In this short sentence of five words there are expressed, therefore, thirteen distinct concepts, of which three are radical and concrete, two derivational, and eight relational. Perhaps the most striking result of the analysis is a renewed realization of the curious lack of accord in our language between function and form. The method of suffixing is used both for derivational and for relational elements; independent words or radical elements express both concrete ideas (objects, activities, qualities) and relational ideas (articles like the and a; words defining case relations, like of, to, for, with, by; words defining local relations, like in, on, at); the same relational concept may be expressed more than once (thus, the singularity of farmer is both negatively expressed in the noun and positively in the verb); and one element may convey a group of interwoven concepts rather than one definite concept alone (thus the -s of kills embodies no less than four logically independent relations).

Our analysis may seem a bit labored, but only because we are so accustomed to our own well-worn grooves of expression that they have come to be felt as inevitable. Yet destructive analysis of the familiar is the only method of approach to an understanding of fundamentally different modes of expression. When one has learned to feel what is fortuitous or illogical or unbalanced in the structure of his own language, he is already well on the way towards a sympathetic grasp of the expression of the various classes of concepts in alien types of speech. Not everything that is "outlandish" is intrinsically illogical or far-fetched. It is often precisely the familiar that a wider perspective reveals as the curiously exceptional. From a purely logical standpoint it is obvious that there is no inherent reason why the concepts expressed in our sentence should have been singled out, treated, and grouped as they have been and not otherwise. The sentence is the outgrowth of historical and of unreasoning psychological forces rather than of a logical synthesis of elements that have been clearly grasped in their individuality. This is the case, to a greater or less degree, in all languages, though in the forms of many we find a more coherent, a more consistent, reflection than in our English forms of that unconscious analysis into individual concepts which is never entirely absent from speech, however it may be complicated with or overlaid by the more irrational factors.

A cursory examination of other languages, near and far, would soon show that some or all of the thirteen concepts that our sentence happens to embody may not only be expressed in different form but that they may be differently grouped among themselves; that some among them may be dispensed with; and that other concepts, not considered worth expressing in English idiom, may be treated as absolutely indispensable to the intelligible rendering of the proposition. First as to a different method of handling such concepts as we have found expressed in the English sentence. If we turn to German, we find that in the equivalent sentence (Der Bauer toetet das Entelein) the definiteness of reference expressed by the English the is unavoidably coupled with three other concepts—number (both der and das are explicitly singular), case (der is subjective; das is subjective or objective, by elimination therefore objective), and gender, a new concept of the relational order that is not in this case explicitly involved in English (der is masculine, das is neuter). Indeed, the chief burden of the expression of case, gender, and number is in the German sentence borne by the particles of reference rather than by the words that express the concrete concepts (Bauer, Entelein) to which these relational concepts ought logically to attach themselves. In the sphere of concrete concepts too it is worth noting that the German splits up the idea of "killing" into the basic concept of "dead" (tot) and the derivational one of "causing to do (or be) so and so" (by the method of vocalic change, toet-); the German toet-et (analytically tot-vowel change-et) "causes to be dead" is, approximately, the formal equivalent of our dead-en-s, though the idiomatic application of this latter 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, for instance, also in Nootka and Sioux.]

Wandering still further afield, we may glance at the Yana method of expression. Literally translated, the equivalent Yana sentence would read something like "kill-s he farmer[56] he to duck-ling," in which "he" and "to" are rather awkward English renderings of a general third personal pronoun (he, she, it, or they) and an objective particle which indicates that the following noun is connected with the verb otherwise than as subject. The suffixed element in "kill-s" corresponds to the English suffix with the important exceptions that it makes no reference to the number of the subject and that the statement is known to be true, that it is vouched for by the speaker. Number is only indirectly expressed in the sentence in so far as there is no specific verb suffix indicating plurality of the subject nor specific plural elements in the two nouns. Had the statement been made on another's authority, a totally different "tense-modal" suffix would have had to be used. The pronouns of reference ("he") imply nothing by themselves as to number, gender, or case. Gender, indeed, is completely absent in Yana as a relational category.

[Footnote 56: Agriculture was not practised by the Yana. The verbal idea of "to farm" would probably be expressed in some such synthetic manner as "to dig-earth" or "to grow-cause." There are suffixed elements corresponding to -er and -ling.]

The Yana sentence has already illustrated the point that certain of our supposedly essential concepts may be ignored; both the Yana and the German sentence illustrate the further point that certain concepts may need expression for which an English-speaking person, or rather the English-speaking habit, finds no need whatever. One could go on and give endless examples of such deviations from English form, but we shall have to content ourselves with a few more indications. In the Chinese sentence "Man kill duck," which may be looked upon as the practical equivalent of "The man kills the duck," there is by no means present for the Chinese consciousness that childish, halting, empty feeling which we experience in the literal English translation. The three concrete concepts—two objects and an action—are each directly expressed by a monosyllabic word which is at the same time a radical element; the two relational concepts—"subject" and "object"—are expressed solely by the position of the concrete words before and after the word of action. And that is all. Definiteness or indefiniteness of reference, number, personality as an inherent aspect of the verb, tense, not to speak of gender—all these are given no expression in the Chinese sentence, which, for all that, is a perfectly adequate communication—provided, of course, there is that context, that background of mutual understanding that is essential to the complete intelligibility of all speech. Nor does this qualification impair our argument, for in the English sentence too we leave unexpressed a large number of ideas which are either taken for granted or which have been developed or are about to be developed in the course of the conversation. Nothing has been said, for example, in the English, German, Yana, or Chinese sentence as to the place relations of the farmer, the duck, the speaker, and the listener. Are the farmer and the duck both visible or is one or the other invisible from the point of view of the speaker, and are both placed within the horizon of the speaker, the listener, or of some indefinite point of reference "off yonder"? In other words, to paraphrase awkwardly certain latent "demonstrative" ideas, does this farmer (invisible to us but standing behind a door not far away from me, you being seated yonder well out of reach) kill that duckling (which belongs to you)? or does that farmer (who lives in your neighborhood and whom we see over there) kill that duckling (that belongs to him)? This type of demonstrative elaboration is 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, the concepts that must be expressed if language is to be a satisfactory means of communication? Clearly we must have, first of all, a large stock of basic or radical concepts, the concrete wherewithal of speech. We must have objects, actions, qualities to talk about, and these must have their corresponding symbols in independent words or in radical elements. No proposition, however abstract its intent, is humanly possible without a tying on at one or more points to the concrete world of sense. In every intelligible proposition at least two of these radical ideas must be expressed, though in exceptional cases one or even both may be understood from the context. And, secondly, such relational concepts must be expressed as moor the concrete concepts to each other and construct a definite, fundamental form of proposition. In this fundamental form there must be no doubt as to the nature of the relations that obtain between the concrete concepts. We must know what concrete concept is directly or indirectly related to what other, and how. If we wish to talk of a thing and an action, we must know if they are cooerdinately related to each other (e.g., "He is fond of wine and gambling"); 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" of which the action is predicated; or if, on the contrary, it is the end point, the "object" of the action. If I wish to communicate an intelligible idea about a farmer, a duckling, and the act of killing, it is not enough to state the linguistic symbols for these concrete ideas in any order, higgledy-piggledy, trusting that the hearer may construct some kind of a relational pattern out of the general probabilities of the case. The fundamental syntactic relations must be unambiguously expressed. I can afford to be silent on the subject of time and place and number and of a host of other possible types of concepts, but I can find no way of dodging the issue as to who is doing the killing. There is no known language that can or does dodge it, any more than it succeeds in saying something without the use of symbols for the concrete concepts.

We are thus once more reminded of the distinction between essential or unavoidable relational concepts and the dispensable type. The former are universally expressed, the latter are but sparsely developed in some languages, elaborated with a bewildering exuberance in others. But what prevents 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 all is said and done, a fundamental difference between a qualifying concept like the negative in unhealthy and a relational one like the number concept in books? If unhealthy may be roughly paraphrased as not healthy, may not books be just as legitimately paraphrased, barring the violence to English idiom, as several book? There are, indeed, languages in which the plural, if expressed at all, is conceived of in the same sober, restricted, one might almost say casual, spirit in which we feel the negative in unhealthy. For such languages the number concept has no syntactic significance whatever, is not essentially conceived of as defining a relation, but falls into the group of derivational or even of basic concepts. In English, however, as in French, German, Latin, Greek—indeed in all the languages that we have most familiarity with—the idea of number is not merely appended to a given concept of a thing. It may have something of this merely qualifying value, but its force extends far beyond. It infects much else in the sentence, molding other concepts, even such as have no intelligible relation to number, into forms that are said to correspond to or "agree with" the basic concept to which it is attached in the first instance. If "a man falls" but "men fall" in English, it is not because of any inherent change that has taken place in the nature of the action or because the idea of plurality inherent in "men" must, in the very nature of ideas, relate itself also to the action performed by these 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 the habit of doing—throwing a bold bridge between the two basically distinct types of concept, the concrete and the abstractly relational, infecting the latter, as it were, with the color and grossness of the former. By a certain violence of metaphor the material concept is forced to 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 two English phrases, "The white woman that comes" and "The white men that come," we are not reminded that gender, as well as number, may be elevated into a secondary relational concept. It would seem a little far-fetched to make of masculinity and femininity, crassly material, philosophically accidental concepts that they are, a means of relating quality 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 to inject into two such highly attenuated relational concepts as are expressed 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 to this: that-one-feminine-doer[57] one-feminine-white-doer feminine-doing-one-woman which-one-feminine-doer other[58]-one-now-come; and: that-several-masculine-doer several-masculine-white-doer masculine-doing-several-man which-several-masculine-doer other-several-now-come. Each word involves no less than four concepts, a radical concept (either properly concrete—white, man, woman, come—or demonstrative—that, which) and three relational concepts, selected from the categories of case, number, gender, person, and tense. Logically, only case[59] (the relation of woman or men to a following verb, of which to its antecedent, of that and white to woman or men, and of which to come) imperatively demands expression, and that only in connection with the concepts directly affected (there is, for instance, no need to be informed that the whiteness is a doing or doer's whiteness[60]). The other relational concepts are either merely parasitic (gender throughout; number in the demonstrative, the adjective, the relative, and the verb) or irrelevant to the essential syntactic form of the sentence (number in the noun; person; tense). An intelligent and sensitive Chinaman, accustomed as he is to cut to the very bone of linguistic form, might well say of the Latin sentence, "How pedantically imaginative!" It must be difficult for him, when first confronted by the illogical complexities of our European languages, to feel at home in an attitude that so largely confounds the subject-matter of speech with its formal pattern or, to be more accurate, that turns certain fundamentally concrete concepts to such attenuated relational uses.

[Footnote 57: "Doer," not "done to." This is a necessarily clumsy tag to represent 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-objective relation but also that of attribution.]

[Footnote 60: Except in so far as Latin uses this method as a rather awkward, roundabout method of establishing the attribution of the color to the particular object or person. In effect one cannot in Latin directly say that a person is white, merely that what is white is identical with the person who is, acts, or is acted upon in such and such a manner. In origin the feel of the Latin illa alba femina is really "that-one, the-white-one, (namely) the-woman"—three substantive ideas that are related to each other by a juxtaposition intended to convey an identity. English and Chinese express the attribution directly by means of order. In Latin the illa and alba may occupy almost any position in the sentence. It is important to observe that the subjective form of illa and alba, does not truly define a relation of these qualifying concepts to femina. Such a relation might be formally expressed via an attributive case, say the genitive (woman of whiteness). In Tibetan both the methods of order and of true case relation may be employed: woman white (i.e., "white woman") or white-of woman (i.e., "woman of whiteness, woman who is white, white woman").]

I have exaggerated somewhat the concreteness of our subsidiary or rather non-syntactical relational concepts In order that the essential facts might come out in bold relief. It goes without saying that a Frenchman has no clear sex notion in his mind when he speaks of un arbre ("a-masculine tree") or of une pomme ("a-feminine apple"). Nor have we, despite the grammarians, a very vivid sense of the present as contrasted 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 past occurrences and possible future ones rather than to present activity). In both the French and English instances the primary ideas of sex and time have become diluted by form-analogy and by extensions into the relational sphere, the concepts ostensibly indicated being now so vaguely delimited that it is rather the tyranny of usage than the need of their concrete expression that sways us in the selection of this or that form. If the thinning-out process continues long enough, we may eventually be left with a system of forms on our hands from which all the color of life has vanished and which merely persist by inertia, duplicating each other's secondary, syntactic functions with endless prodigality. Hence, in part, the complex conjugational systems of so many languages, in which differences of form are attended by no assignable differences of function. There must have been a time, for instance, though it antedates our earliest documentary evidence, when the type of tense formation represented by drove or sank differed in meaning, in however slightly nuanced a degree, from the type (killed, worked) which has now become established in English as the prevailing preterit formation, very much as we recognize a valuable distinction at present between both these types and the "perfect" (has driven, has killed) but may have ceased to do so at some point in the future.[62] Now form lives longer than its own conceptual content. Both are ceaselessly changing, but, on the whole, the form tends to linger on when the spirit has flown or changed its being. Irrational form, form for form's sake—however we term this tendency to hold on to formal distinctions once they have come to be—is as natural to the life of language as is the retention of modes of conduct that have long outlived the meaning they once had.

[Footnote 61: Aside, naturally, from the life and imminence that may be created 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 preterits are more literary or formal in tone than the perfects.]

There is another powerful tendency which makes for a formal elaboration that does not strictly correspond to clear-cut conceptual differences. This is the tendency to construct schemes of classification into which all the concepts of language must be fitted. Once we have made up our minds that all things are either definitely good or bad or definitely black or white, it is difficult to get into the frame of mind that recognizes that any particular thing may be both good and bad (in other words, indifferent) or both black and white (in other words, gray), still more difficult to realize that the good-bad or black-white categories may not apply at all. Language is in many respects as unreasonable and stubborn about its classifications as is such a mind. It must have its perfectly exclusive pigeon-holes and will tolerate no flying vagrants. Any concept that asks for expression must submit to the classificatory rules of the game, just as there are statistical surveys in which even the most convinced atheist must perforce be labeled Catholic, Protestant, or Jew or get no hearing. In English we have made up our minds that all action must be conceived of in reference to three standard times. If, therefore, we desire to state a proposition that is as true to-morrow as it was yesterday, we have to pretend that the present moment may be elongated fore and aft so as to take in all eternity.[63] In French we know once for all that an object is masculine or feminine, whether it be living or not; just as in many American and East Asiatic languages it must be understood to belong to a certain form-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 even said to "be" or "be handled in a definite way" (thus, in the Athabaskan languages and in Yana, "to carry" or "throw" a pebble is quite another thing than to carry or throw a log, linguistically no less than in terms of muscular experience). Such instances might be multiplied at will. It is almost as though at some period in the past the unconscious mind of the race had made a hasty inventory of experience, committed itself to a premature classification that allowed of no revision, and saddled the inheritors of its language with a science that they no longer quite believed in nor had the strength to overthrow. Dogma, rigidly prescribed by tradition, stiffens into formalism. Linguistic categories make up a system of surviving dogma—dogma of the unconscious. They are often but half real as concepts; their life tends ever to languish away into form for form's sake.

[Footnote 63: Hence, "the square root of 4 is 2," precisely as "my uncle is here now." There are many "primitive" languages that are more philosophical 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 mechanical operation of phonetic processes, which may bring about formal distinctions that have not and never had a corresponding functional distinction. Much of the irregularity and general formal complexity of our declensional and conjugational systems is due to this process. The plural of hat is hats, the plural of self is selves. In the former 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 that originally stood for fairly distinct concepts—as we saw was presumably the case with such parallel forms as drove and worked—but a merely mechanical manifolding of the same formal element without a corresponding growth of a new concept. This type of form development, therefore, while of the greatest interest for the general history of language, does not directly concern us now in our effort to understand the nature of grammatical concepts and their tendency to degenerate into purely formal counters.

We may now conveniently revise our first classification of concepts as expressed 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 contrast necessarily implied in defining one concept as against another. "Man" and "white" possess an inherent relation to "woman" and "black," but it is a relation of conceptual content only and is of no direct interest to grammar.]

[Footnote 65: Thus, the -er of farmer may he defined as indicating that particular substantive concept (object or thing) that serves as the habitual subject of the particular verb to which it is affixed. This relation of "subject" (a farmer farms) is inherent in and specific to the word; it does not exist for the sentence as a whole. In the same way the -ling of duckling defines a specific relation of attribution that concerns only the radical element, not the sentence.]

The nature of these four classes of concepts as regards their concreteness or their power to express syntactic relations may be thus symbolized: 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 of analysis difficult problems frequently arise and we may well be in doubt as to how to group a given set of concepts. This is particularly apt to be the case in exotic languages, where we may be quite sure of the analysis of the words in a sentence and yet not succeed in acquiring that inner "feel" of its structure that enables us to tell infallibly what is "material content" and what is "relation." Concepts of class I are essential to all speech, also concepts of class IV. Concepts II and III are both common, but not essential; particularly group III, which represents, in effect, a psychological and formal confusion of types II and IV or of types I and IV, is an avoidable class of concepts. Logically there is an impassable gulf between I and IV, but the illogical, metaphorical genius of speech has wilfully spanned the gulf and set up a continuous gamut of concepts and forms that leads imperceptibly from the crudest of materialities ("house" or "John Smith") to the most subtle of relations. It is particularly significant that the unanalyzable independent word belongs in most cases to either group I or group IV, rather less commonly to II or III. It is possible for a concrete concept, represented by a simple word, to lose its material significance entirely and pass over directly into the relational sphere without at the same time losing its independence as a word. This happens, for instance, in Chinese and Cambodgian when the verb "give" is used in an abstract sense as a mere symbol of the "indirect objective" relation (e.g., Cambodgian "We make story this give all 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 between groups I and II and I and III, as well as of the less radical one between II and III. To the first of these transitions belongs that whole class of examples in which the independent word, after passing through the preliminary stage of functioning as the secondary or qualifying element in a compound, ends up by being a derivational affix pure and simple, yet without losing the memory of its former independence. Such an element and concept is the full of teaspoonfull, which hovers psychologically 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 the primary concreteness is no longer felt. In general, the more highly synthetic our linguistic type, the more difficult and even arbitrary it becomes to distinguish groups I and II.

Not only is there a gradual loss of the concrete as we pass through from group I to group IV, there is also a constant fading away of the feeling of sensible reality within the main groups of linguistic concepts themselves. In many languages it becomes almost imperative, therefore, to make various sub-classifications, to segregate, for instance, the more concrete from the more abstract concepts of group II. Yet we must always beware of reading into such abstracter groups that purely formal, relational feeling that we can hardly help associating with certain of the abstracter concepts which, with us, fall in group III, unless, indeed, there is clear evidence to warrant such a reading in. An example or two should make clear these all-important distinctions.[66] In Nootka we have an unusually large number of derivational affixes (expressing concepts of group II). Some of these are quite material in content (e.g., "in the house," "to dream of"), others, like an element denoting plurality and a diminutive affix, are far more abstract in content. The former type are more closely welded with the radical element than the latter, which can only be suffixed to formations that have the value of complete words. If, therefore, I wish to say "the small fires in the house"—and I can do this in one word—I must form the word "fire-in-the-house," to which elements corresponding to "small," our plural, and "the" are appended. The element indicating the definiteness of reference that is implied in our "the" comes at the very end of the word. So far, so good. "Fire-in-the-house-the" is an intelligible correlate of our "the house-fire."[67] But is the Nootka correlate of "the small fires in the house" the true equivalent of an English "the house-firelets"?[68] By no means. First of all, the plural element precedes the diminutive in Nootka: "fire-in-the-house-plural-small-the," in other words "the house-fires-let," which at once reveals the important fact that the plural concept is not as abstractly, as relationally, felt as in English. A more adequate rendering would be "the house-fire-several-let," in which, however, "several" is too gross a word, "-let" too choice an element ("small" again is too gross). In truth we cannot carry over into English the inherent feeling of the Nootka word, which seems to hover somewhere between "the house-firelets" and "the house-fire-several-small." But what more than anything else cuts 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 corresponds or refers to anything else in the sentence. In English "the house-firelets burn" (not "burns"), in Nootka neither verb, nor adjective, nor anything else in the proposition is in the least concerned with the plurality or the diminutiveness of the fire. Hence, while Nootka recognizes a cleavage between concrete and less concrete concepts within group II, the less concrete do not transcend the group and lead us into that abstracter air into which our plural -s carries us. But at any rate, the reader may object, it is something that the Nootka plural affix is set apart from the concreter group of affixes; and may not the Nootka diminutive have a slenderer, a more elusive content 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 a given grammatical element that has so often led students to misunderstand the nature of languages profoundly alien to their own. Not everything that calls itself "tense" or "mode" or "number" or "gender" or "person" is genuinely comparable to what we mean by these terms in Latin or French.]

[Footnote 67: Suffixed articles occur also in Danish and Swedish and in numerous other languages. The Nootka element for "in the house" differs from our "house-" in that it is suffixed and cannot occur as an independent 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 a feeling-element, an element of nuance, than our -ling. This is shown by the fact that it may be used with verbs as well as with nouns. In speaking to a child, one is likely to add the diminutive to any word in the sentence, regardless of whether there is an inherent diminutive meaning in the word or not.]

Can such a concept as that of plurality ever be classified with the more material concepts of group II? Indeed it can be. In Yana the third person of the verb makes no formal distinction between singular and plural. Nevertheless the plural concept can be, and nearly always is, expressed by the suffixing of an element (-ba-) to the radical element of 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 that the plural affix immediately follows the radical element (ya-), disconnecting it from the local element (-hau-). It needs no labored argument to prove that the concept of plurality is here hardly less concrete than that of location "in the east," and that the Yana form corresponds in feeling not so much to our "They burn in the east" (ardunt oriente) as to a "Burn-several-east-s, it plurally burns in the east," an expression which we cannot adequately assimilate for lack of 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 as an utterly material idea, one that would make of "books" a "plural book," in which the "plural," like the "white" of "white book," falls contentedly into group I? Our "many books" and "several books" are obviously 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 plural concept would still not emerge as clearly as it should for our argument; "many" and "several" are contaminated by certain notions of quantity or scale that are not essential to the idea of plurality itself. We must turn to central and eastern Asia for the type of expression we are seeking. 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 of plurality.[72] If the fact is worth expressing, however, I can say nga-s mi rnams mthong "by me man plural see," where rnams is the perfect conceptual analogue of -s in books, divested of all relational strings. Rnams follows its noun as would any other attributive word—"man plural" (whether two or a million) like "man white." No need to bother about his plurality any more than about his whiteness 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 no commitment on the score of quantity ("a few, several, many").]

What is true of the idea of plurality is naturally just as true of a great many other concepts. They do not necessarily belong where we who speak English are in the habit of putting them. They may be shifted towards I or towards IV, the two poles of linguistic expression. Nor dare we look down on the Nootka Indian and the Tibetan for their material attitude towards a concept which to us is abstract and relational, lest we invite the reproaches of the Frenchman who feels a subtlety of relation in femme blanche and homme blanc that he misses in the coarser-grained white woman and white man. But the Bantu Negro, were he a philosopher, might go further and find it strange that we put in group II a category, the diminutive, which he strongly feels to belong to group III and which he uses, along with a number of other classificatory concepts,[73] to relate his subjects and objects, attributes and predicates, as a Russian or a German handles his genders and, 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 a philosophical analysis of experience that we cannot say in advance just where to put a given concept. We must dispense, in other words, with a well-ordered classification of categories. What boots it to put tense and mode here or number there when the next language one handles puts tense 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 this kind from a general inventory of the types of concepts generally found in groups II, III, and IV. There are too many possibilities. It would be interesting to show what are the most typical noun-forming and verb-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 (singular and plural; singular, dual, and plural; singular, dual, trial, and plural; single, distributive, and collective); what tense distinctions may be made in verb or noun (the "past," for instance, may be an indefinite past, immediate, remote, mythical, completed, prior); how delicately 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 may be recognized (indicative, imperative, potential, dubitative, optative, negative, and a host of others[75]); what distinctions of person are possible (is "we," for instance, conceived of as a plurality of "I" or is it as distinct from "I" as either is from "you" or "he"?—both attitudes are illustrated in language; moreover, does "we" include you to whom I speak or not?—"inclusive" and "exclusive" forms); what may be the general scheme of orientation, the so-called demonstrative categories ("this" and "that" in an endless procession of nuances);[76] how frequently the form expresses the source or nature of the speaker's knowledge (known by actual experience, by hearsay,[77] by inference); how the syntactic relations may be expressed in the noun (subjective and objective; agentive, instrumental, and person affected;[78] various types of "genitive" and indirect relations) and, correspondingly, in the verb (active and passive; active and static; transitive and intransitive; impersonal, reflexive, reciprocal, indefinite as to object, and many other special limitations on the starting-point and end-point of the flow of activity). These details, important as many of them are to an understanding of the "inner form" of language, yield in general significance to the more radical group-distinctions that we have set up. It is enough for the general reader to feel that language struggles towards two poles of linguistic expression—material content and relation—and that these poles tend to be connected by a long series of transitional concepts.

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

[Footnote 75: By "modalities" I do not mean the matter of fact statement, say, of negation or uncertainty as such, rather their implication in terms of form. There are languages, for instance, which have as elaborate an apparatus of negative forms for the verb as Greek has of the optative or wish-modality.]

[Footnote 76: Compare page 97.]

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

[Footnote 77: It is because of this classification of experience that in many languages the verb forms which are proper, say, to a mythical narration differ from those commonly used in daily intercourse. We leave these shades to the context or content ourselves with a more explicit and roundabout mode of expression, e.g., "He is dead, as I happen to know," "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 kill him," but "he kills me." Yet me of the last example is at least as close psychologically to I of "I sleep" as is the latter to I of "I kill him." It is only by form that we can classify the "I" notion of "I sleep" as that of an acting subject. Properly speaking, I am handled by forces beyond my control when I sleep just as truly as when some one is killing me. Numerous languages differentiate clearly between active subject and static subject (I go and I kill him as distinct from I sleep, I am good, I am killed) or between transitive subject and intransitive subject (I kill him as distinct from I sleep, I am good, I am killed, I go). The intransitive or static subjects may or may not be identical with the object of the transitive verb.]

In dealing with words and their varying forms we have had to anticipate much that concerns the sentence as a whole. Every language has its special method or methods of binding words into a larger unity. The importance of these methods is apt to vary with the complexity of the individual word. The more synthetic the language, in other words, the more clearly the status of each word in the sentence is indicated by its own resources, the less need is there for looking beyond the word to the sentence as a whole. The Latin agit "(he) acts" needs no outside help to establish its place in a proposition. Whether I say agit dominus "the master acts" or sic femina agit "thus the woman acts," the net result 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 only be conceived as a statement of activity carried out by a person (or thing) other than you or me. It is not so with such a word as the English act. Act is a syntactic waif until we have defined its status in a proposition—one thing in "they act abominably," quite another in "that was a kindly act." The Latin sentence speaks with the assurance of its individual members, the English word needs the prompting of its fellows. Roughly speaking, to be sure. And yet to say that a sufficiently elaborate word-structure compensates for external syntactic methods is perilously close to begging the question. The elements of the word are related to each other in a specific way and follow each other in a rigorously determined sequence. This is tantamount to saying that a word which consists of more than a radical element is a crystallization of a sentence or of some portion of a sentence, that a form like agit is roughly the psychological[79] equivalent of a form like age is "act he." Breaking down, then, the wall that separates word and sentence, we may ask: What, at last analysis, are the fundamental methods of relating word to word and element to element, in short, of passing from the isolated notions symbolized by each word and by each element to the unified proposition that 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 most fundamental and the most powerful of all relating methods is the method of 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 person or object, setting down its symbol—dog; finally, of a third concrete idea, say an action, setting down its symbol—run. It is hardly possible to set down these three symbols—red dog run—without relating them in some way, for example (the) red dog run(s). I am far from wishing to state that the proposition has always grown up in this analytic manner, merely that the very process of juxtaposing concept to concept, symbol to symbol, forces some kind of relational "feeling," if nothing else, upon us. To certain syntactic adhesions we are very sensitive, for example, to the attributive relation of quality (red dog) or the subjective relation (dog run) or the objective relation (kill dog), to others we are more indifferent, for example, to the attributive relation of circumstance (to-day red dog run or red dog to-day run or red dog run to-day, all of which are equivalent propositions or propositions in embryo). Words and elements, then, once they are listed in a certain order, tend not only to establish some kind of relation among themselves but are attracted to each other in greater or in less degree. It is presumably this very greater or less that ultimately leads to those firmly solidified groups of elements (radical element or elements plus one or more grammatical elements) that we have studied as complex words. They are in all likelihood nothing but sequences that have shrunk together and away from other sequences or isolated elements in the flow of speech. While they are fully alive, in other words, while they are functional at every point, they can keep themselves at a psychological distance from their neighbors. As they gradually lose much of their life, they fall back into the embrace of the sentence as a whole and the sequence of independent words regains the importance it had in part transferred to the crystallized groups of elements. Speech is thus constantly tightening and loosening its sequences. In its highly integrated forms (Latin, Eskimo) the "energy" of sequence is largely locked up in complex word formations, it becomes transformed into a kind of potential energy that may not be released for millennia. In its more analytic forms (Chinese, English) this energy is mobile, ready to hand for such service as we demand of it.

There can be little doubt that stress has frequently played a controlling influence in the formation of element-groups or complex words 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 the following verb and lost its independence as a significant element. In the same way French futures of the type irai "(I) shall go" are but the resultants of a coalescence of originally independent words: ir[81] a'i "to-go I-have," under the influence of a unifying accent. But stress has done more than articulate or unify sequences that in their own right imply a syntactic relation. Stress is the most natural means at our disposal to emphasize a linguistic contrast, to indicate the major element in a sequence. Hence we need not be surprised to find that accent too, no less than sequence, may serve as the unaided symbol of certain relations. Such a contrast as that of go' between ("one who goes between") and to go between' may be of quite secondary origin in English, but there is every reason to believe that analogous distinctions have prevailed at all times in linguistic history. A sequence like see' man might imply some type of relation in which see qualifies the following word, hence "a seeing man" or "a seen (or visible) man," or is its predication, hence "the man sees" or "the man is seen," while a sequence like see man' might indicate that the accented word in some way limits the application of the first, say as direct object, hence "to see a man" or "(he) sees the man." Such alternations of relation, as symbolized by varying stresses, are important 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 to go," i.e., "must go."]

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

It is a somewhat venturesome and yet not an altogether unreasonable speculation that sees in word order and stress the primary methods for the expression of all syntactic relations and looks upon the present relational value of specific words and elements as but a secondary condition due to a transfer of values. Thus, we may surmise that the Latin -m of words like feminam, dominum, and civem did not originally[83] denote that "woman," "master," and "citizen" were objectively related to the verb of the proposition but indicated something far more concrete,[84] that the objective relation was merely implied by the position or accent of the word (radical element) immediately preceding the -m, and that gradually, as its more concrete significance faded away, it took over a syntactic function that did not originally belong to it. This sort of evolution by transfer is traceable in many instances. Thus, the of in an English phrase like "the law of the land" is now as colorless in content, as purely a relational indicator as the "genitive" suffix -is in the Latin lex urbis "the law of the city." We know, however, that it was originally an adverb of considerable concreteness of meaning,[85] "away, moving from," and that the syntactic relation was originally expressed by the case form[86] of the second noun. As the case form lost its vitality, the adverb took over its function. If we are actually justified in assuming that the expression of all syntactic relations is ultimately traceable to these two unavoidable, dynamic features of speech—sequence and stress[87]—an interesting thesis results:—All of the actual content of speech, its clusters of vocalic and consonantal sounds, is in origin limited to the concrete; relations were originally not expressed in outward form but were merely implied and articulated with the help of order and rhythm. In other words, relations were intuitively felt and could only "leak out" with the help of dynamic factors that themselves move on an intuitional plane.

[Footnote 83: By "originally" I mean, of course, some time antedating the earliest period of the Indo-European languages that we can get at by comparative 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 been so often evolved in the history of language that we must glance at it for a moment. This is the method of "concord" or of like signaling. It is based on the same principle as the password or label. All persons or objects that answer to the same counter-sign or that bear the same imprint are thereby stamped as somehow related. It makes little difference, once they are so stamped, where they are to be found or how they behave themselves. They are known to belong together. We are familiar with the principle of concord in Latin and Greek. Many of us have been struck by such relentless rhymes as vidi ilium bonum dominum "I saw that good master" or quarum dearum saevarum "of which stern goddesses." Not that sound-echo, whether in the form of rhyme or of alliteration[88] is necessary to concord, though in its most typical and original forms concord is nearly always accompanied by sound repetition. The essence of the principle is simply this, that words (elements) that belong together, particularly if they are syntactic equivalents or are related in like fashion to another word or element, are outwardly marked by the same or functionally equivalent affixes. The application of the principle varies considerably according to the genius of the particular language. In Latin and Greek, for instance, there is concord between noun and qualifying word (adjective or demonstrative) as regards gender, number, and case, between verb and subject only as regards number, and no concord between verb and object.

[Footnote 88: As in Bantu or Chinook.]

In Chinook there is a more far-reaching concord between noun, whether subject or object, and verb. Every noun is classified according to five categories—masculine, feminine, neuter,[89] dual, and plural. "Woman" is feminine, "sand" is neuter, "table" is masculine. If, therefore, I wish to say "The woman put the sand on the table," I must place in the verb certain class or gender prefixes that accord with corresponding noun 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 are expressed as abstract nouns, each with its inherent class-prefix ("much" is neuter or feminine, "large" is masculine) and with a possessive prefix 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 classification of "table" as masculine is thus three times insisted on—in the noun, in the adjective, and in the verb. In the Bantu languages,[90] the principle of concord works very much as in Chinook. In them also nouns are classified into a number of categories and are brought into relation with adjectives, demonstratives, relative pronouns, and verbs by means of prefixed elements that call off the class and make up a complex system of concordances. In such a sentence as "That fierce lion who came here is dead," the class of "lion," which we may call the animal class, would be referred to by concording prefixes no less than six times,—with the demonstrative ("that"), the qualifying adjective, the noun itself, the relative pronoun, the subjective prefix to the verb of the relative clause, and the subjective prefix to the verb of the main clause ("is dead"). We recognize in this insistence on external clarity of reference the same spirit as moves in the more familiar illum bonum dominum.

[Footnote 89: Perhaps better "general." The Chinook "neuter" may refer to persons as well as things and may also be used as a plural. "Masculine" and "feminine," as in German and French, include a great number 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 River valley. It is impressive to observe how the human mind has arrived at the same form of expression in two such historically unconnected regions.]

Psychologically the methods of sequence and accent lie at the opposite pole to that of concord. Where they are all for implication, for subtlety of feeling, concord is impatient of the least ambiguity but must have its well-certificated tags at every turn. Concord tends to dispense with order. In Latin and Chinook the independent words are free in position, less so in Bantu. In both Chinook and Bantu, however, the methods of concord and order are equally important for the differentiation of subject and object, as the classifying verb prefixes refer to subject, object, or indirect object according to the relative position they occupy. These examples again bring home to us the significant fact that at some point or other order asserts itself in every language as the most fundamental of relating principles.

The observant reader has probably been surprised that all this time we have had so little to say of the time-honored "parts of speech." The reason for this is not far to seek. Our conventional classification of words into parts of speech is only a vague, wavering approximation to a consistently worked out inventory of experience. We imagine, to begin with, that all "verbs" are inherently concerned with action as such, that a "noun" is the name of some definite object or personality that can be pictured by the mind, that all qualities are necessarily expressed by a definite group of words to which we may appropriately apply the term "adjective." As soon as we test our vocabulary, we discover that the parts of speech are far from corresponding to so simple an analysis of reality. We say "it is red" and define "red" as a quality-word or adjective. We should consider it strange to think of an equivalent of "is red" in which the whole predication (adjective and verb of being) is conceived of as a verb in precisely the same way in which we think of "extends" or "lies" or "sleeps" as a verb. Yet as soon as we give the "durative" notion of being red an inceptive or transitional turn, we can avoid the parallel form "it becomes red, it turns red" and say "it reddens." No one denies that "reddens" is as good a verb as "sleeps" or even "walks." Yet "it is red" is related to "it reddens" very much as is "he stands" to "he stands up" or "he rises." It is merely a matter of English or of general Indo-European idiom that we cannot say "it reds" in the sense of "it is red." There are hundreds of languages that can. Indeed there are many that can express what we should 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. We speak of "the height of a building" or "the fall of an apple" quite as though these ideas were parallel to "the roof of a building" or "the skin of an apple," forgetting that the nouns (height, fall) have not ceased to indicate a quality and an act when we have made them speak with the accent of mere objects. And just as there are languages that make verbs of the great mass of adjectives, so there are others that make 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 except by 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" and dodge the preposition altogether, giving the verb a nuance that absorbs the idea of local relation carried by the "to." But let us insist on giving independence to this idea of local relation. Must we not then hold to the preposition? No, we can make a noun of it. We can say something like "he reached the proximity of the house" or "he reached the house-locality." Instead of saying "he looked into the glass" we may say "he scrutinized the glass-interior." Such expressions are stilted in English because they do not easily fit into our formal grooves, but in language after language we find that local relations are expressed in just this way. The local relation is nominalized. And so we might go on examining the various parts of speech and showing how they not merely grade into each other but are to an astonishing degree actually convertible into each other. The upshot of such an examination would be to feel convinced that the "part of speech" reflects not so much our intuitive analysis of reality as our ability to compose that reality into a variety of formal patterns. A part of speech outside of the limitations of syntactic form is but a will o' the wisp. For this reason no logical scheme of the parts of speech—their number, nature, and necessary confines—is of the slightest interest to the linguist. Each language has its own scheme. Everything depends on the formal demarcations which it recognizes.

Yet we must not be too destructive. It is well to remember that speech consists of a series of propositions. There must be something to talk about and something must be said about this subject of discourse once it is selected. This distinction is of such fundamental importance that the vast majority of languages have emphasized it by creating some sort of formal barrier between the two terms of the proposition. The subject of discourse is a noun. As the most common subject of discourse is either a person or a thing, the noun clusters about concrete concepts of that order. As the thing predicated of a subject is generally an activity in the widest sense of the word, a passage from one moment of existence to another, the form which has been set aside for the business of predicating, in other words, the verb, clusters about concepts of activity. No language wholly fails to distinguish noun and verb, though in 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 is imperatively required for the life of language.[91]

[Footnote 91: In Yana the noun and the verb are well distinct, though there are certain features that they hold in common which tend to draw them 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. So are the numeral, the interrogative pronoun (e.g., "to be what?"), and certain "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 prepositions are 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 only with single words and with the relations of words in sentences. We have not envisaged whole languages as conforming to this or that general type. Incidentally we have observed that one language runs to tight-knit synthesis where another contents itself with a more analytic, piece-meal handling of its elements, or that in one language syntactic relations appear pure which in another are combined with certain other notions that have something concrete about them, however abstract they may be felt to be in practice. In this way we may have obtained some inkling of what is meant when we speak of the general form of a language. For it must be obvious to any one who has thought about the question at all or who has felt something of the spirit of a foreign language that there is such a thing as a basic plan, a certain cut, to each language. This type or plan or structural "genius" of the language is something much more fundamental, much more pervasive, than any single feature of it that we can mention, nor can we gain an adequate idea of its nature by a mere recital 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 the same horizon that bounds our view, even though the near, familiar landmarks have changed. When we come to English, we seem to notice that the hills have dipped down a little, yet we recognize the general lay of the land. And when we have arrived at Chinese, it is an utterly different sky that is looking down upon us. We can translate these metaphors and say that all languages differ from one another but that certain ones differ far more than others. This is tantamount to saying that it is possible to group them into morphological types.

Strictly speaking, we know in advance that it is impossible to set up a limited number of types that would do full justice to the peculiarities of the thousands of languages and dialects spoken on the surface of the earth. Like all human institutions, speech is too variable and too elusive to be quite safely ticketed. Even if we operate with a minutely subdivided scale of types, we may be quite certain that many of our languages will need trimming before they fit. To get them into the scheme at all it will be necessary to overestimate the significance of this or that feature or to ignore, for the time being, certain contradictions in their mechanism. Does the difficulty of classification prove the uselessness of the task? I do not think so. It would be too easy to relieve ourselves of the burden of constructive thinking and to take the standpoint that each language has its unique history, therefore its unique structure. Such a standpoint expresses only a half truth. Just as similar social, economic, and religious institutions have grown up in different parts of the world from distinct historical antecedents, so also languages, traveling along different roads, have tended to converge toward similar forms. Moreover, the historical study of language has proven to us beyond all doubt that a language changes not only gradually but consistently, that it moves unconsciously from one type towards another, and that analogous trends are observable in remote quarters of the globe. From this it follows that broadly similar morphologies must have been reached by unrelated languages, independently and frequently. In assuming the existence of comparable types, therefore, we are not gainsaying the individuality of all historical processes; we are merely affirming that back of the face of history are powerful drifts that move language, like other social products, to balanced patterns, in other words, to types. As linguists we shall be content to realize that there are these types and that certain processes in the life of language tend to modify them. Why similar types should be formed, just what is the nature of the forces that make them and dissolve them—these questions are more easily asked than answered. Perhaps the psychologists of the future will be able to give us the ultimate reasons for the formation of linguistic types.

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

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

There is a fourth reason why the classification of languages has generally proved a fruitless undertaking. It is probably the most powerful deterrent of all to clear thinking. This is the evolutionary prejudice which instilled itself into the social sciences towards the middle of the last century and which is only now beginning to abate its tyrannical hold on our mind. Intermingled with this scientific prejudice and largely anticipating it was another, a more human one. The vast majority of linguistic theorists themselves spoke languages of a certain type, of which the most fully developed varieties were the Latin and Greek that they had learned in their childhood. It was not difficult for them to be persuaded that these familiar languages represented the "highest" development that speech had yet attained and that all other types were but steps on the way to this beloved "inflective" type. Whatever conformed to the pattern of Sanskrit and Greek and Latin and German was accepted as expressive of the "highest," whatever departed from it was frowned upon as a shortcoming or was at best an interesting aberration.[93] Now any classification that starts with preconceived values or that works up to sentimental satisfactions is self-condemned as unscientific. A linguist that insists on talking about the Latin type of morphology as though it were necessarily the high-water mark of linguistic development is like the zooelogist that sees in the organic world a huge conspiracy to evolve the race-horse or the Jersey cow. Language in its fundamental forms is the symbolic expression of human intuitions. These may shape themselves in a hundred ways, regardless of the material advancement or backwardness of the people that handle the forms, of which, it need hardly be said, they are in the main unconscious. If, therefore, we wish to understand language in its true inwardness we must disabuse our minds of preferred "values"[94] and accustom ourselves to look upon English and Hottentot with the same cool, yet interested, detachment.

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

[Footnote 94: I have in mind valuations of form as such. Whether or not a language has a large and useful vocabulary is another matter. The actual size of a vocabulary at a given time is not a thing of real interest to the linguist, as all languages have the resources at their disposal for the creation of new words, should need for them arise. Furthermore, we are not in the least concerned with whether or not a language is of great practical value or is the medium of a great culture. 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 adopt for our classification? After all that we have said about grammatical form in the preceding chapter, it is clear that we cannot now make the distinction between form languages and formless languages that used to appeal to some of the older writers. Every language can and must express the fundamental syntactic relations even though there is not a single affix to be found in its vocabulary. We conclude that every language is a form language. Aside from the expression of pure relation a language may, of course, be "formless"—formless, that is, in the mechanical and rather superficial sense that it is not encumbered by the use of non-radical elements. The attempt has sometimes been made to formulate a distinction on the basis of "inner form." Chinese, for instance, has no formal elements pure and simple, no "outer form," but it evidences a keen sense of relations, of the difference between subject and object, attribute and predicate, and so on. In other words, it has an "inner form" in the same sense in which Latin possesses it, though it is outwardly "formless" where Latin is outwardly "formal." On the other hand, there are supposed to be languages[95] which have no true grasp of the fundamental relations but content themselves with the more or less minute expression of material ideas, sometimes with an exuberant display of "outer form," leaving the pure relations to be merely inferred from the context. I am strongly inclined to believe that this supposed "inner formlessness" of certain languages is an illusion. It may well be that in these languages the relations are not expressed in as immaterial a way as in Chinese or even as in Latin,[96] or that the principle of order is subject to greater fluctuations than in Chinese, or that a tendency to complex derivations relieves the language of the necessity of expressing certain relations as explicitly as a more analytic language would have them expressed.[97] All this does not mean that the languages in question have not a true feeling for the fundamental relations. We shall therefore not be able to use the notion of "inner formlessness," except in the greatly modified sense that syntactic relations may be fused with notions of another order. To this criterion 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 no means free from an alloy of the concrete.]

[Footnote 97: Very much as an English cod-liver oil dodges to some extent 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 formal processes[98] most typically developed in the language. Those languages that always identify the word with the radical element would be set off as an "isolating" group against such as either affix modifying elements (affixing languages) or possess the power to change the significance of the radical element by internal changes (reduplication; vocalic and consonantal change; changes in quantity, stress, and pitch). The latter type might be not inaptly termed "symbolic" languages.[99] The affixing languages would naturally subdivide themselves into such as are prevailingly prefixing, like Bantu or Tlingit, and such as are mainly or entirely suffixing, like Eskimo or Algonkin or Latin. There are two serious difficulties with this fourfold classification (isolating, prefixing, suffixing, symbolic). In the first place, most languages fall into more than one of these groups. The Semitic languages, for instance, are prefixing, suffixing, and symbolic at one and the same time. In the second place, the classification in its bare form is superficial. It would throw together languages that differ utterly in spirit merely because of a certain external formal resemblance. There is clearly a world of difference between a prefixing language like Cambodgian, which limits itself, so far as its prefixes (and infixes) are concerned, to the expression of derivational concepts, and the Bantu languages, in which the prefixed elements have a far-reaching significance as symbols of syntactic relations. The classification has much greater value if it is 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 and suffixing languages, we shall find that it is of superior interest to make another distinction, one that is based on the relative firmness with which the affixed elements are united with the core of the word.[101]

[Footnote 98: See Chapter IV.]

[Footnote 99: There is probably a real psychological connection between symbolism and such significant alternations as drink, drank, drunk or Chinese mai (with rising tone) "to buy" and mai (with falling tone) "to sell." The unconscious tendency toward symbolism is justly emphasized by recent psychological literature. Personally I feel that the passage from sing to sang has very much the same feeling as the alternation of symbolic colors—e.g., green for safe, red for danger. But we probably differ greatly as to the intensity with which we feel symbolism 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 difference between a prefixing and a suffixing language, I feel that there is more involved in this difference than linguists have generally recognized. It seems to me that there is a rather important psychological distinction between a language that settles the formal status of a radical element before announcing it—and this, in effect, is what such languages as Tlingit and Chinook and Bantu are in the habit of doing—and one that begins with the concrete nucleus of a word and defines the status of this nucleus by successive limitations, each curtailing in some degree the generality of all that precedes. The spirit of the former method has something diagrammatic or architectural about it, the latter is a method of pruning afterthoughts. In the more highly wrought prefixing languages the 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 form of the whole anew. It is so difficult in practice to apply these elusive, yet important, distinctions that an elementary study has no recourse but to ignore them.]

There is another very useful set of distinctions that can be made, but these too must not be applied exclusively, or our classification will again be superficial. I refer to the notions of "analytic," "synthetic," and "polysynthetic." The terms explain themselves. An analytic language is one that either does not combine concepts into single words at all (Chinese) or does so economically (English, French). In an analytic language the sentence is always of prime importance, the word is of minor interest. In a synthetic language (Latin, Arabic, Finnish) the concepts cluster more thickly, the words are more richly chambered, but there is a tendency, on the whole, to keep the range of concrete significance in the single word down to a moderate compass. A polysynthetic language, as its name implies, is more than ordinarily synthetic. The elaboration of the word is extreme. Concepts which we should never dream of treating in a subordinate fashion are symbolized by derivational affixes or "symbolic" changes in the radical element, while the more abstract notions, including the syntactic relations, may also be conveyed by the word. A polysynthetic language illustrates no principles that are not already exemplified in the more familiar synthetic languages. It is related to them very much as a synthetic language is related to our own analytic English.[102] The three terms are purely quantitative—and relative, that is, a language may be "analytic" from one standpoint, "synthetic" from another. I believe the terms are more useful in defining certain drifts than as absolute counters. It is often illuminating to point out that a language has been becoming more and more analytic in the course of its history or that it shows signs of having crystallized from a simple analytic base into a highly synthetic form.[103]

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

[Footnote 103: The former process is demonstrable for English, French, Danish, Tibetan, Chinese, and a host of other languages. The latter tendency may be proven, I believe, for a number of American Indian languages, e.g., Chinook, Navaho. Underneath their present moderately polysynthetic form is discernible an analytic base that in the one case may 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 is a useful, even a necessary, one, but it has been generally obscured by a number of irrelevancies and by the unavailing effort to make the terms cover all languages that are not, like Chinese, of a definitely isolating cast. The meaning that we had best assign to the term "inflective" can be gained by considering very briefly what are some of the basic features of Latin and Greek that have been looked upon as peculiar to the inflective languages. First of all, they are synthetic rather than analytic. This does not help us much. Relatively to many another language that resembles them in broad structural respects, Latin and Greek are not notably synthetic; on the other hand, their modern descendants, Italian and Modern Greek, while far more analytic[104] than they, have not departed so widely in structural outlines as to warrant their being put in a distinct major group. An inflective language, we must 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 so clearly analytic.]

Latin and Greek are mainly affixing in their method, with the emphasis heavily on suffixing. The agglutinative languages are just as typically affixing as they, some among them favoring prefixes, others running to the use of suffixes. Affixing alone does not define inflection. Possibly everything depends on just what kind of affixing we have to deal with. If we compare our English words farmer and goodness with such words as height and depth, we cannot fail to be struck by a notable difference in the affixing technique of the two sets. The -er and -ness are affixed quite mechanically to radical elements which are at the same time independent words (farm, good). They are in no sense independently significant elements, but they convey their meaning (agentive, abstract quality) with unfailing directness. Their use is simple and regular and we should have no difficulty in appending them to any verb or to any adjective, however recent in origin. From a verb to camouflage 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. Functionally they are related to high and deep precisely as is goodness to good, but the degree of coalescence between radical element and affix is greater. Radical element and affix, while measurably distinct, cannot be 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 of affixing as "fusing" and "juxtaposing." The juxtaposing technique we may call an "agglutinative" one, if we like.

Is the fusing technique thereby set off as the essence of inflection? I am afraid that we have not yet reached our goal. If our language were crammed full of coalescences of the type of depth, but if, on the other hand, it used the plural independently of verb concord (e.g., the books falls like the book falls, or the book fall like the books fall), the personal endings independently of tense (e.g., the book fells 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 sees me, or him see the man like the man sees him), we should hesitate to describe it as inflective. The mere fact of fusion does not seem to satisfy us as a clear indication of the inflective process. There are, indeed, a large number of languages that fuse radical element and affix in as complete and intricate a fashion as one could hope to find anywhere without thereby giving signs of that particular kind of formalism that marks off such languages as Latin and Greek as inflective.

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 of spiritualized 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), it is rather the peculiar alternation of the first person singular -a of the perfect with the -o of the present that gives them their inflective cast. Nothing could be more erroneous than to imagine that symbolic changes of the radical element, even for the expression of such abstract concepts as those of number and tense, is always associated with the syntactic peculiarities of an inflective language. If by an "agglutinative" language we mean one that affixes according to the juxtaposing technique, then we can only say that there are hundreds of fusing and symbolic languages—non-agglutinative by definition—that are, for all that, quite alien in spirit to the inflective type of Latin and Greek. We can call such languages inflective, if we like, but we must 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 on line 4081.]

It is necessary to understand that fusion of the radical element and the affix may be taken in a broader psychological sense than I have yet indicated. 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 of plurality, there is little doubt that the fusion of the elements book and -s into the unified word books would be felt as a little less complete than it actually is. One reasons, or feels, unconsciously about the matter somewhat as follows:—If the form pattern represented by the word books is identical, as far as use is concerned, with that of the word oxen, the pluralizing elements -s and -en cannot have quite so definite, quite so autonomous, a value as we might at first be inclined to suppose. They are plural elements only in so far as plurality is predicated of certain selected concepts. The words books and oxen are therefore a little other than mechanical combinations of the symbol of a thing (book, ox) and a clear symbol of plurality. There is a slight psychological uncertainty or haze about the juncture in book-s and ox-en. A little of the force of -s and -en is anticipated by, or appropriated by, the words book and ox themselves, just as the conceptual force of -th in dep-th is appreciably weaker than that of -ness in good-ness in spite of the functional parallelism between depth and goodness. Where there is uncertainty about the juncture, where the affixed element cannot rightly claim to possess its full share of significance, the unity of the complete word is more strongly emphasized. The mind must rest on something. If it cannot linger on the constituent elements, it hastens all the more eagerly to the acceptance of the word as a whole. A word like 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 are mathematically 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 any mystic value in the process of fusion. It is quite likely to have developed as a purely mechanical product of phonetic forces that brought about irregularities of various sorts.]

The psychological distinctness of the affixed elements in an agglutinative term may be even more marked than in the -ness of goodness. To be strictly accurate, the significance of the -ness is not quite as inherently determined, as autonomous, as it might be. It is at the mercy of the preceding radical element to this extent, that it requires to be preceded by a particular type of such element, an adjective. Its own power is thus, in a manner, checked in advance. The fusion here, however, is so vague and elementary, so much a matter of course in the great majority of all cases of affixing, that it is natural to overlook its reality and to emphasize rather the juxtaposing or agglutinative nature of the affixing process. If the -ness could be affixed as an abstractive element to each and every type of radical element, if we could say fightness ("the act or quality of fighting") or waterness ("the quality or state of water") or awayness ("the state of being away") as we can say goodness ("the state of being good"), we should have moved appreciably nearer the agglutinative pole. A language that runs to synthesis of this loose-jointed sort may be looked upon as an example of the ideal agglutinative type, particularly if the concepts expressed by the agglutinated elements are relational or, at the least, belong to the abstracter class of derivational ideas.

Instructive forms may be cited from Nootka. We shall return to our "fire in the house."[107] The Nootka word inikw-ihl "fire in the house" is not as definitely formalized a word as its translation, suggests. The radical element inikw- "fire" is really as much of a verbal as of a nominal term; it may be rendered now by "fire," now by "burn," according to 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 may be definitely nominalized or verbalized by the affixing of elements that are 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, with its indicative suffix, is just as clearly verbal: "it burns in the house." How weak must be the degree of fusion between "fire in the house" and the nominalizing or verbalizing suffix is apparent from the fact that the formally indifferent inikwihl is not an abstraction gained by analysis but a full-fledged word, ready for use in the sentence. The nominalizing -'i and the indicative -ma are not fused form-affixes, they are simply additions of formal import. But we can continue to hold the verbal or nominal nature of inikwihl in abeyance long before we reach the -'i or -ma. We can pluralize it: inikw-ihl-'minih; it is still either "fires in the house" or "burn plurally in the house." We can diminutivize this plural: inikw-ihl-'minih-'is, "little fires in the house" or "burn plurally and slightly in the house." What if we add the preterit tense suffix -it? Is not inikw-ihl-'minih-'is-it necessarily a verb: "several small fires were burning in the house"? It is not. It may still be nominalized; inikwihl'minih'isit-'i means "the former small fires in the house, the little fires that were once burning in the house." It is not an unambiguous verb until it is given a form that excludes every other possibility, as in the indicative inikwihl-minih'isit-a "several small fires were burning in the house." We recognize at once that the elements -ihl, -'minih, -'is, and -it, quite aside from the relatively concrete or abstract nature of their content and aside, further, from the degree of their outer (phonetic) cohesion with the elements that precede them, have a psychological independence that our own affixes never have. They are typically agglutinated elements, though they have no greater external independence, are no more capable of living apart from the radical element to which they are suffixed, than the -ness and goodness or the -s of books. It does not follow that an agglutinative language may not make use of the principle of fusion, both external and psychological, or even of symbolism to a considerable extent. It is a question of tendency. Is the formative slant clearly towards the agglutinative method? Then the language is "agglutinative." As such, it may be prefixing or suffixing, analytic, synthetic, or polysynthetic.