number of tendencies whichat first sight seem unconnected. Certain well-established forms, likethe present third person singular _-s_ of _works_ or the plural _-s_ of_books_, have resisted the drift to invariable words, possibly becausethey symbolize certain stronger form cravings that we do not yet fullyunderstand. It is interesting to note that derivations that get awaysufficiently from the concrete notion of the radical word to exist asindependent conceptual centers are not affected by this elusive drift. As soon as the derivation runs danger of being felt as a mere nuancingof, a finicky play on, the primary concept, it tends to be absorbed bythe radical word, to disappear as such. English words crave spacesbetween them, they do not like to huddle in clusters of slightlydivergent centers of meaning, each edging a little away from the rest. _Goodness_, a noun of quality, almost a noun of relation, that takes itscue from the concrete idea of "good" without necessarily predicatingthat quality (e. G. , _I do not think much of his goodness_) issufficiently spaced from _good_ itself not to need fear absorption. Similarly, _unable_ can hold its own against _able_ because it destroysthe latter's sphere of influence; _unable_ is psychologically asdistinct from _able_ as is _blundering_ or _stupid_. It is differentwith adverbs in _-ly_. These lean too heavily on their adjectives tohave the kind of vitality that English demands of its words. _Do itquickly!_ drags psychologically. The nuance expressed by _quickly_ istoo close to that of _quick_, their circles of concreteness are toonearly the same, for the two words to feel comfortable together. Theadverbs in _-ly_ are likely to go to the wall in the not too distantfuture for this very reason and in face of their obvious usefulness. Another instance of the sacrifice of highly useful forms to thisimpatience of nuancing is the group _whence_, _whither_, _hence_, _hither_, _thence_, _thither_. They could not persist in live usagebecause they impinged too solidly upon the circles of meaningrepresented by the words _where_, _here_ and _there_. In saying_whither_ we feel too keenly that we repeat all of _where_. That we addto _where_ an important nuance of direction irritates rather thansatisfies. We prefer to merge the static and the directive (_Where doyou live?_ like _Where are you going?_) or, if need be, to overdo alittle the concept of direction (_Where are you running to?_). 

Now it is highly symptomatic of the nature of the drift away from wordclusters that we do not object to nuances as such, we object to havingthe nuances formally earmarked for us. As a matter of fact ourvocabulary is rich in near-synonyms and in groups of words that arepsychologically near relatives, but these near-synonyms and these groupsdo not hang together by reason of etymology. We are satisfied with_believe_ and _credible_ just because they keep aloof from each other. _Good_ and _well_ go better together than _quick_ and _quickly_. TheEnglish vocabulary is a rich medley because each English word wants itsown castle. Has English long been peculiarly receptive to foreign wordsbecause it craves the staking out of as many word areas as possible, or, conversely, has the mechanical imposition of a flood of French and Latinloan-words, unrooted in our earlier tradition, so dulled our feeling forthe possibilities of our native resources that we are allowing these toshrink by default? I suspect that both propositions are true. Each feedson the other. I do not think it likely, however, that the borrowings inEnglish have been as mechanical and external a process as they aregenerally represented to have been. There was something about theEnglish drift as early as the period following the Norman Conquest thatwelcomed the new words. They were a compensation for something that wasweakening within. 

VIII

LANGUAGE AS A HISTORICAL PRODUCT: PHONETIC LAW

I have preferred to take up in some detail the analysis of ourhesitation in using a locution like "Whom did you see?" and to point tosome of the English drifts, particular and general, that are implied bythis hesitation than to discuss linguistic change in the abstract. Whatis true of the particular idiom that we started with is true ofeverything else in language. Nothing is perfectly static. Every word, every grammatical element, every locution, every sound and accent is aslowly changing configuration, molded by the invisible and impersonaldrift that is the life of language. The evidence is overwhelming thatthis drift has a certain consistent direction. Its speed variesenormously according to circumstances that it is not always easy todefine. We have already seen that Lithuanian is to-day nearer itsIndo-European prototype than was the hypothetical Germanic mother-tonguefive hundred or a thousand years before Christ. German has moved moreslowly than English; in some respects it stands roughly midway betweenEnglish and Anglo-Saxon, in others it has of course diverged from theAnglo-Saxon line. When I pointed out in the preceding chapter thatdialects formed because a language broken up into local segments couldnot move along the same drift in all of these segments, I meant ofcourse that it could not move along identically the same drift. Thegeneral drift of a language has its depths. At the surface the currentis relatively fast. In certain features dialects drift apart rapidly. Bythat very fact these features betray themselves as less fundamental tothe genius of the language than the more slowly modifiable features inwhich the dialects keep together long after they have grown to bemutually alien forms of speech. But this is not all. The momentum of themore fundamental, the pre-dialectic, drift is often such that languageslong disconnected will pass through the same or strikingly similarphases. In many such cases it is perfectly clear that there could havebeen no dialectic interinfluencing. 

These parallelisms in drift may operate in the phonetic as well as inthe morphological sphere, or they may affect both at the same time. Hereis an interesting example. The English type of plural represented by_foot_: _feet_, _mouse_: _mice_ is strictly parallel to the German_Fuss_: _Füsse_, _Maus_: _Mäuse_. One would be inclined to surmise thatthese dialectic forms go back to old Germanic or West-Germanicalternations of the same type. But the documentary evidence showsconclusively that there could have been no plurals of this type inprimitive Germanic. There is no trace of such vocalic mutation("umlaut") in Gothic, our most archaic Germanic language. Moresignificant still is the fact that it does not appear in our oldest OldHigh German texts and begins to develop only at the very end of the OldHigh German period (circa 1000 A. D. ). In the Middle High German periodthe mutation was carried through in all dialects. The typical Old HighGerman forms are singular _fuoss_, plural _fuossi_;[145] singular _mus_, plural _musi_. The corresponding Middle High German forms are _fuoss_, _füesse_; _mus_, _müse_. Modern German _Fuss_: _Füsse_, _Maus_: _Mäuse_are the regular developments of these medieval forms. Turning toAnglo-Saxon, we find that our modern English forms correspond to _fot_, _fet_; _mus_, _mys_. [146] These forms are already in use in the earliestEnglish monuments that we possess, dating from the eighth century, andthus antedate the Middle High German forms by three hundred years ormore. In other words, on this particular point it took German at leastthree hundred years to catch up with a phonetic-morphological drift[147]that had long been under way in English. The mere fact that the affectedvowels of related words (Old High German _uo_, Anglo-Saxon _o_) are notalways the same shows that the affection took place at different periodsin German and English. [148] There was evidently some general tendency orgroup of tendencies at work in early Germanic, long before English andGerman had developed as such, that eventually drove both of thesedialects along closely parallel paths. 

[Footnote 145: I have changed the Old and Middle High German orthographyslightly in order to bring it into accord with modern usage. Thesepurely orthographical changes are immaterial. The _u_ of _mus_ is a longvowel, very nearly like the _oo_ of English _moose_. ]

[Footnote 146: The vowels of these four words are long; _o_ as in_rode_, _e_ like _a_ of _fade_, _u_ like _oo_ of _brood_, _y_ likeGerman _ü_. ]

[Footnote 147: Or rather stage in a drift. ]

[Footnote 148: Anglo-Saxon _fet_ is "unrounded" from an older _föt_, which is phonetically related to _fot_ precisely as is _mys_ (i. E. , _müs_) to _mus_. Middle High German _ue_ (Modern German _u_) did notdevelop from an "umlauted" prototype of Old High German _uo_ andAnglo-Saxon _o_, but was based directly on the dialectic _uo_. Theunaffected prototype was long _o_. Had this been affected in theearliest Germanic or West-Germanic period, we should have had apre-German alternation _fot_: _föti_; this older _ö_ could not well haveresulted in _ue_. Fortunately we do not need inferential evidence inthis case, yet inferential comparative methods, if handled with care, may be exceedingly useful. They are indeed indispensable to thehistorian of language. ]

How did such strikingly individual alternations as _fot_: _fet_, _fuoss_: _füesse_ develop? We have now reached what is probably themost central problem in linguistic history, gradual phonetic change. "Phonetic laws" make up a large and fundamental share of thesubject-matter of linguistics. Their influence reaches far beyond theproper sphere of phonetics and invades that of morphology, as we shallsee. A drift that begins as a slight phonetic readjustment orunsettlement may in the course of millennia bring about the mostprofound structural changes. The mere fact, for instance, that there isa growing tendency to throw the stress automatically on the firstsyllable of a word may eventually change the fundamental type of thelanguage, reducing its final syllables to zero and driving it to the useof more and more analytical or symbolic[149] methods. The Englishphonetic laws involved in the rise of the words _foot_, _feet_, _mouse_and _mice_ from their early West-Germanic prototypes _fot_, _foti_, _mus_, _musi_[150] may be briefly summarized as follows:

[Footnote 149: See page 133. ]

[Transcriber's note: Footnote 149 refers to the paragraph beginning online 4081. ]

[Footnote 150: Primitive Germanic _fot(s)_, _fotiz_, _mus_, _musiz_;Indo-European _pods_, _podes_, _mus_, _muses_. The vowels of the firstsyllables are all long. ]

1. In _foti_ "feet" the long _o_ was colored by the following _i_ tolong _ö_, that is, _o_ kept its lip-rounded quality and its middleheight of tongue position but anticipated the front tongue position ofthe _i_; _ö_ is the resulting compromise. This assimilatory change wasregular, i. E. , every accented long _o_ followed by an _i_ in thefollowing syllable automatically developed to long _ö_; hence _tothi_"teeth" became _töthi_, _fodian_ "to feed" became _födian_. At firstthere is no doubt the alternation between _o_ and _ö_ was not felt asintrinsically significant. It could only have been an unconsciousmechanical adjustment such as may be observed in the speech of manyto-day who modify the "oo" sound of words like _you_ and _few_ in thedirection of German _ü_ without, however, actually departing far enoughfrom the "oo" vowel to prevent their acceptance of _who_ and _you_ assatisfactory rhyming words. Later on the quality of the _ö_ vowel musthave departed widely enough from that of _o_ to enable _ö_ to rise inconsciousness[151] as a neatly distinct vowel. As soon as this happened, the expression of plurality in _föti_, _töthi_, and analogous wordsbecame symbolic and fusional, not merely fusional. 

[Footnote 151: Or in that unconscious sound patterning which is ever onthe point of becoming conscious. See page 57. ]

[Transcriber's note: Footnote 151 refers to the paragraph beginning online 1797. ]

2. In _musi_ "mice" the long _u_ was colored by the following _i_ tolong _ü_. This change also was regular; _lusi_ "lice" became _lüsi_, _kui_ "cows" became _küi_ (later simplified to _kü_; still preserved as_ki-_ in _kine_), _fulian_ "to make foul" became _fülian_ (stillpreserved as _-file_ in _defile_). The psychology of this phonetic lawis entirely analogous to that of 1. 

3. The old drift toward reducing final syllables, a rhythmic consequenceof the strong Germanic stress on the first syllable, now manifesteditself. The final _-i_, originally an important functional element, hadlong lost a great share of its value, transferred as that was to thesymbolic vowel change (_o_: _ö_). It had little power of resistance, therefore, to the drift. It became dulled to a colorless _-e_; _föti_became _föte_. 

4. The weak _-e_ finally disappeared. Probably the forms _föte_ and_föt_ long coexisted as prosodic variants according to the rhythmicrequirements of the sentence, very much as _Füsse_ and _Füss'_ nowcoexist in German. 

5. The _ö_ of _föt_ became "unrounded" to long _e_ (our present _a_ of_fade_). The alternation of _fot_: _foti_, transitionally _fot_: _föti_, _föte_, _föt_, now appears as _fot_: _fet_. Analogously, _töth_ appearsas _teth_, _födian_ as _fedian_, later _fedan_. The new long _e_-vowel"fell together" with the older _e_-vowel already existent (e. G. , _her_"here, " _he_ "he"). Henceforward the two are merged and their laterhistory is in common. Thus our present _he_ has the same vowel as_feet_, _teeth_, and _feed_. In other words, the old sound pattern _o_, _e_, after an interim of _o_, _ö_, _e_, reappeared as _o_, _e_, exceptthat now the _e_ had greater "weight" than before. 

6. _Fot_: _fet_, _mus_: _müs_ (written _mys_) are the typical forms ofAnglo-Saxon literature. At the very end of the Anglo-Saxon period, sayabout 1050 to 1100 A. D. , the _ü_, whether long or short, becameunrounded to _i_. _Mys_ was then pronounced _mis_ with long _i_ (rhymingwith present _niece_). The change is analogous to 5, but takes placeseveral centuries later. 

7. In Chaucer's day (circa 1350-1400 A. D. ) the forms were still_fot_: _fet_ (written _foot_, _feet_) and _mus_: _mis_ (written veryvariably, but _mous_, _myse_ are typical). About 1500 all the long_i_-vowels, whether original (as in _write_, _ride_, _wine_) orunrounded from Anglo-Saxon _ü_ (as in _hide_, _bride_, _mice_, _defile_), became diphthongized to _ei_ (i. E. , _e_ of _met_ + short_i_). Shakespeare pronounced _mice_ as _meis_ (almost the same as thepresent Cockney pronunciation of _mace_). 

8. About the same time the long _u_-vowels were diphthongized to _ou_(i. E. , _o_ of present Scotch _not_ + _u_ of _full_). The Chaucerian_mus_: _mis_ now appears as the Shakespearean _mous_: _meis_. Thischange may have manifested itself somewhat later than 7; all Englishdialects have diphthongized old Germanic long _i_, [152] but the longundiphthongized _u_ is still preserved in Lowland Scotch, in which_house_ and _mouse_ rhyme with our _loose_. 7 and 8 are analogousdevelopments, as were 5 and 6; 8 apparently lags behind 7 as 6, centuries earlier, lagged behind 7. 

[Footnote 152: As have most Dutch and German dialects. ]

9. Some time before 1550 the long _e_ of _fet_ (written _feet_) took theposition that had been vacated by the old long _i_, now diphthongized(see 7), i. E. , _e_ took the higher tongue position of _i_. Our (andShakespeare's) "long _e_" is, then, phonetically the same as the oldlong _i_. _Feet_ now rhymed with the old _write_ and the present _beat_. 

10. About the same time the long _o_ of _fot_ (written _foot_) took theposition that had been vacated by the old long _u_, now diphthongized(see 8), i. E. , _o_ took the higher tongue position of _u_. Our (andShakespeare's) "long _oo_" is phonetically the same as the old long _u_. _Foot_ now rhymed with the old _out_ and the present _boot_. Tosummarize 7 to 10, Shakespeare pronounced _meis_, _mous_, _fit_, _fut_, of which _meis_ and _mous_ would affect our ears as a rather "mincing"rendering of our present _mice_ and _mouse_, _fit_ would soundpractically identical with (but probably a bit more "drawled" than) ourpresent _feet_, while _foot_, rhyming with _boot_, would now be set downas "broad Scotch. "

11. Gradually the first vowel of the diphthong in _mice_ (see 7) wasretracted and lowered in position. The resulting diphthong now varies indifferent English dialects, but _ai_ (i. E. , _a_ of _father_, butshorter, + short _i_) may be taken as a fairly accurate rendering of itsaverage quality. [153] What we now call the "long _i_" (of words like_ride, bite, mice_) is, of course, an _ai_-diphthong. _Mice_ is nowpronounced _mais_. 

[Footnote 153: At least in America. ]

12. Analogously to 11, the first vowel of the diphthong in _mouse_ (see8) was unrounded and lowered in position. The resulting diphthong may bephonetically rendered _au_, though it too varies considerably accordingto dialect. _Mouse_, then, is now pronounced _maus_. 

13. The vowel of _foot_ (see 10) became "open" in quality and shorter inquantity, i. E. , it fell together with the old short _u_-vowel of wordslike _full_, _wolf_, _wool_. This change has taken place in a number ofwords with an originally long _u_ (Chaucerian long close _o_), such as_forsook_, _hook_, _book_, _look_, _rook_, _shook_, all of whichformerly had the vowel of _boot_. The older vowel, however, is stillpreserved in most words of this class, such as _fool_, _moon_, _spool_, _stoop_. It is highly significant of the nature of the slow spread of a"phonetic law" that there is local vacillation at present in severalwords. One hears _roof_, _soot_, and _hoop_, for instance, both with the"long" vowel of _boot_ and the "short" of _foot_. It is impossible now, in other words, to state in a definitive manner what is the "phoneticlaw" that regulated the change of the older _foot_ (rhyming with _boot_)to the present _foot_. We know that there is a strong drift towards theshort, open vowel of _foot_, but whether or not all the old "long _oo_"words will eventually be affected we cannot presume to say. If they all, or practically all, are taken by the drift, phonetic law 13 will be as"regular, " as sweeping, as most of the twelve that have preceded it. Ifnot, it may eventually be possible, if past experience is a safe guide, to show that the modified words form a natural phonetic group, that is, that the "law" will have operated under certain definable limitingconditions, e. G. , that all words ending in a voiceless consonant (suchas _p_, _t_, _k_, _f_) were affected (e. G. , _hoof_, _foot_, _look_, _roof_), but that all words ending in the _oo_-vowel or in a voicedconsonant remained unaffected (e. G. , _do_, _food_, _move_, _fool_). Whatever the upshot, we may be reasonably certain that when the"phonetic law" has run its course, the distribution of "long" and"short" vowels in the old _oo_-words will not seem quite as erratic asat the present transitional moment. [154] We learn, incidentally, thefundamental fact that phonetic laws do not work with spontaneousautomatism, that they are simply a formula for a consummated drift thatsets in at a psychologically exposed point and gradually worms its waythrough a gamut of phonetically analogous forms. 

[Footnote 154: It is possible that other than purely phonetic factorsare also at work in the history of these vowels. ]

It will be instructive to set down a table of form sequences, a kind ofgross history of the words _foot_, _feet_, _mouse_, _mice_ for the last1500 years:[155]

[Footnote 155: The orthography is roughly phonetic. Pronounce allaccented vowels long except where otherwise indicated, unaccented vowelsshort; give continental values to vowels, not present English ones. ]

 I. _fot_: _foti_; _mus_: _musi_ (West Germanic) II. _fot_: _föti_; _mus_: _müsi_ III. _fot_: _föte_; _mus_: _müse_ IV. _fot_: _föt_; _mus_: _müs_ V. _fot_: _fet_; _mus_: _müs_ (Anglo-Saxon) VI. _fot_: _fet_; _mus_: _mis_(Chaucer) VII. _fot_: _fet_; _mous_: _meis_VIII. _fut_ (rhymes with _boot_): _fit_; _mous_: _meis_ (Shakespeare) IX. _fut_: _fit_; _maus_: _mais_ X. _fut_ (rhymes with _put_): _fit_; _maus_: _mais_ (English of 1900)

It will not be necessary to list the phonetic laws thatgradually differentiated the modern German equivalentsof the original West Germanic forms from theirEnglish cognates. The following table gives a roughidea of the form sequences in German:[156]

[Footnote 156: After I. The numbers are not meant to correspondchronologically to those of the English table. The orthography is againroughly phonetic. ]

 I. _fot_: _foti_; _mus_: _musi_ (West Germanic) II. _foss_:[157] _fossi_; _mus_: _musi_ III. _fuoss_: _fuossi_; _mus_: _musi_ (Old High German) IV. _fuoss_: _füessi_; _mus_: _müsi_ V. _fuoss_: _füesse_; _mus_: _müse_ (Middle High German) VI. _fuoss_: _füesse_; _mus_: _müze_[158] VII. _fuos_: _füese_; _mus_: _müze_VIII. _fuos_: _füese_; _mous_: _möüze_ IX. _fus_: _füse_; _mous_: _möüze_ (Luther) X. _fus_: _füse_; _maus_: _moize_ (German of 1900)

[Footnote 157: I use _ss_ to indicate a peculiar long, voiceless_s_-sound that was etymologically and phonetically distinct from the oldGermanic _s_. It always goes back to an old _t_. In the old sources itis generally written as a variant of _z_, though it is not to beconfused with the modern German _z_ (= _ts_). It was probably a dental(lisped) _s_. ]

[Footnote 158: _Z_ is to be understood as French or English _z_, not inits German use. Strictly speaking, this "z" (intervocalic _-s-_) was notvoiced but was a soft voiceless sound, a sibilant intermediate betweenour _s_ and _z_. In modern North German it has become voiced to _z_. Itis important not to confound this _s_--_z_ with the voicelessintervocalic _s_ that soon arose from the older lisped _ss_. In ModernGerman (aside from certain dialects), old _s_ and _ss_ are not nowdifferentiated when final (_Maus_ and _Fuss_ have identical sibilants), but can still be distinguished as voiced and voiceless _s_ betweenvowels (_Mäuse_ and _Füsse_). ]

We cannot even begin to ferret out and discuss all the psychologicalproblems that are concealed behind these bland tables. Their generalparallelism is obvious. Indeed we might say that to-day the English andGerman forms resemble each other more than does either set the WestGermanic prototypes from which each is independently derived. Each tableillustrates the tendency to reduction of unaccented syllables, thevocalic modification of the radical element under the influence of thefollowing vowel, the rise in tongue position of the long middle vowels(English _o_ to _u_, _e_ to _i_; German _o_ to _uo_ to _u_, _üe_ to_ü_), the diphthongizing of the old high vowels (English _i_ to _ei_ to_ai_; English and German _u_ to _ou_ to _au_; German _ü_ to _öü_ to_oi_). These dialectic parallels cannot be accidental. They are rootedin a common, pre-dialectic drift. 

Phonetic changes are "regular. " All but one (English table, X. ), andthat as yet uncompleted, of the particular phonetic laws represented inour tables affect all examples of the sound in question or, if thephonetic change is conditional, all examples of the same sound that areanalogously circumstanced. [159] An example of the first type of changeis the passage in English of all old long _i_-vowels to diphthongal _ai_via _ei_. The passage could hardly have been sudden or automatic, but itwas rapid enough to prevent an irregularity of development due to crossdrifts. The second type of change is illustrated in the development ofAnglo-Saxon long _o_ to long _e_, via _ö_, under the influence of afollowing _i_. In the first case we may say that _au_ mechanicallyreplaced long _u_, in the second that the old long _o_ "split" into twosounds--long _o_, eventually _u_, and long _e_, eventually _i_. Theformer type of change did no violence to the old phonetic pattern, theformal distribution of sounds into groups; the latter type rearrangedthe pattern somewhat. If neither of the two sounds into which an old one"splits" is a new sound, it means that there has been a phoneticleveling, that two groups of words, each with a distinct sound or soundcombination, have fallen together into one group. This kind of levelingis quite frequent in the history of language. In English, for instance, we have seen that all the old long _ü_-vowels, after they had becomeunrounded, were indistinguishable from the mass of long _i_-vowels. Thismeant that the long _i_-vowel became a more heavily weighted point ofthe phonetic pattern than before. It is curious to observe how oftenlanguages have striven to drive originally distinct sounds into certainfavorite positions, regardless of resulting confusions. [160] In ModernGreek, for instance, the vowel _i_ is the historical resultant of noless than ten etymologically distinct vowels (long and short) anddiphthongs of the classical speech of Athens. There is, then, goodevidence to show that there are general phonetic drifts towardparticular sounds. 

[Footnote 159: In practice phonetic laws have their exceptions, but moreintensive study almost invariably shows that these exceptions are moreapparent than real. They are generally due to the disturbing influenceof morphological groupings or to special psychological reasons whichinhibit the normal progress of the phonetic drift. It is remarkable withhow few exceptions one need operate in linguistic history, aside from"analogical leveling" (morphological replacement). ]

[Footnote 160: These confusions are more theoretical than real, however. A language has countless methods of avoiding practical ambiguities. ]

More often the phonetic drift is of a more general character. It is notso much a movement toward a particular set of sounds as towardparticular types of articulation. The vowels tend to become higher orlower, the diphthongs tend to coalesce into monophthongs, the voicelessconsonants tend to become voiced, stops tend to become spirants. As amatter of fact, practically all the phonetic laws enumerated in the twotables are but specific instances of such far-reaching phonetic drifts. The raising of English long _o_ to _u_ and of long _e_ to _i_, forinstance, was part of a general tendency to raise the position of thelong vowels, just as the change of _t_ to _ss_ in Old High German waspart of a general tendency to make voiceless spirants of the oldvoiceless stopped consonants. A single sound change, even if there is nophonetic leveling, generally threatens to upset the old phonetic patternbecause it brings about a disharmony in the grouping of sounds. Toreëstablish the old pattern without going back on the drift the onlypossible method is to have the other sounds of the series shift inanalogous fashion. If, for some reason or other, _p_ becomes shifted toits voiced correspondent _b_, the old series _p_, _t_, _k_ appears inthe unsymmetrical form _b_, _t_, _k_. Such a series is, in phoneticeffect, not the equivalent of the old series, however it may answer toit in etymology. The general phonetic pattern is impaired to thatextent. But if _t_ and _k_ are also shifted to their voicedcorrespondents _d_ and _g_, the old series is reëstablished in a newform: _b_, _d_, _g_. The pattern as such is preserved, or restored. _Provided that_ the new series _b_, _d_, _g_ does not become confusedwith an old series _b_, _d_, _g_ of distinct historical antecedents. Ifthere is no such older series, the creation of a _b_, _d_, _g_ seriescauses no difficulties. If there is, the old patterning of sounds can bekept intact only by shifting the old _b_, _d_, _g_ sounds in some way. They may become aspirated to _bh_, _dh_, _gh_ or spirantized ornasalized or they may develop any other peculiarity that keeps themintact as a series and serves to differentiate them from other series. And this sort of shifting about without loss of pattern, or with aminimum loss of it, is probably the most important tendency in thehistory of speech sounds. Phonetic leveling and "splitting" counteractit to some extent but, on the whole, it remains the central unconsciousregulator of the course and speed of sound changes. 

The desire to hold on to a pattern, the tendency to "correct" adisturbance by an elaborate chain of supplementary changes, often spreadover centuries or even millennia--these psychic undercurrents oflanguage are exceedingly difficult to understand in terms of individualpsychology, though there can be no denial of their historical reality. What is the primary cause of the unsettling of a phonetic pattern andwhat is the cumulative force that selects these or those particularvariations of the individual on which to float the pattern readjustmentswe hardly know. Many linguistic students have made the fatal error ofthinking of sound change as a quasi-physiological instead of as astrictly psychological phenomenon, or they have tried to dispose of theproblem by bandying such catchwords as "the tendency to increased easeof articulation" or "the cumulative result of faulty perception" (on thepart of children, say, in learning to speak). These easy explanationswill not do. "Ease of articulation" may enter in as a factor, but it isa rather subjective concept at best. Indians find hopelessly difficultsounds and sound combinations that are simple to us; one languageencourages a phonetic drift that another does everything to fight. "Faulty perception" does not explain that impressive drift in speechsounds which I have insisted upon. It is much better to admit that we donot yet understand the primary cause or causes of the slow drift inphonetics, though we can frequently point to contributing factors. It islikely that we shall not advance seriously until we study theintuitional bases of speech. How can we understand the nature of thedrift that frays and reforms phonetic patterns when we have neverthought of studying sound patterning as such and the "weights" andpsychic relations of the single elements (the individual sounds) inthese patterns?

Every linguist knows that phonetic change is frequently followed bymorphological rearrangements, but he is apt to assume that morphologyexercises little or no influence on the course of phonetic history. I aminclined to believe that our present tendency to isolate phonetics andgrammar as mutually irrelevant linguistic provinces is unfortunate. There are likely to be fundamental relations between them and theirrespective histories that we do not yet fully grasp. After all, ifspeech sounds exist merely because they are the symbolic carriers ofsignificant concepts and groupings of concepts, why may not a strongdrift or a permanent feature in the conceptual sphere exercise afurthering or retarding influence on the phonetic drift? I believe thatsuch influences may be demonstrated and that they deserve far morecareful study than they have received. 

This brings us back to our unanswered question: How is it that bothEnglish and German developed the curious alternation of unmodified vowelin the singular (_foot_, _Fuss_) and modified vowel in the plural(_feet_, _Füsse_)? Was the pre-Anglo-Saxon alternation of _fot_ and_föti_ an absolutely mechanical matter, without other than incidentalmorphological interest? It is always so represented, and, indeed, allthe external facts support such a view. The change from _o_ to _ö_, later _e_, is by no means peculiar to the plural. It is found also inthe dative singular (_fet_), for it too goes back to an older _foti_. Moreover, _fet_ of the plural applies only to the nominative andaccusative; the genitive has _fota_, the dative _fotum_. Only centurieslater was the alternation of _o_ and _e_ reinterpreted as a means ofdistinguishing number; _o_ was generalized for the singular, _e_ for theplural. Only when this reassortment of forms took place[161] was themodern symbolic value of the _foot_: _feet_ alternation clearlyestablished. Again, we must not forget that _o_ was modified to _ö (e)_in all manner of other grammatical and derivative formations. Thus, apre-Anglo-Saxon _hohan_ (later _hon_) "to hang" corresponded to a_höhith_, _hehith_ (later _hehth_) "hangs"; to _dom_ "doom, " _blod_"blood, " and _fod_ "food" corresponded the verbal derivatives _dömian_(later _deman_) "to deem, " _blödian_ (later _bledan_) "to bleed, " and_födian_ (later _fedan_) "to feed. " All this seems to point to thepurely mechanical nature of the modification of _o_ to _ö_ to _e_. Somany unrelated functions were ultimately served by the vocalic changethat we cannot believe that it was motivated by any one of them. 

[Footnote 161: A type of adjustment generally referred to as "analogicalleveling. "]

The German facts are entirely analogous. Only later in the history ofthe language was the vocalic alternation made significant for number. And yet consider the following facts. The change of _foti_ to _föti_antedated that of _föti_ to _föte_, _föt_. This may be looked upon as a"lucky accident, " for if _foti_ had become _fote_, _fot_ before the _-i_had had the chance to exert a retroactive influence on the _o_, therewould have been no difference between the singular and the plural. Thiswould have been anomalous in Anglo-Saxon for a masculine noun. But wasthe sequence of phonetic changes an "accident"? Consider two furtherfacts. All the Germanic languages were familiar with vocalic change aspossessed of functional significance. Alternations like _sing_, _sang_, _sung_ (Anglo-Saxon _singan_, _sang_, _sungen_) were ingrained in thelinguistic consciousness. Further, the tendency toward the weakening offinal syllables was very strong even then and had been manifestingitself in one way and another for centuries. I believe that thesefurther facts help us to understand the actual sequence of phoneticchanges. We may go so far as to say that the _o_ (and _u_) could affordto stay the change to _ö_ (and _ü_) until the destructive drift hadadvanced to the point where failure to modify the vowel would soonresult in morphological embarrassment. At a certain moment the _-i_ending of the plural (and analogous endings with _i_ in otherformations) was felt to be too weak to quite bear its functional burden. The unconscious Anglo-Saxon mind, if I may be allowed a somewhat summaryway of putting the complex facts, was glad of the opportunity affordedby certain individual variations, until then automatically canceled out, to have some share of the burden thrown on them. These particularvariations won through because they so beautifully allowed the generalphonetic drift to take its course without unsettling the morphologicalcontours of the language. And the presence of symbolic variation(_sing_, _sang_, _sung_) acted as an attracting force on the rise of anew variation of similar character. All these factors were equally trueof the German vocalic shift. Owing to the fact that the destructivephonetic drift was proceeding at a slower rate in German than inEnglish, the preservative change of _uo_ to _üe_ (_u_ to _ü_) did notneed to set in until 300 years or more after the analogous Englishchange. Nor did it. And this is to my mind a highly significant fact. Phonetic changes may sometimes be unconsciously encouraged in order tokeep intact the psychological spaces between words and word forms. Thegeneral drift seizes upon those individual sound variations that help topreserve the morphological balance or to lead to the new balance thatthe language is striving for. 

I would suggest, then, that phonetic change is compacted of at leastthree basic strands: (1) A general drift in one direction, concerningthe nature of which we know almost nothing but which may be suspected tobe of prevailingly dynamic character (tendencies, e. G. , to greater orless stress, greater or less voicing of elements); (2) A readjustingtendency which aims to preserve or restore the fundamental phoneticpattern of the language; (3) A preservative tendency which sets in whena too serious morphological unsettlement is threatened by the maindrift. I do not imagine for a moment that it is always possible toseparate these strands or that this purely schematic statement doesjustice to the complex forces that guide the phonetic drift. Thephonetic pattern of a language is not invariable, but it changes farless readily than the sounds that compose it. Every phonetic elementthat it possesses may change radically and yet the pattern remainunaffected. It would be absurd to claim that our present English patternis identical with the old Indo-European one, yet it is impressive tonote that even at this late day the English series of initialconsonants:

_p_ _t_ _k__b_ _d_ _g__f_ _th_ _h_

corresponds point for point to the Sanskrit series:

_b_ _d_ _g__bh_ _dh_ _gh__p_ _t_ _k_

The relation between phonetic pattern and individual sound is roughlyparallel to that which obtains between the morphologic type of alanguage and one of its specific morphological features. Both phoneticpattern and fundamental type are exceedingly conservative, allsuperficial appearances to the contrary notwithstanding. Which is moreso we cannot say. I suspect that they hang together in a way that wecannot at present quite understand. 

If all the phonetic changes brought about by the phonetic drift wereallowed to stand, it is probable that most languages would present suchirregularities of morphological contour as to lose touch with theirformal ground-plan. Sound changes work mechanically. Hence they arelikely to affect a whole morphological group here--this does notmatter--, only part of a morphological group there--and this may bedisturbing. Thus, the old Anglo-Saxon paradigm:

 Sing. Plur. N. Ac. _fot_ _fet_ (older _foti_)G. _fotes_ _fota_D. _fet_ (older _foti_) _fotum_

could not long stand unmodified. The _o_--_e_ alternation was welcome inso far as it roughly distinguished the singular from the plural. Thedative singular _fet_, however, though justified historically, was soonfelt to be an intrusive feature. The analogy of simpler and morenumerously represented paradigms created the form _fote_ (compare, e. G. , _fisc_ "fish, " dative singular _fisce_). _Fet_ as a dative becomesobsolete. The singular now had _o_ throughout. But this very fact madethe genitive and dative _o_-forms of the plural seem out of place. Thenominative and accusative _fet_ was naturally far more frequently in usethan were the corresponding forms of the genitive and dative. These, inthe end, could not but follow the analogy of _fet_. At the verybeginning of the Middle English period, therefore, we find that the oldparadigm has yielded to a more regular one:

 Sing. Plur. N. Ac. *_fot_ *_fet_G. *_fotes_ _fete_D. _fote_ _feten_

The starred forms are the old nucleus around which the new paradigm isbuilt. The unstarred forms are not genealogical kin of their formalprototypes. They are analogical replacements. 

The history of the English language teems with such levelings orextensions. _Elder_ and _eldest_ were at one time the only possiblecomparative and superlative forms of _old_ (compare German _alt_, _älter_, _der älteste_; the vowel following the _old-_, _alt-_ wasoriginally an _i_, which modified the quality of the stem vowel). Thegeneral analogy of the vast majority of English adjectives, however, hascaused the replacement of the forms _elder_ and _eldest_ by the formswith unmodified vowel, _older_ and _oldest_. _Elder_ and _eldest_survive only as somewhat archaic terms for the older and oldest brotheror sister. This illustrates the tendency for words that arepsychologically disconnected from their etymological or formal group topreserve traces of phonetic laws that have otherwise left norecognizable trace or to preserve a vestige of a morphological processthat has long lost its vitality. A careful study of these survivals oratrophied forms is not without value for the reconstruction of theearlier history of a language or for suggestive hints as to its remoteraffiliations. 

Analogy may not only refashion forms within the confines of a relatedcluster of forms (a "paradigm") but may extend its influence far beyond. Of a number of functionally equivalent elements, for instance, only onemay survive, the rest yielding to its constantly widening influence. This is what happened with the English _-s_ plural. Originally confinedto a particular class of masculines, though an important class, the _-s_plural was gradually generalized for all nouns but a mere handful thatstill illustrate plural types now all but extinct (_foot_: feet, _goose_: _geese_, _tooth_: _teeth_, _mouse_: _mice_, _louse_: _lice_;_ox_: _oxen_; _child_: _children_; _sheep_: _sheep_, _deer_: _deer_). Thus analogy not only regularizes irregularities that have come in thewake of phonetic processes but introduces disturbances, generally infavor of greater simplicity or regularity, in a long established systemof forms. These analogical adjustments are practically always symptomsof the general morphological drift of the language. 

A morphological feature that appears as the incidental consequence of aphonetic process, like the English plural with modified vowel, mayspread by analogy no less readily than old features that owe theirorigin to other than phonetic causes. Once the _e_-vowel of MiddleEnglish _fet_ had become confined to the plural, there was notheoretical reason why alternations of the type _fot_: _fet_ and_mus_: _mis_ might not have become established as a productive type ofnumber distinction in the noun. As a matter of fact, it did not sobecome established. The _fot_: _fet_ type of plural secured but amomentary foothold. It was swept into being by one of the surface driftsof the language, to be swept aside in the Middle English period by themore powerful drift toward the use of simple distinctive forms. It wastoo late in the day for our language to be seriously interested in suchpretty symbolisms as _foot_: _feet_. What examples of the type aroselegitimately, in other words _via_ purely phonetic processes, weretolerated for a time, but the type as such never had a serious future. 

It was different in German. The whole series of phonetic changescomprised under the term "umlaut, " of which _u_: _ü_ and _au_: _oi_(written _äu_) are but specific examples, struck the German language ata time when the general drift to morphological simplification was not sostrong but that the resulting formal types (e. G. , _Fuss_: _Füsse_;_fallen_ "to fall": _fällen_ "to fell"; _Horn_ "horn": _Gehörne_ "groupof horns"; _Haus_ "house": _Häuslein_ "little house") could keepthemselves intact and even extend to forms that did not legitimatelycome within their sphere of influence. "Umlaut" is still a very livesymbolic process in German, possibly more alive to-day than in medievaltimes. Such analogical plurals as _Baum_ "tree": _Bäume_ (contrastMiddle High German _boum_: _boume_) and derivatives as _lachen_ "tolaugh": _Gelächter_ "laughter" (contrast Middle High German _gelach_)show that vocalic mutation has won through to the status of a productivemorphologic process. Some of the dialects have even gone further thanstandard German, at least in certain respects. In Yiddish, [162] forinstance, "umlaut" plurals have been formed where there are no MiddleHigh German prototypes or modern literary parallels, e. G. , _tog_ "day":_teg_ "days" (but German _Tag_: _Tage_) on the analogy of _gast_"guest": _gest_ "guests" (German _Gast_: _Gäste_), _shuch_[163] "shoe":_shich_ "shoes" (but German _Schuh_: _Schuhe_) on the analogy of _fus_"foot": _fis_ "feet. " It is possible that "umlaut" will run its courseand cease to operate as a live functional process in German, but thattime is still distant. Meanwhile all consciousness of the merelyphonetic nature of "umlaut" vanished centuries ago. It is now a strictlymorphological process, not in the least a mechanical phoneticadjustment. We have in it a splendid example of how a simple phoneticlaw, meaningless in itself, may eventually color or transform largereaches of the morphology of a language. 

[Footnote 162: Isolated from other German dialects in the late fifteenthand early sixteenth centuries. It is therefore a good test for gaugingthe strength of the tendency to "umlaut, " particularly as it hasdeveloped a strong drift towards analytic methods. ]

[Footnote 163: _Ch_ as in German _Buch_. ]

IX

HOW LANGUAGES INFLUENCE EACH OTHER

Languages, like cultures, are rarely sufficient unto themselves. Thenecessities of intercourse bring the speakers of one language intodirect or indirect contact with those of neighboring or culturallydominant languages. The intercourse may be friendly or hostile. It maymove on the humdrum plane of business and trade relations or it mayconsist of a borrowing or interchange of spiritual goods--art, science, religion. It would be difficult to point to a completely isolatedlanguage or dialect, least of all among the primitive peoples. The tribeis often so small that intermarriages with alien tribes that speak otherdialects or even totally unrelated languages are not uncommon. It mayeven be doubted whether intermarriage, intertribal trade, and generalcultural interchanges are not of greater relative significance onprimitive levels than on our own. Whatever the degree or nature ofcontact between neighboring peoples, it is generally sufficient to leadto some kind of linguistic interinfluencing. Frequently the influenceruns heavily in one direction. The language of a people that is lookedupon as a center of culture is naturally far more likely to exert anappreciable influence on other languages spoken in its vicinity than tobe influenced by them. Chinese has flooded the vocabularies of Corean, Japanese, and Annamite for centuries, but has received nothing inreturn. In the western Europe of medieval and modern times French hasexercised a similar, though probably a less overwhelming, influence. English borrowed an immense number of words from the French of theNorman invaders, later also from the court French of Isle de France, appropriated a certain number of affixed elements of derivational value(e. G. , _-ess_ of _princess_, _-ard_ of _drunkard_, _-ty_ of _royalty_), may have been somewhat stimulated in its general analytic drift bycontact with French, [164] and even allowed French to modify its phoneticpattern slightly (e. G. , initial _v_ and _j_ in words like _veal_ and_judge_; in words of Anglo-Saxon origin _v_ and _j_ can only occur aftervowels, e. G. , _over_, _hedge_). But English has exerted practically noinfluence on French. 

[Footnote 164: The earlier students of English, however, grosslyexaggerated the general "disintegrating" effect of French on middleEnglish. English was moving fast toward a more analytic structure longbefore the French influence set in. ]

The simplest kind of influence that one language may exert on another isthe "borrowing" of words. When there is cultural borrowing there isalways the likelihood that the associated words may be borrowed too. When the early Germanic peoples of northern Europe first learned ofwine-culture and of paved streets from their commercial or warlikecontact with the Romans, it was only natural that they should adopt theLatin words for the strange beverage (_vinum_, English _wine_, German_Wein_) and the unfamiliar type of road (_strata [via]_, English_street_, German _Strasse_). Later, when Christianity was introducedinto England, a number of associated words, such as _bishop_ and_angel_, found their way into English. And so the process has continueduninterruptedly down to the present day, each cultural wave bringing tothe language a new deposit of loan-words. The careful study of suchloan-words constitutes an interesting commentary on the history ofculture. One can almost estimate the rôle which various peoples haveplayed in the development and spread of cultural ideas by taking note ofthe extent to which their vocabularies have filtered into those of otherpeoples. When we realize that an educated Japanese can hardly frame asingle literary sentence without the use of Chinese resources, that tothis day Siamese and Burmese and Cambodgian bear the unmistakableimprint of the Sanskrit and Pali that came in with Hindu Buddhismcenturies ago, or that whether we argue for or against the teaching ofLatin and Greek our argument is sure to be studded with words that havecome to us from Rome and Athens, we get some inkling of what earlyChinese culture and Buddhism and classical Mediterranean civilizationhave meant in the world's history. There are just five languages thathave had an overwhelming significance as carriers of culture. They areclassical Chinese, Sanskrit, Arabic, Greek, and Latin. In comparisonwith these even such culturally important languages as Hebrew and Frenchsink into a secondary position. It is a little disappointing to learnthat the general cultural influence of English has so far been all butnegligible. The English language itself is spreading because the Englishhave colonized immense territories. But there is nothing to show that itis anywhere entering into the lexical heart of other languages as Frenchhas colored the English complexion or as Arabic has permeated Persianand Turkish. This fact alone is significant of the power of nationalism, cultural as well as political, during the last century. There are nowpsychological resistances to borrowing, or rather to new sources ofborrowing, [165] that were not greatly alive in the Middle Ages or duringthe Renaissance. 

[Footnote 165: For we still name our new scientific instruments andpatent medicines from Greek and Latin. ]

Are there resistances of a more intimate nature to the borrowing ofwords? It is generally assumed that the nature and extent of borrowingdepend entirely on the historical facts of culture relation; that ifGerman, for instance, has borrowed less copiously than English fromLatin and French it is only because Germany has had less intimaterelations than England with the culture spheres of classical Rome andFrance. This is true to a considerable extent, but it is not the wholetruth. We must not exaggerate the physical importance of the Normaninvasion nor underrate the significance of the fact that Germany'scentral geographical position made it peculiarly sensitive to Frenchinfluences all through the Middle Ages, to humanistic influences in thelatter fifteenth and early sixteenth centuries, and again to thepowerful French influences of the seventeenth and eighteenth centuries. It seems very probable that the psychological attitude of the borrowinglanguage itself towards linguistic material has much to do with itsreceptivity to foreign words. English has long been striving for thecompletely unified, unanalyzed word, regardless of whether it ismonosyllabic or polysyllabic. Such words as _credible_, _certitude_, _intangible_ are entirely welcome in English because each represents aunitary, well-nuanced idea and because their formal analysis(_cred-ible_, _cert-itude_, _in-tang-ible_) is not a necessary act ofthe unconscious mind (_cred-_, _cert-_, and _tang-_ have no realexistence in English comparable to that of _good-_ in _goodness_). Aword like _intangible_, once it is acclimated, is nearly as simple apsychological entity as any radical monosyllable (say _vague_, _thin_, _grasp_). In German, however, polysyllabic words strive to analyzethemselves into significant elements. Hence vast numbers of French andLatin words, borrowed at the height of certain cultural influences, could not maintain themselves in the language. Latin-German words like_kredibel_ "credible" and French-German words like _reussieren_ "tosucceed" offered nothing that the unconscious mind could assimilate toits customary method of feeling and handling words. It is as though thisunconscious mind said: "I am perfectly willing to accept _kredibel_ ifyou will just tell me what you mean by _kred-_. " Hence German hasgenerally found it easier to create new words out of its own resources, as the necessity for them arose. 

The psychological contrast between English and German as regards thetreatment of foreign material is a contrast that may be studied in allparts of the world. The Athabaskan languages of America are spoken bypeoples that have had astonishingly varied cultural contacts, yetnowhere do we find that an Athabaskan dialect has borrowed at allfreely[166] from a neighboring language. These languages have alwaysfound it easier to create new words by compounding afresh elements readyto hand. They have for this reason been highly resistant to receivingthe linguistic impress of the external cultural experiences of theirspeakers. Cambodgian and Tibetan offer a highly instructive contrast intheir reaction to Sanskrit influence. Both are analytic languages, eachtotally different from the highly-wrought, inflective language of India. Cambodgian is isolating, but, unlike Chinese, it contains manypolysyllabic words whose etymological analysis does not matter. LikeEnglish, therefore, in its relation to French and Latin, it welcomedimmense numbers of Sanskrit loan-words, many of which are in common useto-day. There was no psychological resistance to them. Classical Tibetanliterature was a slavish adaptation of Hindu Buddhist literature andnowhere has Buddhism implanted itself more firmly than in Tibet, yet itis strange how few Sanskrit words have found their way into thelanguage. Tibetan was highly resistant to the polysyllabic words ofSanskrit because they could not automatically fall into significantsyllables, as they should have in order to satisfy the Tibetan feelingfor form. Tibetan was therefore driven to translating the great majorityof these Sanskrit words into native equivalents. The Tibetan craving forform was satisfied, though the literally translated foreign terms mustoften have done violence to genuine Tibetan idiom. Even the proper namesof the Sanskrit originals were carefully translated, element forelement, into Tibetan; e. G. , _Suryagarbha_ "Sun-bosomed" was carefullyTibetanized into _Nyi-mai snying-po_ "Sun-of heart-the, the heart (oressence) of the sun. " The study of how a language reacts to the presenceof foreign words--rejecting them, translating them, or freely acceptingthem--may throw much valuable light on its innate formal tendencies. 

[Footnote 166: One might all but say, "has borrowed at all. "]

The borrowing of foreign words always entails their phoneticmodification. There are sure to be foreign sounds or accentualpeculiarities that do not fit the native phonetic habits. They are thenso changed as to do as little violence as possible to these habits. Frequently we have phonetic compromises. Such an English word as therecently introduced _camouflage_, as now ordinarily pronounced, corresponds to the typical phonetic usage of neither English nor French. The aspirated _k_, the obscure vowel of the second syllable, the precisequality of the _l_ and of the last _a_, and, above all, the strongaccent on the first syllable, are all the results of unconsciousassimilation to our English habits of pronunciation. They differentiateour _camouflage_ clearly from the same word as pronounced by theFrench. On the other hand, the long, heavy vowel in the third syllableand the final position of the "zh" sound (like _z_ in _azure_) aredistinctly un-English, just as, in Middle English, the initial _j_ and_v_[167] must have been felt at first as not strictly in accord withEnglish usage, though the strangeness has worn off by now. In all fourof these cases--initial _j_, initial _v_, final "zh, " and unaccented _a_of _father_--English has not taken on a new sound but has merelyextended the use of an old one. 

[Footnote 167: See page 206. ]

[Transcriber's note: Footnote 167 refers to the paragraph beginning online 6329. ]

Occasionally a new sound is introduced, but it is likely to melt awaybefore long. In Chaucer's day the old Anglo-Saxon _ü_ (written _y_) hadlong become unrounded to _i_, but a new set of _ü_-vowels had come infrom the French (in such words as _due_, _value_, _nature_). The new _ü_did not long hold its own; it became diphthongized to _iu_ and wasamalgamated with the native _iw_ of words like _new_ and _slew_. Eventually this diphthong appears as _yu_, with change of stress--_dew_(from Anglo-Saxon _deaw_) like _due_ (Chaucerian _dü_). Facts like theseshow how stubbornly a language resists radical tampering with itsphonetic pattern. 

Nevertheless, we know that languages do influence each other in phoneticrespects, and that quite aside from the taking over of foreign soundswith borrowed words. One of the most curious facts that linguistics hasto note is the occurrence of striking phonetic parallels in totallyunrelated or very remotely related languages of a restrictedgeographical area. These parallels become especially impressive whenthey are seen contrastively from a wide phonetic perspective. Here are afew examples. The Germanic languages as a whole have not developednasalized vowels. Certain Upper German (Suabian) dialects, however, have now nasalized vowels in lieu of the older vowel + nasal consonant(_n_). Is it only accidental that these dialects are spoken in proximityto French, which makes abundant use of nasalized vowels? Again, thereare certain general phonetic features that mark off Dutch and Flemish incontrast, say, to North German and Scandinavian dialects. One of theseis the presence of unaspirated voiceless stops (_p_, _t_, _k_), whichhave a precise, metallic quality reminiscent of the corresponding Frenchsounds, but which contrast with the stronger, aspirated stops ofEnglish, North German, and Danish. Even if we assume that theunaspirated stops are more archaic, that they are the unmodifieddescendants of the old Germanic consonants, is it not perhaps asignificant historical fact that the Dutch dialects, neighbors ofFrench, were inhibited from modifying these consonants in accordancewith what seems to have been a general Germanic phonetic drift? Evenmore striking than these instances is the peculiar resemblance, incertain special phonetic respects, of Russian and other Slavic languagesto the unrelated Ural-Altaic languages[168] of the Volga region. Thepeculiar, dull vowel, for instance, known in Russian as "yeri"[169] hasUral-Altaic analogues, but is entirely wanting in Germanic, Greek, Armenian, and Indo-Iranian, the nearest Indo-European congeners ofSlavic. We may at least suspect that the Slavic vowel is nothistorically unconnected with its Ural-Altaic parallels. One of the mostpuzzling cases of phonetic parallelism is afforded by a large number ofAmerican Indian languages spoken west of the Rockies. Even at the mostradical estimate there are at least four totally unrelated linguisticstocks represented in the region from southern Alaska to centralCalifornia. Nevertheless all, or practically all, the languages of thisimmense area have some important phonetic features in common. Chief ofthese is the presence of a "glottalized" series of stopped consonants ofvery distinctive formation and of quite unusual acoustic effect. [170] Inthe northern part of the area all the languages, whether related or not, also possess various voiceless _l_-sounds and a series of "velar"(back-guttural) stopped consonants which are etymologically distinctfrom the ordinary _k_-series. It is difficult to believe that three suchpeculiar phonetic features as I have mentioned could have evolvedindependently in neighboring groups of languages. 

[Footnote 168: Ugro-Finnic and Turkish (Tartar)]

[Footnote 169: Probably, in Sweet's terminology, high-back (or, better, between back and "mixed" positions)-narrow-unrounded. It generallycorresponds to an Indo-European long _u_. ]

[Footnote 170: There seem to be analogous or partly analogous sounds incertain languages of the Caucasus. ]

How are we to explain these and hundreds of similar phoneticconvergences? In particular cases we may really be dealing with archaicsimilarities due to a genetic relationship that it is beyond our presentpower to demonstrate. But this interpretation will not get us far. Itmust be ruled entirely out of court, for instance, in two of the threeEuropean examples I have instanced; both nasalized vowels and the Slavic"yeri" are demonstrably of secondary origin in Indo-European. However weenvisage the process in detail, we cannot avoid the inference that thereis a tendency for speech sounds or certain distinctive manners ofarticulation to spread over a continuous area in somewhat the same waythat elements of culture ray out from a geographical center. We maysuppose that individual variations arising at linguisticborderlands--whether by the unconscious suggestive influence of foreignspeech habits or by the actual transfer of foreign sounds into thespeech of bilingual individuals--have gradually been incorporated intothe phonetic drift of a language. So long as its main phonetic concernis the preservation of its sound patterning, not of its sounds as such, there is really no reason why a language may not unconsciouslyassimilate foreign sounds that have succeeded in worming their way intoits gamut of individual variations, provided always that these newvariations (or reinforced old variations) are in the direction of thenative drift. 

A simple illustration will throw light on this conception. Let ussuppose that two neighboring and unrelated languages, A and B, eachpossess voiceless _l_-sounds (compare Welsh _ll_). We surmise that thisis not an accident. Perhaps comparative study reveals the fact that inlanguage A the voiceless _l_-sounds correspond to a sibilant series inother related languages, that an old alternation _s_: _sh_ has beenshifted to the new alternation _l_ (voiceless): _s_. [171] Does it followthat the voiceless _l_ of language B has had the same history? Not inthe least. Perhaps B has a strong tendency toward audible breath releaseat the end of a word, so that the final _l_, like a final vowel, wasoriginally followed by a marked aspiration. Individuals perhaps tendedto anticipate a little the voiceless release and to "unvoice" the latterpart of the final _l_-sound (very much as the _l_ of English words like_felt_ tends to be partly voiceless in anticipation of the voicelessnessof the _t_). Yet this final _l_ with its latent tendency to unvoicingmight never have actually developed into a fully voiceless _l_ had notthe presence of voiceless _l_-sounds in A acted as an unconsciousstimulus or suggestive push toward a more radical change in the line ofB's own drift. Once the final voiceless _l_ emerged, its alternation inrelated words with medial voiced _l_ is very likely to have led to itsanalogical spread. The result would be that both A and B have animportant phonetic trait in common. Eventually their phonetic systems, judged as mere assemblages of sounds, might even become completelyassimilated to each other, though this is an extreme case hardly everrealized in practice. The highly significant thing about such phoneticinterinfluencings is the strong tendency of each language to keep itsphonetic pattern intact. So long as the respective alignments of thesimilar sounds is different, so long as they have differing "values" and"weights" in the unrelated languages, these languages cannot be said tohave diverged materially from the line of their inherent drift. Inphonetics, as in vocabulary, we must be careful not to exaggerate theimportance of interlinguistic influences. 

[Footnote 171: This can actually be demonstrated for one of theAthabaskan dialects of the Yukon. ]

I have already pointed out in passing that English has taken over acertain number of morphological elements from French. English also usesa number of affixes that are derived from Latin and Greek. Some of theseforeign elements, like the _-ize_ of _materialize_ or the _-able_ of_breakable_, are even productive to-day. Such examples as these arehardly true evidences of a morphological influence exerted by onelanguage on another. Setting aside the fact that they belong to thesphere of derivational concepts and do not touch the centralmorphological problem of the expression of relational ideas, they haveadded nothing to the structural peculiarities of our language. Englishwas already prepared for the relation of _pity_ to _piteous_ by such anative pair as _luck_ and _lucky_; _material_ and _materialize_ merelyswelled the ranks of a form pattern familiar from such instances as_wide_ and _widen_. In other words, the morphological influence exertedby foreign languages on English, if it is to be gauged by such examplesas I have cited, is hardly different in kind from the mere borrowing ofwords. The introduction of the suffix _-ize_ made hardly more differenceto the essential build of the language than did the mere fact that itincorporated a given number of words. Had English evolved a new futureon the model of the synthetic future in French or had it borrowed fromLatin and Greek their employment of reduplication as a functional device(Latin _tango_: _tetigi_; Greek _leipo_: _leloipa_), we should have theright to speak of true morphological influence. But such far-reachinginfluences are not demonstrable. Within the whole course of the historyof the English language we can hardly point to one importantmorphological change that was not determined by the native drift, thoughhere and there we may surmise that this drift was hastened a little bythe suggestive influence of French forms. [172]

[Footnote 172: In the sphere of syntax one may point to certain Frenchand Latin influences, but it is doubtful if they ever reached deeperthan the written language. Much of this type of influence belongs ratherto literary style than to morphology proper. ]

It is important to realize the continuous, self-contained morphologicaldevelopment of English and the very modest extent to which itsfundamental build has been affected by influences from without. Thehistory of the English language has sometimes been represented as thoughit relapsed into a kind of chaos on the arrival of the Normans, whoproceeded to play nine-pins with the Anglo-Saxon tradition. Students aremore conservative today. That a far-reaching analytic development maytake place without such external foreign influence as English wassubjected to is clear from the history of Danish, which has gone evenfurther than English in certain leveling tendencies. English may beconveniently used as an _a fortiori_ test. It was flooded with Frenchloan-words during the later Middle Ages, at a time when its drift towardthe analytic type was especially strong. It was therefore changingrapidly both within and on the surface. The wonder, then, is not that ittook on a number of external morphological features, mere accretions onits concrete inventory, but that, exposed as it was to remoldinginfluences, it remained so true to its own type and historic drift. Theexperience gained from the study of the English language is strengthenedby all that we know of documented linguistic history. Nowhere do we findany but superficial morphological interinfluencings. We may infer one ofseveral things from this:--That a really serious morphological influenceis not, perhaps, impossible, but that its operation is so slow that ithas hardly ever had the chance to incorporate itself in the relativelysmall portion of linguistic history that lies open to inspection; orthat there are certain favorable conditions that make for profoundmorphological disturbances from without, say a peculiar instability oflinguistic type or an unusual degree of cultural contact, conditionsthat do not happen to be realized in our documentary material; or, finally, that we have not the right to assume that a language may easilyexert a remolding morphological influence on another. 

Meanwhile we are confronted by the baffling fact that important traitsof morphology are frequently found distributed among widely differinglanguages within a large area, so widely differing, indeed, that it iscustomary to consider them genetically unrelated. Sometimes we maysuspect that the resemblance is due to a mere convergence, that asimilar morphological feature has grown up independently in unrelatedlanguages. Yet certain morphological distributions are too specific incharacter to be so lightly dismissed. There must be some historicalfactor to account for them. Now it should be remembered that the conceptof a "linguistic stock" is never definitive[173] in an exclusive sense. We can only say, with reasonable certainty, that such and such languagesare descended from a common source, but we cannot say that such and suchother languages are not genetically related. All we can do is to saythat the evidence for relationship is not cumulative enough to make theinference of common origin absolutely necessary. May it not be, then, that many instances of morphological similarity between divergentlanguages of a restricted area are merely the last vestiges of acommunity of type and phonetic substance that the destructive work ofdiverging drifts has now made unrecognizable? There is probably stillenough lexical and morphological resemblance between modern English andIrish to enable us to make out a fairly conclusive case for theirgenetic relationship on the basis of the present-day descriptiveevidence alone. It is true that the case would seem weak in comparisonto the case that we can actually make with the help of the historicaland the comparative data that we possess. It would not be a bad casenevertheless. In another two or three millennia, however, the points ofresemblance are likely to have become so obliterated that English andIrish, in the absence of all but their own descriptive evidence, willhave to be set down as "unrelated" languages. They will still have incommon certain fundamental morphological features, but it will bedifficult to know how to evaluate them. Only in the light of thecontrastive perspective afforded by still more divergent languages, suchas Basque and Finnish, will these vestigial resemblances receive theirtrue historic value. 

[Footnote 173: See page 163. ]

[Transcriber's note: Footnote 173 refers to the paragraph beginning online 5037. ]

I cannot but suspect that many of the more significant distributions ofmorphological similarities are to be explained as just such vestiges. The theory of "borrowing" seems totally inadequate to explain thosefundamental features of structure, hidden away in the very core of thelinguistic complex, that have been pointed out as common, say, toSemitic and Hamitic, to the various Soudanese languages, toMalayo-Polynesian and Mon-Khmer[174] and Munda, [175] to Athabaskan andTlingit and Haida. We must not allow ourselves to be frightened away bythe timidity of the specialists, who are often notably lacking in thesense of what I have called "contrastive perspective. "

[Footnote 174: A group of languages spoken in southeastern Asia, ofwhich Khmer (Cambodgian) is the best known representative. ]

[Footnote 175: A group of languages spoken in northeastern India. ]

Attempts have sometimes been made to explain the distribution of thesefundamental structural features by the theory of diffusion. We know thatmyths, religious ideas, types of social organization, industrialdevices, and other features of culture may spread from point to point, gradually making themselves at home in cultures to which they were atone time alien. We also know that words may be diffused no less freelythan cultural elements, that sounds also may be "borrowed, " and thateven morphological elements may be taken over. We may go further andrecognize that certain languages have, in all probability, taken onstructural features owing to the suggestive influence of neighboringlanguages. An examination of such cases, [176] however, almost invariablyreveals the significant fact that they are but superficial additions onthe morphological kernel of the language. So long as such directhistorical testimony as we have gives us no really convincing examplesof profound morphological influence by diffusion, we shall do well notto put too much reliance in diffusion theories. On the whole, therefore, we shall ascribe the major concordances and divergences in linguisticform--phonetic pattern and morphology--to the autonomous drift oflanguage, not to the complicating effect of single, diffused featuresthat cluster now this way, now that. Language is probably the mostself-contained, the most massively resistant of all social phenomena. Itis easier to kill it off than to disintegrate its individual form. 

[Footnote 176: I have in mind, e. G. , the presence of postpositions inUpper Chinook, a feature that is clearly due to the influence ofneighboring Sahaptin languages; or the use by Takelma of instrumentalprefixes, which are likely to have been suggested by neighboring "Hokan"languages (Shasta, Karok). ]

X

LANGUAGE, RACE AND CULTURE

Language has a setting. The people that speak it belong to a race (or anumber of races), that is, to a group which is set off by physicalcharacteristics from other groups. Again, language does not exist apartfrom culture, that is, from the socially inherited assemblage ofpractices and beliefs that determines the texture of our lives. Anthropologists have been in the habit of studying man under the threerubrics of race, language, and culture. One of the first things they dowith a natural area like Africa or the South Seas is to map it out fromthis threefold point of view. These maps answer the questions: What andwhere are the major divisions of the human animal, biologicallyconsidered (e. G. , Congo Negro, Egyptian White; Australian Black, Polynesian)? What are the most inclusive linguistic groupings, the"linguistic stocks, " and what is the distribution of each (e. G. , theHamitic languages of northern Africa, the Bantu languages of the south;the Malayo-Polynesian languages of Indonesia, Melanesia, Micronesia, andPolynesia)? How do the peoples of the given area divide themselves ascultural beings? what are the outstanding "cultural areas" and what arethe dominant ideas in each (e. G. , the Mohammedan north of Africa; theprimitive hunting, non-agricultural culture of the Bushmen in the south;the culture of the Australian natives, poor in physical respects butrichly developed in ceremonialism; the more advanced and highlyspecialized culture of Polynesia)?

The man in the street does not stop to analyze his position in thegeneral scheme of humanity. He feels that he is the representative ofsome strongly integrated portion of humanity--now thought of as a"nationality, " now as a "race"--and that everything that pertains to himas a typical representative of this large group somehow belongstogether. If he is an Englishman, he feels himself to be a member of the"Anglo-Saxon" race, the "genius" of which race has fashioned the Englishlanguage and the "Anglo-Saxon" culture of which the language is theexpression. Science is colder. It inquires if these three types ofclassification--racial, linguistic, and cultural--are congruent, iftheir association is an inherently necessary one or is merely a matterof external history. The answer to the inquiry is not encouraging to"race" sentimentalists. Historians and anthropologists find that races, languages, and cultures are not distributed in parallel fashion, thattheir areas of distribution intercross in the most bewildering fashion, and that the history of each is apt to follow a distinctive course. Races intermingle in a way that languages do not. On the other hand, languages may spread far beyond their original home, invading theterritory of new races and of new culture spheres. A language may evendie out in its primary area and live on among peoples violently hostileto the persons of its original speakers. Further, the accidents ofhistory are constantly rearranging the borders of culture areas withoutnecessarily effacing the existing linguistic cleavages. If we can oncethoroughly convince ourselves that race, in its only intelligible, thatis biological, sense, is supremely indifferent to the history oflanguages and cultures, that these are no more directly explainable onthe score of race than on that of the laws of physics and chemistry, weshall have gained a viewpoint that allows a certain interest to suchmystic slogans as Slavophilism, Anglo-Saxondom, Teutonism, and the Latingenius but that quite refuses to be taken in by any of them. A carefulstudy of linguistic distributions and of the history of suchdistributions is one of the driest of commentaries on these sentimentalcreeds. 

That a group of languages need not in the least correspond to a racialgroup or a culture area is easily demonstrated. We may even show how asingle language intercrosses with race and culture lines. The Englishlanguage is not spoken by a unified race. In the United States there areseveral millions of negroes who know no other language. It is theirmother-tongue, the formal vesture of their inmost thoughts andsentiments. It is as much their property, as inalienably "theirs, " asthe King of England's. Nor do the English-speaking whites of Americaconstitute a definite race except by way of contrast to the negroes. Ofthe three fundamental white races in Europe generally recognized byphysical anthropologists--the Baltic or North European, the Alpine, andthe Mediterranean--each has numerous English-speaking representatives inAmerica. But does not the historical core of English-speaking peoples, those relatively "unmixed" populations that still reside in England andits colonies, represent a race, pure and single? I cannot see that theevidence points that way. The English people are an amalgam of manydistinct strains. Besides the old "Anglo-Saxon, " in other words NorthGerman, element which is conventionally represented as the basicstrain, the English blood comprises Norman French, [177] Scandinavian, "Celtic, "[178] and pre-Celtic elements. If by "English" we mean alsoScotch and Irish, [179] then the term "Celtic" is loosely used for atleast two quite distinct racial elements--the short, dark-complexionedtype of Wales and the taller, lighter, often ruddy-haired type of theHighlands and parts of Ireland. Even if we confine ourselves to theSaxon element, which, needless to say, nowhere appears "pure, " we arenot at the end of our troubles. We may roughly identify this strain withthe racial type now predominant in southern Denmark and adjoining partsof northern Germany. If so, we must content ourselves with thereflection that while the English language is historically most closelyaffiliated with Frisian, in second degree with the other West Germanicdialects (Low Saxon or "Plattdeutsch, " Dutch, High German), only inthird degree with Scandinavian, the specific "Saxon" racial type thatoverran England in the fifth and sixth centuries was largely the same asthat now represented by the Danes, who speak a Scandinavian language, while the High German-speaking population of central and southernGermany[180] is markedly distinct. 

[Footnote 177: Itself an amalgam of North "French" and Scandinavianelements. ]

[Footnote 178: The "Celtic" blood of what is now England and Wales is byno means confined to the Celtic-speaking regions--Wales and, untilrecently, Cornwall. There is every reason to believe that the invadingGermanic tribes (Angles, Saxons, Jutes) did not exterminate theBrythonic Celts of England nor yet drive them altogether into Wales andCornwall (there has been far too much "driving" of conquered peoplesinto mountain fastnesses and land's ends in our histories), but simplyintermingled with them and imposed their rule and language upon them. ]

[Footnote 179: In practice these three peoples can hardly be keptaltogether distinct. The terms have rather a local-sentimental than aclearly racial value. Intermarriage has gone on steadily for centuriesand it is only in certain outlying regions that we get relatively puretypes, e. G. , the Highland Scotch of the Hebrides. In America, English, Scotch, and Irish strands have become inextricably interwoven. ]

[Footnote 180: The High German now spoken in northern Germany is not ofgreat age, but is due to the spread of standardized German, based onUpper Saxon, a High German dialect, at the expense of "Plattdeutsch. "]

But what if we ignore these finer distinctions and simply assume thatthe "Teutonic" or Baltic or North European racial type coincided in itsdistribution with that of the Germanic languages? Are we not on safeground then? No, we are now in hotter water than ever. First of all, themass of the German-speaking population (central and southern Germany, German Switzerland, German Austria) do not belong to the tall, blond-haired, long-headed[181] "Teutonic" race at all, but to theshorter, darker-complexioned, short-headed[182] Alpine race, of whichthe central population of France, the French Swiss, and many of thewestern and northern Slavs (e. G. , Bohemians and Poles) are equally goodrepresentatives. The distribution of these "Alpine" populationscorresponds in part to that of the old continental "Celts, " whoselanguage has everywhere given way to Italic, Germanic, and Slavicpressure. We shall do well to avoid speaking of a "Celtic race, " but ifwe were driven to give the term a content, it would probably be moreappropriate to apply it to, roughly, the western portion of the Alpinepeoples than to the two island types that I referred to before. Theselatter were certainly "Celticized, " in speech and, partly, in blood, precisely as, centuries later, most of England and part of Scotland was"Teutonized" by the Angles and Saxons. Linguistically speaking, the"Celts" of to-day (Irish Gaelic, Manx, Scotch Gaelic, Welsh, Breton) areCeltic and most of the Germans of to-day are Germanic precisely as theAmerican Negro, Americanized Jew, Minnesota Swede, and German-Americanare "English. " But, secondly, the Baltic race was, and is, by no meansan exclusively Germanic-speaking people. The northernmost "Celts, " suchas the Highland Scotch, are in all probability a specialized offshoot ofthis race. What these people spoke before they were Celticized nobodyknows, but there is nothing whatever to indicate that they spoke aGermanic language. Their language may quite well have been as remotefrom any known Indo-European idiom as are Basque and Turkish to-day. Again, to the east of the Scandinavians are non-Germanic members of therace--the Finns and related peoples, speaking languages that are notdefinitely known to be related to Indo-European at all. 

[Footnote 181: "Dolichocephalic. "]

[Footnote 182: "Brachycephalic. "]

We cannot stop here. The geographical position of the Germanic languagesis such[183] as to make it highly probable that they represent but anoutlying transfer of an Indo-European dialect (possibly a Celto-Italicprototype) to a Baltic people speaking a language or a group oflanguages that was alien to Indo-European. [184] Not only, then, isEnglish not spoken by a unified race at present but its prototype, morelikely than not, was originally a foreign language to the race withwhich English is more particularly associated. We need not seriouslyentertain the idea that English or the group of languages to which itbelongs is in any intelligible sense the expression of race, that thereare embedded in it qualities that reflect the temperament or "genius" ofa particular breed of human beings. 

[Footnote 183: By working back from such data as we possess we can makeit probable that these languages were originally confined to acomparatively small area in northern Germany and Scandinavia. This areais clearly marginal to the total area of distribution of theIndo-European-speaking peoples. Their center of gravity, say 1000 B. C. , seems to have lain in southern Russia. ]

[Footnote 184: While this is only a theory, the technical evidence forit is stronger than one might suppose. There are a surprising number ofcommon and characteristic Germanic words which cannot be connected withknown Indo-European radical elements and which may well be survivals ofthe hypothetical pre-Germanic language; such are _house_, _stone_, _sea_, _wife_ (German _Haus_, _Stein_, _See_, _Weib_). ]

Many other, and more striking, examples of the lack of correspondencebetween race and language could be given if space permitted. Oneinstance will do for many. The Malayo-Polynesian languages form awell-defined group that takes in the southern end of the Malay Peninsulaand the tremendous island world to the south and east (except Australiaand the greater part of New Guinea). In this vast region we findrepresented no less than three distinct races--the Negro-like Papuans ofNew Guinea and Melanesia, the Malay race of Indonesia, and thePolynesians of the outer islands. The Polynesians and Malays all speaklanguages of the Malayo-Polynesian group, while the languages of thePapuans belong partly to this group (Melanesian), partly to theunrelated languages ("Papuan") of New Guinea. [185] In spite of the factthat the greatest race cleavage in this region lies between the Papuansand the Polynesians, the major linguistic division is of Malayan on theone side, Melanesian and Polynesian on the other. 

[Footnote 185: Only the easternmost part of this island is occupied byMelanesian-speaking Papuans. ]

As with race, so with culture. Particularly in more primitive levels, where the secondarily unifying power of the "national"[186] ideal doesnot arise to disturb the flow of what we might call naturaldistributions, is it easy to show that language and culture are notintrinsically associated. Totally unrelated languages share in oneculture, closely related languages--even a single language--belong todistinct culture spheres. There are many excellent examples inaboriginal America. The Athabaskan languages form as clearly unified, asstructurally specialized, a group as any that I know of. [187] Thespeakers of these languages belong to four distinct culture areas--thesimple hunting culture of western Canada and the interior of Alaska(Loucheux, Chipewyan), the buffalo culture of the Plains (Sarcee), thehighly ritualized culture of the southwest (Navaho), and the peculiarlyspecialized culture of northwestern California (Hupa). The culturaladaptability of the Athabaskan-speaking peoples is in the strangestcontrast to the inaccessibility to foreign influences of the languagesthemselves. [188] The Hupa Indians are very typical of the culture areato which they belong. Culturally identical with them are the neighboringYurok and Karok. There is the liveliest intertribal intercourse betweenthe Hupa, Yurok, and Karok, so much so that all three generally attendan important religious ceremony given by any one of them. It isdifficult to say what elements in their combined culture belong inorigin to this tribe or that, so much at one are they in communalaction, feeling, and thought. But their languages are not merely aliento each other; they belong to three of the major American linguisticgroups, each with an immense distribution on the northern continent. Hupa, as we have seen, is Athabaskan and, as such, is also distantlyrelated to Haida (Queen Charlotte Islands) and Tlingit (southernAlaska); Yurok is one of the two isolated Californian languages of theAlgonkin stock, the center of gravity of which lies in the region of theGreat Lakes; Karok is the northernmost member of the Hokan group, whichstretches far to the south beyond the confines of California and hasremoter relatives along the Gulf of Mexico. 

[Footnote 186: A "nationality" is a major, sentimentally unified, group. The historical factors that lead to the feeling of national unity arevarious--political, cultural, linguistic, geographic, sometimesspecifically religious. True racial factors also may enter in, thoughthe accent on "race" has generally a psychological rather than astrictly biological value. In an area dominated by the nationalsentiment there is a tendency for language and culture to become uniformand specific, so that linguistic and cultural boundaries at least tendto coincide. Even at best, however, the linguistic unification is neverabsolute, while the cultural unity is apt to be superficial, of aquasi-political nature, rather than deep and far-reaching. ]

[Footnote 187: The Semitic languages, idiosyncratic as they are, are nomore definitely ear-marked. ]

[Footnote 188: See page 209. ]

[Transcriber's note: Footnote 188 refers to the paragraph beginning online 6448. ]

Returning to English, most of us would readily admit, I believe, thatthe community of language between Great Britain and the United States isfar from arguing a like community of culture. It is customary to saythat they possess a common "Anglo-Saxon" cultural heritage, but are notmany significant differences in life and feeling obscured by thetendency of the "cultured" to take this common heritage too much forgranted? In so far as America is still specifically "English, " it isonly colonially or vestigially so; its prevailing cultural drift ispartly towards autonomous and distinctive developments, partly towardsimmersion in the larger European culture of which that of England isonly a particular facet. We cannot deny that the possession of a commonlanguage is still and will long continue to be a smoother of the way toa mutual cultural understanding between England and America, but it isvery clear that other factors, some of them rapidly cumulative, areworking powerfully to counteract this leveling influence. A commonlanguage cannot indefinitely set the seal on a common culture when thegeographical, political, and economic determinants of the culture are nolonger the same throughout its area. 

Language, race, and culture are not necessarily correlated. This doesnot mean that they never are. There is some tendency, as a matter offact, for racial and cultural lines of cleavage to correspond tolinguistic ones, though in any given case the latter may not be of thesame degree of importance as the others. Thus, there is a fairlydefinite line of cleavage between the Polynesian languages, race, andculture on the one hand and those of the Melanesians on the other, inspite of a considerable amount of overlapping. [189] The racial andcultural division, however, particularly the former, are of majorimportance, while the linguistic division is of quite minorsignificance, the Polynesian languages constituting hardly more than aspecial dialectic subdivision of the combined Melanesian-Polynesiangroup. Still clearer-cut coincidences of cleavage may be found. Thelanguage, race, and culture of the Eskimo are markedly distinct fromthose of their neighbors;[190] in southern Africa the language, race, and culture of the Bushmen offer an even stronger contrast to those oftheir Bantu neighbors. Coincidences of this sort are of the greatestsignificance, of course, but this significance is not one of inherentpsychological relation between the three factors of race, language, andculture. The coincidences of cleavage point merely to a readilyintelligible historical association. If the Bantu and Bushmen are sosharply differentiated in all respects, the reason is simply that theformer are relatively recent arrivals in southern Africa. The twopeoples developed in complete isolation from each other; their presentpropinquity is too recent for the slow process of cultural and racialassimilation to have set in very powerfully. As we go back in time, weshall have to assume that relatively scanty populations occupied largeterritories for untold generations and that contact with other masses ofpopulation was not as insistent and prolonged as it later became. Thegeographical and historical isolation that brought about racedifferentiations was naturally favorable also to far-reaching variationsin language and culture. The very fact that races and cultures which arebrought into historical contact tend to assimilate in the long run, while neighboring languages assimilate each other only casually and insuperficial respects[191], indicates that there is no profound causalrelation between the development of language and the specificdevelopment of race and of culture. 

[Footnote 189: The Fijians, for instance, while of Papuan (negroid)race, are Polynesian rather than Melanesian in their cultural andlinguistic affinities. ]

[Footnote 190: Though even here there is some significant overlapping. The southernmost Eskimo of Alaska were assimilated in culture to theirTlingit neighbors. In northeastern Siberia, too, there is no sharpcultural line between the Eskimo and the Chukchi. ]

[Footnote 191: The supersession of one language by another is of coursenot truly a matter of linguistic assimilation. ]

But surely, the wary reader will object, there must be some relationbetween language and culture, and between language and at least thatintangible aspect of race that we call "temperament". Is it notinconceivable that the particular collective qualities of mind that havefashioned a culture are not precisely the same as were responsible forthe growth of a particular linguistic morphology? This question takes usinto the heart of the most difficult problems of social psychology. Itis doubtful if any one has yet attained to sufficient clarity on thenature of the historical process and on the ultimate psychologicalfactors involved in linguistic and cultural drifts to answer itintelligently. I can only very briefly set forth my own views, or rathermy general attitude. It would be very difficult to prove that"temperament", the general emotional disposition of a people[192], isbasically responsible for the slant and drift of a culture, however muchit may manifest itself in an individual's handling of the elements ofthat culture. But granted that temperament has a certain value for theshaping of culture, difficult though it be to say just how, it does notfollow that it has the same value for the shaping of language. It isimpossible to show that the form of a language has the slightestconnection with national temperament. Its line of variation, its drift, runs inexorably in the channel ordained for it by its historicantecedents; it is as regardless of the feelings and sentiments of itsspeakers as is the course of a river of the atmospheric humors of thelandscape. I am convinced that it is futile to look in linguisticstructure for differences corresponding to the temperamental variationswhich are supposed to be correlated with race. In this connection it iswell to remember that the emotional aspect of our psychic life is butmeagerly expressed in the build of language[193]. 

[Footnote 192: "Temperament" is a difficult term to work with. A greatdeal of what is loosely charged to national "temperament" is reallynothing but customary behavior, the effect of traditional ideals ofconduct. In a culture, for instance, that does not look kindly upondemonstrativeness, the natural tendency to the display of emotionbecomes more than normally inhibited. It would be quite misleading toargue from the customary inhibition, a cultural fact, to the nativetemperament. But ordinarily we can get at human conduct only as it isculturally modified. Temperament in the raw is a highly elusive thing. ]

[Footnote 193: See pages 39, 40. ]

[Transcriber's note: Footnote 193 refers to the paragraph beginning online 1256. ]

Language and our thought-grooves are inextricably interwoven, are, in asense, one and the same. As there is nothing to show that there aresignificant racial differences in the fundamental conformation ofthought, it follows that the infinite variability of linguistic form, another name for the infinite variability of the actual process ofthought, cannot be an index of such significant racial differences. Thisis only apparently a paradox. The latent content of all languages is thesame--the intuitive _science_ of experience. It is the manifest formthat is never twice the same, for this form, which we call linguisticmorphology, is nothing more nor less than a collective _art_ of thought, an art denuded of the irrelevancies of individual sentiment. At lastanalysis, then, language can no more flow from race as such than can thesonnet form. 

Nor can I believe that culture and language are in any true sensecausally related. Culture may be defined as _what_ a society does andthinks. Language is a particular _how_ of thought. It is difficult tosee what particular causal relations may be expected to subsist betweena selected inventory of experience (culture, a significant selectionmade by society) and the particular manner in which the societyexpresses all experience. The drift of culture, another way of sayinghistory, is a complex series of changes in society's selectedinventory--additions, losses, changes of emphasis and relation. Thedrift of language is not properly concerned with changes of content atall, merely with changes in formal expression. It is possible, inthought, to change every sound, word, and concrete concept of a languagewithout changing its inner actuality in the least, just as one can pourinto a fixed mold water or plaster or molten gold. If it can be shownthat culture has an innate form, a series of contours, quite apart fromsubject-matter of any description whatsoever, we have a something inculture that may serve as a term of comparison with and possibly ameans of relating it to language. But until such purely formal patternsof culture are discovered and laid bare, we shall do well to hold thedrifts of language and of culture to be non-comparable and unrelatedprocesses. From this it follows that all attempts to connect particulartypes of linguistic morphology with certain correlated stages ofcultural development are vain. Rightly understood, such correlations arerubbish. The merest _coup d'oeil_ verifies our theoretical argument onthis point. Both simple and complex types of language of an indefinitenumber of varieties may be found spoken at any desired level of culturaladvance. When it comes to linguistic form, Plato walks with theMacedonian swineherd, Confucius with the head-hunting savage of Assam. 

It goes without saying that the mere content of language is intimatelyrelated to culture. A society that has no knowledge of theosophy needhave no name for it; aborigines that had never seen or heard of a horsewere compelled to invent or borrow a word for the animal when they madehis acquaintance. In the sense that the vocabulary of a language more orless faithfully reflects the culture whose purposes it serves it isperfectly true that the history of language and the history of culturemove along parallel lines. But this superficial and extraneous kind ofparallelism is of no real interest to the linguist except in so far asthe growth or borrowing of new words incidentally throws light on theformal trends of the language. The linguistic student should never makethe mistake of identifying a language with its dictionary. 

If both this and the preceding chapter have been largely negative intheir contentions, I believe that they have been healthily so. There isperhaps no better way to learn the essential nature of speech than torealize what it is not and what it does not do. Its superficialconnections with other historic processes are so close that it needs tobe shaken free of them if we are to see it in its own right. Everythingthat we have so far seen to be true of language points to the fact thatit is the most significant and colossal work that the human spirit hasevolved--nothing short of a finished form of expression for allcommunicable experience. This form may be endlessly varied by theindividual without thereby losing its distinctive contours; and it isconstantly reshaping itself as is all art. Language is the most massiveand inclusive art we know, a mountainous and anonymous work ofunconscious generations. 

XI

LANGUAGE AND LITERATURE

Languages are more to us than systems of thought-transference. They areinvisible garments that drape themselves about our spirit and give apredetermined form to all its symbolic expression. When the expressionis of unusual significance, we call it literature. [194] Art is sopersonal an expression that we do not like to feel that it is bound topredetermined form of any sort. The possibilities of individualexpression are infinite, language in particular is the most fluid ofmediums. Yet some limitation there must be to this freedom, someresistance of the medium. In great art there is the illusion of absolutefreedom. The formal restraints imposed by the material--paint, black andwhite, marble, piano tones, or whatever it may be--are not perceived; itis as though there were a limitless margin of elbow-room between theartist's fullest utilization of form and the most that the material isinnately capable of. The artist has intuitively surrendered to theinescapable tyranny of the material, made its brute nature fuse easilywith his conception. [195] The material "disappears" precisely becausethere is nothing in the artist's conception to indicate that any othermaterial exists. For the time being, he, and we with him, move in theartistic medium as a fish moves in the water, oblivious of the existenceof an alien atmosphere. No sooner, however, does the artist transgressthe law of his medium than we realize with a start that there is amedium to obey. 

[Footnote 194: I can hardly stop to define just what kind of expressionis "significant" enough to be called art or literature. Besides, I donot exactly know. We shall have to take literature for granted. ]

[Footnote 195: This "intuitive surrender" has nothing to do withsubservience to artistic convention. More than one revolt in modern arthas been dominated by the desire to get out of the material just what itis really capable of. The impressionist wants light and color becausepaint can give him just these; "literature" in painting, the sentimentalsuggestion of a "story, " is offensive to him because he does not wantthe virtue of his particular form to be dimmed by shadows from anothermedium. Similarly, the poet, as never before, insists that words meanjust what they really mean. ]

Language is the medium of literature as marble or bronze or clay are thematerials of the sculptor. Since every language has its distinctivepeculiarities, the innate formal limitations--and possibilities--of oneliterature are never quite the same as those of another. The literaturefashioned out of the form and substance of a language has the color andthe texture of its matrix. The literary artist may never be conscious ofjust how he is hindered or helped or otherwise guided by the matrix, butwhen it is a question of translating his work into another language, thenature of the original matrix manifests itself at once. All his effectshave been calculated, or intuitively felt, with reference to the formal"genius" of his own language; they cannot be carried over without lossor modification. Croce[196] is therefore perfectly right in saying thata work of literary art can never be translated. Nevertheless literaturedoes get itself translated, sometimes with astonishing adequacy. Thisbrings up the question whether in the art of literature there are notintertwined two distinct kinds or levels of art--a generalized, non-linguistic art, which can be transferred without loss into an alienlinguistic medium, and a specifically linguistic art that is nottransferable. [197] I believe the distinction is entirely valid, thoughwe never get the two levels pure in practice. Literature moves inlanguage as a medium, but that medium comprises two layers, the latentcontent of language--our intuitive record of experience--and theparticular conformation of a given language--the specific how of ourrecord of experience. Literature that draws its sustenance mainly--neverentirely--from the lower level, say a play of Shakespeare's, istranslatable without too great a loss of character. If it moves in theupper rather than in the lower level--a fair example is a lyric ofSwinburne's--it is as good as untranslatable. Both types of literaryexpression may be great or mediocre. 

[Footnote 196: See Benedetto Croce, "Aesthetic. "]

[Footnote 197: The question of the transferability of art productionsseems to me to be of genuine theoretic interest. For all that we speakof the sacrosanct uniqueness of a given art work, we know very well, though we do not always admit it, that not all productions are equallyintractable to transference. A Chopin étude is inviolate; it movesaltogether in the world of piano tone. A Bach fugue is transferable intoanother set of musical timbres without serious loss of estheticsignificance. Chopin plays with the language of the piano as though noother language existed (the medium "disappears"); Bach speaks thelanguage of the piano as a handy means of giving outward expression to aconception wrought in the generalized language of tone. ]

There is really no mystery in the distinction. It can be clarified alittle by comparing literature with science. A scientific truth isimpersonal, in its essence it is untinctured by the particularlinguistic medium in which it finds expression. It can as readilydeliver its message in Chinese[198] as in English. Nevertheless it musthave some expression, and that expression must needs be a linguisticone. Indeed the apprehension of the scientific truth is itself alinguistic process, for thought is nothing but language denuded of itsoutward garb. The proper medium of scientific expression is therefore ageneralized language that may be defined as a symbolic algebra of whichall known languages are translations. One can adequately translatescientific literature because the original scientific expression isitself a translation. Literary expression is personal and concrete, butthis does not mean that its significance is altogether bound up with theaccidental qualities of the medium. A truly deep symbolism, forinstance, does not depend on the verbal associations of a particularlanguage but rests securely on an intuitive basis that underlies alllinguistic expression. The artist's "intuition, " to use Croce's term, isimmediately fashioned out of a generalized human experience--thought andfeeling--of which his own individual experience is a highly personalizedselection. The thought relations in this deeper level have no specificlinguistic vesture; the rhythms are free, not bound, in the firstinstance, to the traditional rhythms of the artist's language. Certainartists whose spirit moves largely in the non-linguistic (better, in thegeneralized linguistic) layer even find a certain difficulty in gettingthemselves expressed in the rigidly set terms of their accepted idiom. One feels that they are unconsciously striving for a generalized artlanguage, a literary algebra, that is related to the sum of all knownlanguages as a perfect mathematical symbolism is related to all theroundabout reports of mathematical relations that normal speech iscapable of conveying. Their art expression is frequently strained, itsounds at times like a translation from an unknown original--which, indeed, is precisely what it is. These artists--Whitmans andBrownings--impress us rather by the greatness of their spirit than thefelicity of their art. Their relative failure is of the greatestdiagnostic value as an index of the pervasive presence in literature ofa larger, more intuitive linguistic medium than any particular language. 

[Footnote 198: Provided, of course, Chinese is careful to provide itselfwith the necessary scientific vocabulary. Like any other language, itcan do so without serious difficulty if the need arises. ]

Nevertheless, human expression being what it is, the greatest--or shallwe say the most satisfying--literary artists, the Shakespeares andHeines, are those who have known subconsciously to fit or trim thedeeper intuition to the provincial accents of their daily speech. Inthem there is no effect of strain. Their personal "intuition" appears asa completed synthesis of the absolute art of intuition and the innate, specialized art of the linguistic medium. With Heine, for instance, oneis under the illusion that the universe speaks German. The material"disappears. "

Every language is itself a collective art of expression. There isconcealed in it a particular set of esthetic factors--phonetic, rhythmic, symbolic, morphological--which it does not completely sharewith any other language. These factors may either merge their potencieswith those of that unknown, absolute language to which I havereferred--this is the method of Shakespeare and Heine--or they may weavea private, technical art fabric of their own, the innate art of thelanguage intensified or sublimated. The latter type, the moretechnically "literary" art of Swinburne and of hosts of delicate "minor"poets, is too fragile for endurance. It is built out of spiritualizedmaterial, not out of spirit. The successes of the Swinburnes are asvaluable for diagnostic purposes as the semi-failures of the Brownings. They show to what extent literary art may lean on the collective art ofthe language itself. The more extreme technical practitioners may soover-individualize this collective art as to make it almost unendurable. One is not always thankful to have one's flesh and blood frozen toivory. 

An artist must utilize the native esthetic resources of his speech. Hemay be thankful if the given palette of colors is rich, if thespringboard is light. But he deserves no special credit for felicitiesthat are the language's own. We must take for granted this language withall its qualities of flexibility or rigidity and see the artist's workin relation to it. A cathedral on the lowlands is higher than a stick onMont Blanc. In other words, we must not commit the folly of admiring aFrench sonnet because the vowels are more sonorous than our own or ofcondemning Nietzsche's prose because it harbors in its texturecombinations of consonants that would affright on English soil. To sojudge literature would be tantamount to loving "Tristan und Isolde"because one is fond of the timbre of horns. There are certain thingsthat one language can do supremely well which it would be almost vainfor another to attempt. Generally there are compensations. The vocalismof English is an inherently drabber thing than the vowel scale ofFrench, yet English compensates for this drawback by its greaterrhythmical alertness. It is even doubtful if the innate sonority of aphonetic system counts for as much, as esthetic determinant, as therelations between the sounds, the total gamut of their similarities andcontrasts. As long as the artist has the wherewithal to lay out hissequences and rhythms, it matters little what are the sensuous qualitiesof the elements of his material. 

The phonetic groundwork of a language, however, is only one of thefeatures that give its literature a certain direction. Far moreimportant are its morphological peculiarities. It makes a great deal ofdifference for the development of style if the language can or cannotcreate compound words, if its structure is synthetic or analytic, if thewords of its sentences have considerable freedom of position or arecompelled to fall into a rigidly determined sequence. The majorcharacteristics of style, in so far as style is a technical matter ofthe building and placing of words, are given by the language itself, quite as inescapably, indeed, as the general acoustic effect of verse isgiven by the sounds and natural accents of the language. These necessaryfundamentals of style are hardly felt by the artist to constrain hisindividuality of expression. They rather point the way to thosestylistic developments that most suit the natural bent of the language. It is not in the least likely that a truly great style can seriouslyoppose itself to the basic form patterns of the language. It not onlyincorporates them, it builds on them. The merit of such a style as W. H. Hudson's or George Moore's[199] is that it does with ease and economywhat the language is always trying to do. Carlylese, though individualand vigorous, is yet not style; it is a Teutonic mannerism. Nor is theprose of Milton and his contemporaries strictly English; it issemi-Latin done into magnificent English words. 

[Footnote 199: Aside from individual peculiarities of diction, theselection and evaluation of particular words as such. ]

It is strange how long it has taken the European literatures to learnthat style is not an absolute, a something that is to be imposed on thelanguage from Greek or Latin models, but merely the language itself, running in its natural grooves, and with enough of an individual accentto allow the artist's personality to be felt as a presence, not as anacrobat. We understand more clearly now that what is effective andbeautiful in one language is a vice in another. Latin and Eskimo, withtheir highly inflected forms, lend themselves to an elaborately periodicstructure that would be boring in English. English allows, even demands, a looseness that would be insipid in Chinese. And Chinese, with itsunmodified words and rigid sequences, has a compactness of phrase, aterse parallelism, and a silent suggestiveness that would be too tart, too mathematical, for the English genius. While we cannot assimilate theluxurious periods of Latin nor the pointilliste style of the Chineseclassics, we can enter sympathetically into the spirit of these alientechniques. 

I believe that any English poet of to-day would be thankful for theconcision that a Chinese poetaster attains without effort. Here is anexample:[200]

[Footnote 200: Not by any means a great poem, merely a bit of occasionalverse written by a young Chinese friend of mine when he left Shanghaifor Canada. ]

Wu-river[201] stream mouth evening sun sink, North look Liao-Tung, [202] not see home. Steam whistle several noise, sky-earth boundless, Float float one reed out Middle-Kingdom. 

[Footnote 201: The old name of the country about the mouth of theYangtsze. ]

[Footnote 202: A province of Manchuria. ]

These twenty-eight syllables may be clumsily interpreted: "At the mouthof the Yangtsze River, as the sun is about to sink, I look north towardLiao-Tung but do not see my home. The steam-whistle shrills severaltimes on the boundless expanse where meet sky and earth. The steamer, floating gently like a hollow reed, sails out of the MiddleKingdom. "[203] But we must not envy Chinese its terseness unduly. Ourmore sprawling mode of expression is capable of its own beauties, andthe more compact luxuriance of Latin style has its loveliness too. There are almost as many natural ideals of literary style as there arelanguages. Most of these are merely potential, awaiting the hand ofartists who will never come. And yet in the recorded texts of primitivetradition and song there are many passages of unique vigor and beauty. The structure of the language often forces an assemblage of conceptsthat impresses us as a stylistic discovery. Single Algonkin words arelike tiny imagist poems. We must be careful not to exaggerate afreshness of content that is at least half due to our freshness ofapproach, but the possibility is indicated none the less of utterlyalien literary styles, each distinctive with its disclosure of thesearch of the human spirit for beautiful form. 

[Footnote 203: I. E. , China. ]

Probably nothing better illustrates the formal dependence of literatureon language than the prosodic aspect of poetry. Quantitative verse wasentirely natural to the Greeks, not merely because poetry grew up inconnection with the chant and the dance, [204] but because alternationsof long and short syllables were keenly live facts in the daily economyof the language. The tonal accents, which were only secondarily stressphenomena, helped to give the syllable its quantitative individuality. When the Greek meters were carried over into Latin verse, there wascomparatively little strain, for Latin too was characterized by an acuteawareness of quantitative distinctions. However, the Latin accent wasmore markedly stressed than that of Greek. Probably, therefore, thepurely quantitative meters modeled after the Greek were felt as a shademore artificial than in the language of their origin. The attempt tocast English verse into Latin and Greek molds has never been successful. The dynamic basis of English is not quantity, [205] but stress, thealternation of accented and unaccented syllables. This fact givesEnglish verse an entirely different slant and has determined thedevelopment of its poetic forms, is still responsible for the evolutionof new forms. Neither stress nor syllabic weight is a very keenpsychologic factor in the dynamics of French. The syllable has greatinherent sonority and does not fluctuate significantly as to quantityand stress. Quantitative or accentual metrics would be as artificial inFrench as stress metrics in classical Greek or quantitative or purelysyllabic metrics in English. French prosody was compelled to develop onthe basis of unit syllable-groups. Assonance, later rhyme, could not butprove a welcome, an all but necessary, means of articulating orsectioning the somewhat spineless flow of sonorous syllables. Englishwas hospitable to the French suggestion of rhyme, but did not seriouslyneed it in its rhythmic economy. Hence rhyme has always been strictlysubordinated to stress as a somewhat decorative feature and has beenfrequently dispensed with. It is no psychologic accident that rhyme camelater into English than in French and is leaving it sooner. [206] Chineseverse has developed along very much the same lines as French verse. Thesyllable is an even more integral and sonorous unit than in French, while quantity and stress are too uncertain to form the basis of ametric system. Syllable-groups--so and so many syllables per rhythmicunit--and rhyme are therefore two of the controlling factors in Chineseprosody. The third factor, the alternation of syllables with level toneand syllables with inflected (rising or falling) tone, is peculiar toChinese. 

[Footnote 204: Poetry everywhere is inseparable in its origins from thesinging voice and the measure of the dance. Yet accentual and syllabictypes of verse, rather than quantitative verse, seem to be theprevailing norms. ]

[Footnote 205: Quantitative distinctions exist as an objective fact. They have not the same inner, psychological value that they had inGreek. ]

[Footnote 206: Verhaeren was no slave to the Alexandrine, yet heremarked to Symons, _à propos_ of the translation of _Les Aubes_, thatwhile he approved of the use of rhymeless verse in the English version, he found it "meaningless" in French. ]

To summarize, Latin and Greek verse depends on the principle ofcontrasting weights; English verse, on the principle of contrastingstresses; French verse, on the principles of number and echo; Chineseverse, on the principles of number, echo, and contrasting pitches. Eachof these rhythmic systems proceeds from the unconscious dynamic habit ofthe language, falling from the lips of the folk. Study carefully thephonetic system of a language, above all its dynamic features, and youcan tell what kind of a verse it has developed--or, if history hasplayed pranks with its phychology, what kind of verse it should havedeveloped and some day will. 

Whatever be the sounds, accents, and forms of a language, however theselay hands on the shape of its literature, there is a subtle law ofcompensations that gives the artist space. If he is squeezed a bit here, he can swing a free arm there. And generally he has rope enough to hanghimself with, if he must. It is not strange that this should be so. Language is itself the collective art of expression, a summary ofthousands upon thousands of individual intuitions. The individual goeslost in the collective creation, but his personal expression has leftsome trace in a certain give and flexibility that are inherent in allcollective works of the human spirit. The language is ready, or can bequickly made ready, to define the artist's individuality. If noliterary artist appears, it is not essentially because the language istoo weak an instrument, it is because the culture of the people is notfavorable to the growth of such personality as seeks a truly individualverbal expression. 

INDEX

_Note_. Italicized entries are names of languages or groups of languages. 

A

Abbreviation of stem, Accent, stress, as grammatical process, importance of, metrical value of"Accent, ""Adam's apple, "Adjective, Affixation, Affixing languages, African languages, pitch in, Agglutination, Agglutinative languages, Agglutinative-fusional, Agglutinative-isolating, _Algonkin_ languages (N. Amer. ), Alpine race, Analogical leveling, Analytic tendency, Angles, _Anglo-Saxon_, Anglo-Saxon: culture, race, _Annamite_ (S. E. Asia), _Apache_ (N. Amer. ), _Arabic_, _Armenian_, Art, language as, transferability of, Articulation: ease of, types of, drift toward, Articulations: laryngeal, manner of consonantal, nasal, oral, place of consonantal, vocalic, _Aryan_. See _Indo-European_. Aspect, Association of concepts and speech elements, Associations fundamental to speech, _Athabaskan_ languages (N. Amer. ), Athabaskans, cultures of, _Attic_ dialect, Attribution, Auditory cycle in language, Australian culture, _Avestan_, B

Bach, Baltic race, _Bantu_ languages (Africa), Bantus, _Basque_ (Pyrenees), _Bengali_ (India), _Berber_. See _Hamitic_. Bohemians, _Bontoc Igorot_ (Philippines), Borrowing, morphological, Borrowing, word, phonetic adaptation in, resistances to, _Breton_, Bronchial tubes, Browning, Buddhism, influence of, _Burmese_, _Bushman_ (S. Africa), Bushmen, C

_Cambodgian_ (S. E. Asia), Carlyle, _Carrier_ (British Columbia), Case, See _Attribution_; _Object_; _Personal relations_; _Subject_. Case-system, history of, Caucasus, languages of, Celtic. See _Celts_. _Celtic_ languages, Celts, Brythonic, "Cerebral" articulations, Chaucer, English of, _Chimariko_ (N. California), _Chinese_: absence of affixes, analytic character, attribution, compounds, grammatical concepts illustrated, influence, "inner form, ", pitch accent, radical words, relational use of material words, sounds, stress, structure, style, survivals, morphological, symbolism, verse, word duplication, word order, _Chinook_ (N. Amer. ), _Chipewyan_ (N. Amer. ), C. Indians, Chopin, Christianity, influence of, Chukchi, Classification: of concepts, rigid, of linguistic types, See _Structure, linguistic_. "Clicks, "Composition, absence of, in certain languages, types of, word order as related to, Concepts, Concepts, grammatical: analysis of, in sentence, classification of, concrete, concrete relational, concreteness in, varying degree of, derivational, derivational, abstract, essential, grouping of, non-logical, lack of expression of certain, pure relational, radical, redistribution of, relational, thinning-out of significance of, types of, typical categories of, See _Structure, linguistic_. Concord, Concrete concepts. See _Concepts_. Conflict, Consonantal change, Consonants, combinations of, Coördinate sentences, _Corean_, Croce, Benedetto, Culture, language and, language as aspect of, language, race and, reflection of history of, in language, Culture areas, D

_Danish_, Demonstrative ideas, Dental articulations, Derivational concepts. See _Concepts_. Determinative structure, Dialects: causes of, compromise between, distinctness of, drifts in, diverging, drifts in, parallel, splitting up of, unity of, Diffusion, morphological, Diphthongs, Drift, linguistic, components of, determinants of, in English, direction of, direction of, illustrated in English, examples of general, in English, parallelisms in, speed of, See _Phonetic Law_; _Phonetic processes_. Duplication of words, _Dutch_, E

Elements of speech, Emotion, expression of: involuntary, linguistic, _English_: agentive suffix, analogical leveling, analytic tendency, animate and inanimate, aspect, attribution, case, history of, compounds, concepts, grammatical, in sentence, concepts, passage of concrete into derivational, consonantal change, culture of speakers of, desire, expression of, diminutive suffix, drift, duplication, word, esthetic qualities, feeling-tone, form, word, French influence on, function and form, fusing and juxtaposing, gender, Greek influence on, influence of, influence on, morphological, lack of deep, interrogative words, invariable words, tendency to, infixing, Latin influence on, loan-words, modality, number, order, word, parts of speech, patterning, formal, personal relations, phonetic drifts, history of, phonetic leveling, phonetic pattern, plurality, race of speakers of, reference, definiteness of, relational words, relations, genetic, rhythm, sentence, analysis of, sentence, dependence of word on, sound-imitative words, sounds, stress and pitch, structure, survivals, morphological, symbolism, syntactic adhesions, syntactic values, transfer of, tense, verb, syntactic relations of, verse, vocalic change, word and element, analysis of, _English, Middle_, English people, _Eskimo_, Eskimos, _Ewe_ (Guinea coast, Africa), Expiratory sounds, "Explosives, "

F

Faucal position, Feeling-tones of words, Fijians, _Finnish_, Finns, _Flemish_, "Foot, feet" (English), history of, Form, cultural, feeling of language for, "inner, "Form, linguistic: conservatism of, differences of, mechanical origin of, elaboration of, reasons for, function and, independence of, grammatical concepts embodied in, grammatical processes embodying, permanence of different aspects of, relative, twofold consideration of, See _Structure, linguistic_. Form-classes, See _Gender_. Formal units of speech, "Formlessness, inner, "_Fox_ (N. Amer. ), _French_: analytical tendency, esthetic qualities, gender, influence, order, word, plurality, sounds, sounds as words, single, stress, structure, tense forms, verse, French, Norman, French people, Freud, Fricatives, _Frisian_, _Ful_ (Soudan), Function, independence of form and, Functional units of speech, Fusion, Fusional languages, See _Fusion_. Fusional-agglutinative, Fusional-isolating, "Fuss, Füsse" (German), history of, G

_Gaelic_, Gender, _German_: French influence on, grammatical concepts in sentence, Latin influence on, phonetic drifts, history of, plurality, relations, sound-imitative words, sounds, tense forms, "umlaut, " unanalyzable words, resistance to, _German, High_, _German, Middle High_, _German, Old High_, _Germanic_ languages, _Germanic, West_, Germans, Gesture languages, Ginneken, Jac van, Glottal cords, action of, Glottal stop, _Gothic_, Grammar, Grammatical element, Grammatical concepts. See _Concepts, grammatical_. Grammatical processes: classified by, languages, particular, development by each language of, types of, variety of, use in one language of, _Greek_, dialectic history of, _Greek, classical_: affixing, compounds, concord, infixing, influence, pitch accent, plurality, reduplicated perfects, stress, structure, synthetic character, verse, _Greek, modern_, H

_Haida_ (British Columbia), _Hamitic_ languages (N. Africa), _Hausa_ (Soudan), _Hebrew_, Heine, Hesitation, History, linguistic, _Hokan_ languages (N. Amer. ), _Hottentot_ (S. Africa), Hudson, W. H. , Humming, _Hupa_ (N. California), Hupa Indians, I

_Icelandic, Old_, India, languages of, Indians, American, languages of, See also _Algonkin_; _Athabaskan_; _Chimariko_; _Chinook_; _Eskimo_; _Fox_; _Haida_; _Hokan_; _Hupa_; _Iroquois_; _Karok_; _Kwakiutl_; _Nahuatl_; _Nass_; _Navaho_; _Nootka_; _Ojibwa_; _Paiute_; _Sahaptin_; _Salinan_; _Shasta_; _Siouan_; _Sioux_; _Takelma_; _Tlingit_; _Tsimshian_; _Washo_; _Yana_; _Yokuts_; _Yurok_. _Indo-Chinese_ languages, _Indo-European_, _Indo-Iranian_ languages, Infixes, Inflection. See _Inflective languages_. Inflective languages, Influence: cultural, reflected in language, morphological, of alien language, phonetic, of alien language, Inspiratory sounds, Interjections, Irish, _Irish_, _Iroquois_ (N. Amer. ), Isolating languages, _Italian_, "Its, " history of, J

_Japanese_, Jutes, Juxtaposing. See _Agglutination_. 

K

_Karok_ (N. California), K. Indians, _Khmer_. See _Cambodgian_. Knowledge, source of, as grammatical category, _Koine_, _Kwakiutl_ (British Columbia), L

Labial trills, Language: associations in, associations underlying elements of, auditory cycle in, concepts expressed in, a cultural function, definition of, diversity of, elements of, emotion expressed in, feeling-tones in, grammatical concepts of, grammatical processes of, historical aspects of, imitations of sounds, not evolved from, influences on, exotic, interjections, not evolved from, literature and, modifications and transfers of typical form of, an "overlaid" function, psycho-physical basis of, race, culture and, simplification of experience in, sounds of, structure of, thought and, universality of, variability of, volition expressed in, Larynx, Lateral sounds, _Latin_: attribution, concord, infixing, influence of, objective _-m_, order of words, plurality, prefixes and suffixes, reduplicated perfects, relational concepts expressed, sentence-word, sound as word in, single, structure, style, suffixing character, syntactic nature of sentence, synthetic character, verse, word and element in, analysis of, _Lettish_, Leveling, phonetic, See _Analogical leveling_. Lips, action of, Literature: compensations in, formal, language and, levels in, linguistic, medium of, language as, science and, Literature, determinants of: linguistic, metrical, morphological, phonetic, _Lithuanian_, Localism, Localization of speech, _Loucheux_ (N. Amer. ), L. Indians, Lungs, Luther, German of, M

_Malay_, M. Race, _Malayan_, _Malayo-Polynesian_ languages, _Manchu_, _Manx_, "Maus, Mäuse" (German), history of, Mediterranean race, _Melanesian_ languages, Meter. See _Verse_. Milton, Mixed-relational languages, complex, simple, Modality, _Mon-Khmer_ (S. E. Asia), Moore, George, Morphological features, diffusion of, Morphology. See _Structure, linguistic_. "Mouse, mice" (English), history of, _Munda_ languages (E. India), Murmuring, Mutation, vocalic, N

_Nahuatl_ (Mexico), Nasal sounds, "Nasal twang, "Nasalized stops, _Nass_ (British Columbia), Nationality, _Navaho_ (Arizona, New Mexico), N. Indians, Nietzsche, _Nootka_ (Vancouver Id. ), Nose, action of, Noun, Nouns, classification of, Number, See _Plurality_. 

O

Object, See _Personal relations_. _Ojibwa_ (N, Amer. ), Onomatopoetic theory of origin of speech, Oral sounds, Order, word, composition as related to, fixed, English tendency, sentence molded by, significance of, fundamental, Organs of speech, action of, P

_Paiute_ (N. Amer. ), Palate, action of soft, articulations of, _Pali_ (India), _Papuan_ languages, Papuans, Parts of speech, Pattern: formal, phonetic, _Persian_, Person, Personal relations, Phonetic adaptation, Phonetic diffusion, Phonetic law: basis of, direction of, examples of, influence of, on morphology, influence of morphology on, regularity of, significance of, spread of, slow, See _Leveling, phonetic_; _Pattern, phonetic_. Phonetic processes, form caused by, differences of, parallel drifts in, Pitch, grammatical use of, metrical use of, production of, significant differences in, Plains Indians, gesture language of, "Plattdeutsch, "Plurality: classification of concept of, variable, a concrete relational category, a derivational or radical concept, expression of, multiple, See _Number_. Poles, _Polynesian_, Polynesians, Polysynthetic languages, _Portuguese_, Predicate, Prefixes, Prefixing languages, Preposition, Psycho-physical aspect of speech, Pure-relational languages, complex, simple, Q

Qualifying concepts. See _Concepts, derivational_. Quality: of speech sounds, of individual's voice, Quantity of speech sounds, R

Race, language and, lack of correspondence between, language and, theoretical relation between, language as correlated with, English, language, culture and, correspondence between, language, culture and, independence of, Radical concepts. See _Concepts_. Radical element, Radical word, "Reading from the lips, "Reduplication, Reference, definite and indefinite, Repetition of stem, See _Reduplication_. Repression of impulse, Rhyme, Rolled consonants, _Romance_ languages, Root, _Roumanian_, Rounded vowels, _Russian_, S

_Sahaptin_ languages (N. Amer. ), _Salinan_ (S. W. California), _Sanskrit_ (India), Sarcee Indians, _Saxon_: _Low_, _Old_, _Upper_, Saxons, _Scandinavian_, See _Danish_; _Icelandic_; _Swedish_. Scandinavians, Scotch, _Scotch, Lowland_, _Semitic languages_, Sentence, binding words into, methods of, stress in, influence of, word-order in, Sequence. See _Order of words_. Shakespeare: art of, English of, _Shasta_ (N. California), _Shilh_ (Morocco), _Shilluk_ (Nile headwaters), _Siamese_, Singing, _Siouan_ languages (N. Amer. ), _Sioux_ (Dakota), _Slavic_ languages, Slavs, _Somali_ (E. Africa), _Soudanese_ languages, Sound-imitative words, Sounds of speech, adjustments involved in, muscular, adjustments involved in certain, inhibition of, basic importance of, classification of, combinations of, conditioned appearance of, dynamics of, illusory feelings in regard to, "inner" or "ideal" system of, place in phonetic pattern of, production of, values of, psychological, variability of, _Spanish_, Speech. See _Language_. Spirants, Splitting of sounds, Stem, Stock, linguistic, Stopped consonants (_or_ stops), Stress. See _Accent_. Structure, linguistic, conservatism of, differences of, intuitional forms of, Structure, linguistic, types of: classification of, by character of concepts, by degree of fusion, by degree of synthesis, by formal processes, from threefold standpoint, into "formal" and "formless, " classifying, difficulties in, examples of, mixed, reality of, validity of conceptual, historical test of, Style, Subject, See _Personal relations_. Subject of discourse, Suffixes, Suffixing, Suffixing languages, Survivals, morphological, _Swedish_, Swinburne, Swiss, French, Syllabifying, Symbolic languages, Symbolic processes, Symbolic-fusional, Symbolic-isolating, Symons, Syntactic adhesions, Syntactic relations: primary methods of expressing, transfer of values in, See _Concepts, relational_; _Concord_; _Order, word_; _Personal relations_; _Sentence_. Synthetic tendency, T

_Takelma_ (S. W. Oregon), Teeth, articulations of, Telegraph code, Temperament, Tense, Teutonic race. See _Baltic race_. Thinking, types of, Thought, relation of language to, Throat, articulations of, _Tibetan_, Time. See _Tense_. _Tlingit_ (S. Alaska), T. Indians, Tongue, action of, Transfer, types of linguistic, Trills, _Tsimshian_ (British Columbia), See _Nass_. _Turkish_, Types, linguistic, change of, See _Structure, linguistic_. 

U

_Ugro-Finnic_, "Umlaut. " See _Mutation, vocalic_. United States: culture in, race in, _Ural-Altaic_ languages, Uvula, V

Values: "hesitation, " morphologic, phonetic, variability in, of components of drift, Variations, linguistic: dialect, historical, individual, Verb, syntactic relations expressed in, Verhaeren, Verse: accentual, linguistic determinants of, quantitative, syllabic, Vocalic change, See _Mutation, vocalic_. Voice, production of, Voiced sounds, Voiceless: laterals, nasals, sounds, trills, vowels, "Voicelessness, " production of, Volition expressed in speech, Vowels, W

Walking, a biological function, _Washo_ (Nevada), _Welsh_, Westermann, D. , Whisper, Whitman, "Whom, " use and drift of, Word, definition of, syntactic origin of complex, "twilight" type of, types of, formal, Written language, Y

_Yana_ (N. California), _Yiddish_, _Yokuts_ (S. California), _Yurok_ (N. W. California), Y. Indians, Z

_Zaconic_ dialect of Greek,