BRANCHES OF DIFFERENT SCIENCES often use the same name; thus “morphology” is used to designate both a branch of biology and a branch of glottology (linguistics). In linguistics itself the term “phonetics” is employed in two different senses that may cause a confusion of concepts: on the one hand, it refers to the “phonetics” or “phonology” of an individual or collective language or to “comparative phonetics” of several languages and, on the other hand to anthropophonetics, i.e., experimental, instrumental, or general phonetics, for which special chairs have even been set up at some universities.
Like so-called “experimental psychology,” experimental phonetics is actually a natural science; neither of them deals with truly psychological or linguistic processes, but they examine only those psysiological and other phenomena that underlie and make possible the psychological and linguistic processes.
General or experimental phonetics includes: description of the speech apparatus (pronunciation, phonation); description of the auditory apparatus; description of the cerebral apparatus, the instrument of the linguistic thought processes; description of the medium (air) or natural channel of linguistic communication. The data of anthropophonetics or experimental phonetics are the phenomena which affect our senses in the process of social-linguistic intercourse and which are associated with mental concepts.
All sciences operate with representations and concepts. The world provides data for thought only insofar as these are reworked into representations and concepts and become a part of man’s psychological activity.
Sciences are divided into (1) those which do not conspicuously deal with the reflection of phenomena and their impressions and which include such outstanding products of the human mind as mathematics, logic, and gnoseology, or the theory of knowledge, and (2) those which deal with phenomena that are reflected in the human mind. The latter sciences study: (a) either the psychological surrogates of what is assumed to occur outside ourselves, that is, in nature, including our own body; and (b) what is assumed to occur only in us, that is, the psychological processes in the broadest sense, including individual psychology, social psychology, sociology, etc. However, inasmuch as interpersonal communication is impossible without the mediacy of physical means acting upon our senses, psychology itself must at each step take into account physical and physiological phenomena.
Like Janus, therefore, the life of man (and animal) presents a double face: one turned to the external world, to nature, and the other to the world of the mind, to the person. Halfway between the natural sciences and the so-called humanistic (or we should say animalistic) sciences stand such sciences as biology, anthropology, psychology, sociology, ethnology, and glottology (or linguistics). Some of them deal primarily with the natural element, others with the animalistic element.
The functions, processes, and impressions which, in the field of linguistics, refer to the external world and are transient, constitute the subject matter of the natural sciences, whereas insofar as they are constant and fixed concepts of the mind, they are the subject matter of linguistics proper.
The terms “sounding,” “sound,” “resonance,” etc., designate the transient representations of linguistic thought and, therefore, belong to (natural) science; but insofar as they are representations of linguistic thought based on individual and individual-collective mental processes, they should be replaced by (proper linguistic) terms, such as:
phoneme, the psychological equivalent of physical “sound,” the actual and reproducible phonetic unit of linguistic thought. The phoneme consists, in turn, of constituent elements of which we are not aware during linguistic intercourse but which can be obtained by analysis; they are:
the kineme, the articulatory, phonational element of linguistic thought;
the acousmeme, the simplest psychological element of audition or acoustic perception; and
the kinakeme, the complex representative of both the articulatory (phonational) and auditory elements.
This is the way things stand in psychologically oriented linguistic acoustics. In the graphic-visual sphere, in psychologically interpreted linguistic optics, we need similar terms, such as grapheme, etc., instead of such terms as “letter” and others, in order to designate the rendering of graphemes and their elements that are a part of socio-linguistic behavior.
The mechanical-acoustic and mechanical-optic phenomena that serve to render linguistic thought present neither continuity nor identity, but solely infinite change and diversity. They owe their identity only to the articulatory-auditory and graphic-visual representations that are ingrained in the human mind, although the representations themselves are, by no means, petrified or fixed; they are mobile, variable and accompanied by countless nuances.
The articulatory-auditory and graphic-visual representations exist in linguistic thought only insofar as they are semasiologized and morphologized.
The foregoing remarks lead us to conclude that we must distinguish two separate sciences: (1) a natural science, phonetics (phonology) or anthropophonetics, closely related to mechanics (dynamics, kinetics) and physics (acoustics, optics), and (2) psychophonetics, which is a “humanistic” science closely related to psychology and sociology. Of course, these two sciences are not altogether separated but rather overlap. As a linguistic science psychophonetics studies only that which exists in the form of articulatory-auditory representations.
But even psychophonetics, or phonetics which is related to the collective-individual (i.e., tribal, or national) forms of linguistic thought, must take into account the results of experimental anthropophonetics or natural phonetics, which enable us to discern that which is residually semasiologized (and morphologized) and that which contains the germ of future distinctions among articulatory-acoustic elements and their eventual semasiologizations.
An example of residual distinctions is the difference between the two j’ s, i.e., between j continuing the proto-Indo-European , and the j which stems historically from proto-Slavic d (dj), a difference that Belić established, in my presence, for the Slavic dialects of Istria.
An example of phenomena that point to future distinctions is the historical transition from the distinction of the velars k, g, x depending on the following palatal or nonpalatal vowels into an autonomous distinction independent of the phonetic sequence: k, g, x, and k’, g’, x’, ) č, ž, š.
Just as it is necessary to distinguish phonetics and psychophonetics, sounds and phonemes, etc., so one must keep apart letters and sounds, graphemes and phonemes. There is no direct relationship between the optical impressions of letters and the acoustic impressions of sounds; there exist only associations of written-visual representations with articulatory-acoustic representations, of graphemes with phonemes, and vice versa.
Two types of confusion must then be guarded against: (a) that of letters with sounds, graphemes with phonemes (and their respective constituent elements), and (b) that of phonetic-auditory phenomena with their fixed psychological counterparts. For example, in Polish there is no psychological distinction between the phoneme rendered by the grapheme y and the phoneme rendered by the grapheme i. Instead of two apparently different phonemes y and i, Polish has only one-im (i mutabile).
All psychophonetic and psychographic representations, insofar as they exist in linguistic thinking, are semasiologized and morphologized. Acoustic or optic impressions may exist outside the sphere of semasiologization and morphologization, but they are not a part of linguistic thinking.
With respect to their morphological role, individual phonemes may:
form a part of the syntagm, i.e., of the word as a morphological element of the sentence; e.g., Polish o, a, u, im;
or constitute a morpheme within a word; e.g., a in vod-a, śan—a, bik-a, gad—a; u in stoł-u, ojc-u, pis-u-je. . . ; o in śan-o, žon-o. . . ; e in pol-e, stol-e, ńeś—e. . . ; im in vod-im, m’eƷ-im, ńić-i m, stoj-im, vol-im. . . ; b in lič-b-a, śej-b-a. . . ; n in trud-n-, lič-n-, da-n. . . ; t in b’i-t-, dar-t-, bi-t-. . ;
or enter into the structure of a morpheme as its main semasiologized and morphologized component; e.g., Polish o || a in mog- || mag-, noś-i || naš-, vol-i- || val-. . . .
The morphological structure of words must be distinguished from their articulatory-acoustic structure, as evidenced, for example, by the existence of rhymes.
Semasiologization and morphologization contribute to preservation of phonological distinctions acquired in the history of a national language.
The morphologization of articulatory-acoustic differences may show different tendencies at different periods of the history of a given linguistic community. For example, in Polish the connection between the various endings of the declension and the final consonants of the stem was once different from what it is now: sto-l-e | król-u, muř-e | knř-u, pań-e | koń-u... ; śćań-e, ręc-e | putsim-ń-i, m’eƷ-i, šij-i. In the past the chief distinction was between palatal and nonpalatal consonants; at present it is between consonants which change psychophonetically in the declension vs. consonants which remain unchanged.
Morphologization affects only certain articulatory-auditory distinctions and different ones in different languages; for example, intonation, length (i.e., length or shortness, quantitas temporalis), accent, palatalization or nonpalatalization, the degree of opening of the oral cavity (o || u, a || å, e || ei), the place of articulation (e || a), voicing or voicelessness. . . . Semasiologization, on the other hand, affects all articulatory-auditory distinctions: to || po, kot || pot, do || no, tom || dom, kos || kąs, sam || tam, car || čar. . . .
Semantic differences allow for a different morphological analysis of homonyms: dam-Ø “of ladies” (Ø = zero) | da-m “I give”. . . .
The degree of morphologization affects the varying historical susceptibility of the phonemes of a national language to articulatory-auditory change; for example:
-m > -n when the morphologization of -m is weak, but
-m > -m when the morphologization of -m is strong;
-t > (-t) (facultative t) when the morphologization of -t is weak, but
-t > -t when the morphologization of -t is strong.
The relationship between the individual, psychological and the collective, social element may vary in different phonemes: in some of them there is a preponderance of the individual and changeable, of physiology and phonation (in the phonemes t ..., s..., x...), and in others a preponderance of the collective and general, of acoustics and audition (in the phonemes of nasal resonance, in l, r. . . , in the vowels).
The assumptions of “sound laws” operating “without exceptions,” of gradual, purely physiological change in a definite direction, and of gradual “transition” of some phonetic nuances into others (n1, n2, n3, n4, . . . nn) are not confirmed by the actual facts of language and are in contradiction with its socio-psychological character. <. . .> The first of these assumptions is confirmed neither by scientific phonetics nor by historical psychophonetics. “Soundlaws” would be possible only if we ignored the existence of individuals, collectivity, social life, linguistic intercourse—in brief, the impact of sociology on linguistic behavior.
The physiological-mechanical production of phonemes and their combinations yields various kinds of temporary and transitory adaptation, whereas psychophonetics involves psychological habits (for example, the habit of narrowing vowels before nasal consonants).
Psychophonetic differences and identities exert an influence upon the morphological types. For example, in the Bohinj-Posavian dialects of Slovenia, the merger of l and v into a single phoneme w yields the declension: gwawa, gwale, gwali; mrtw, mrtli instead of the older głava, głave, głavi; mrtv, mrtvi. Compare also the Polish kakao “cocoa” | v kakale.