ProtoLanguage-1.htm, 1 of 5


"Languages, in spite of their outward differences, are
formed by identical principles."

Winfred P. Lehmann, Syntactic Typology (1978)


I am proposing a monogenetic origin for all human language. The majority of linguists today do not accept this idea.

There is a minority of linguists who believe that all human language did have a monogenetic origin but that because of lexical replacement and radical phonetic transformations, it will never be possible to successfully reconstruct it.

The arguments against both the idea of monogenesis and the possibility of its reconstruction are many.

I will be providing links to counter-arguments which the reader will be able to access through this central essay --- as well as the arguments presented by its opponents --- so that anyone may obtain a balanced view of the question.

The Proto-Language

I. Time of Appearance

    A. circa 100,000 B.P.E. (Cavalli-Sforza 1995)

      1. I attribute the rapid and successful expansion of modern humans (homo sapiens sapiens) from their earliest attested bases in South Africa and Israel to the emergence of language and the value of its employment as an evolutionary advantage.

II. General Characteristics

    A. Phonology

      1. If all human language has a common origin, then it is possible to reconstruct the phonology of the earliest language from general linguistic principles;

        a. the general principles to be applied are:

          1) all (phonological) "systems strive toward perfect symmetry..." (Anttila 1972: 184);

          2) phonological systems tend toward "maximal differentiation" (Anttila 1972: 186);

          3) universal infantile speech development patterns show the primacy of stops before spirants and affricates (Anttila 1972: 112);

            a) bilabials ("labials"), apicals ("dentals"), and dorsals ("velars") are the primary consonantal categories (Anttila 1972: 112);

            b) front, central, and back are the primary vocalic categories (Anttila 1972: 112; 186);

          4) syllables of the form CV are primary (Anttila 1972: 199);

          5) languages attempt to avoid polysemy and homophony (Anttila 1972: 184);

          6) languages avoid unnecessary expenditures of energy;

          7) in infantile speech, consonants and vowels are first produced separately, and only later connected into monosyllables;

            a) since consonants and vowels (really ?V) were first separately produced, the resulting monosyllables had the form C?V (C+?V)

        b. based on A.1.a.3) and A.1.a.3)a), the earliest language contained at least three stops:

          1) P, representing a bilabial stop; T, representing an apical stop; and K, representing a dorsal stop;

            a) from A.1.a.2), we define the earliest consonants as voiceless (surd) so that consonants and vowels are "maximally differentiated" by the feature voice;

            b) from A.1.a.7)a), the earliest three consonants were /p[?]/, /t[?]/, and /k[?]/ when analyzed from their monosyllabic matrices (surd, glottalized, stops);

              1)) the convenience of webreaders, I indicate superscript by square brackets ([ ]);

        c. from A.1.a.3)b), it contained a minimum of three vowels:

          1) E, a front vowel; A, a central vowel; and O, a back vowel

            a) from A.1.a.6), these vowels were low (open) [æ as in Apple; ä as in fAther; å as cAught]

        d. based on A.1.a.3), the stops were followed by spirants and affricates:

          1) this increased the consonantal inventory to p[?], f, p[?]f, t[?], s, t[?]s, k[?], x, k[?]x;

            a) when spirants, and subsequently affricates, were introduced, the earliest C[?]V monosyllables had already been re-analyzed by the earliest speakers as consisting of C[?]+V so that spirants (and later affricates) were not glottalized;

          2) the earlier ?V articulations were re-analyzed by the earliest speakers as consisting of /?/+V so the consonantal inventory was increased by ?, the surd glottal (laryngal) stop;

        e. in addition, the existence of a sonant (voiced) pharyngal spirant ¿ is postulated for the Proto-Language because of its occurrence in some of the world's languages (notably Afrasian and Caucasian). It is extremely difficult to imagine another phoneme from which it could have arisen;

        f. though some languages have no nasals, the majority of the world's languages do;

          1) based on A.1.a.1) and A.1.a.3)a), three nasals are postulated for the Proto-Language: m, n, and /ng/, which, for webpurposes, we indicate as q;

        g. finally, the frequency of /r/ among the world's languages strongly suggests its originality;

    B. Lexicon

      1. based on A.1.a.4), man's earliest language consisted of 45 monosyllables: the fifteen consonants combined with one of the three vowels;

      2. the first area of interest for children is the human body; it was a primary focus of interest for the earliest speakers also; consequently, the earliest primary referents of these 45 monosyllables were bodily parts; though this may seem a naif idea, Thomas R. Damasio, a foremost neurologist, has written: " . . . the body, as represented in the brain, may constitute the indispensable frame of reference for the neural processes that we experience as the mind; . . . our very organism rather than some absolute external reality is used as the ground reference for the constructions we make of the world around us."(1)

      3. some languages (Egyptian, Sumerian) that were recorded very early offer tantalizing hints at the original meanings of these earliest monosyllables after allowances are made for the modifications to the original consonants in the earliest inventory;

        a. an example is the Egyptian sign d, which illustrates fingers and thumb (PL t[?]a, "thumb, side"); or the Sumerian da, which illustrates the right hand and arm, and means "side";

          1) the application of the rebus principle as an explanation for the assignment of phonetic values for signs in ancient writing system is here explicitly rejected;

        b. these provisional assignments of meaning to the early monosyllables can be confirmed by analysis of the constituents of CVC-roots in other languages, which are the result of earlier CV+CV combinations, with the final vowel reduced by expiratory stress-accentuation of the first monosyllable of the combination;

          1) an example for IE is Pokorny's *ba[x]b-, "swell" (PL p[?]a, "buttock"+ p[?]a, "buttock"); or *3. bher-, "incise" (PL p[?]fa, "chin, projection"+ re "finger/toe nail, scratch")

        c. these assignments of meaning to the early monosyllables were not randomly made. Using the terminology established by Mary LeCron Foster in several recent publications, each consonant and vowel was a phememe, and carried a specific if generalized meaning (E, 'away from the reference'; A, 'at the reference', O, 'towards the reference'), which correlates theoretically with the ideas that are currently being developed around the rubric of sound symbolism.
        These phememes were defined originally by the imaginary projection of oral movements of the tongue and other organs to the extra-oral environment, an idea pioneered by Sir Richard Paget. I will be presenting a short essay to explain the theoretical generalized interpretations of these oral movements.

      4. based on extensive analysis of meanings attached to monosyllables, and the analysis of CVC roots into their constituents, the bodily parts associated with each of the forty-five earliest monosyllables can be assigned;

        a. at some very early date, "verbal" associations were made with the monosyllables based on the perceived properties of the referenced bodily parts. For the convenience of the reader, I will list the primary "verbal" associations along with the primary bodily part referent:

          To see the the earliest forty-five monosyllables with inanimate reference, press here.

          1) as one would expect in a language that was ultimately would be composed of 90 monosyllables (see below), each monosyllable had a specific but, by our standards, very wide but integrally related semantic range. Klimov recognizes this characteristic of the "active" type (see below), and cites as examples "blood=(plant-)juice, ear=leaf" (Klimov 1977: 317);

            a) e.g., P[?]FE means "toe" and "root" (vegetable) and "flake" (mineral); P[?]FA means "chin(-bristle)" and "awn" (vegetable) and "rays (around a shining object)" (mineral); and P[?]FO means "leg" and "(tree-)trunk" (vegetable) and "place(-marker)";

              1)) Thus it can be seen that if 200 words of Basic English suffice to constitute a language, the Proto-Language, monosyllabic though it was, could also function fully as a language;

      5. at an early date, simple aspiration, /h/, was phonemicized; and HA was assigned the meaning "air";

        a. it was, of course, inevitable that interest would expand to naming the animate entities in the Proto-Language environment. What generally characterizes animates is air-breathing, and additionally self-initiated (visible) movement; there is a physiological basis for this primary linguistic classification. In an interesting essay based on the deliberations of members of the LOS (Language Origins Society) located at, we read:

          "For the open categories (Nouns and Verbs), the existence of these classes is seen as deriving both from the actual manner of operation of vision - the pattern of movements of the eye (saccades, fixations, tremor and accommodation). These aspects of the motor functioning of the eye already show a clear distinction between visual operations relating to static elements in the visual scene and visual operations involving perception of movement and change in the visual scene."

          1) the association of HA with "air" led to the formation of another set of monosyllables patterned after the inanimate set, characterized by an aspirated articulation in place of glottalization;

          2) since /?/ corresponded to /h/ as laryngal stop to laryngal spirant, a corresponding surd (voiceless) spirant to sonant (voiced) /¿/ was introduced, which we notate as HH for webpurposes. This is the dotted -h- of Arabic, a surd pharyngal spirant;

          3) accordingly, an additional forty-five monosyllables with animate reference were created in the Proto-Language vocabulary; for the reader's convenience, the primary "verbal" associations will also be listed.

          To see the the earliest animate forty-five monosyllables, press here.

          b) If we knew certainly the Urheimat of homo sapiens sapiens, we would be able to improve the animal identifications based on the then local fauna.

          c) From the meanings, it is obvious that phenomena such as "odor", which we would consider inanimate, were considered animate by the speakers of the Proto-Language.

          d) The level of confidence for the correctness of the identifications for the animates is much lower than for the inanimates, which occur in many compounds in which semantic analysis can act as a control. Typically, the animates occur in one or two animal names, for which several identifications are possible; and where semantic analysis is of little assistance. As I investigate new languages, these identifications are being regularly revised.

          e) an associated animate problem is that, in some cases, we have no method to distinguish primary from secondary nominal meanings. Whereas the "verbal" meaning of "fall" for R[H]E seems to be rather firmly established, the nominal "rain" may be primary; or secondary while "panther" (which lurks on branches overhanging game-paths, and "falls" on prey) may be primary.

          f) the sound system reconstructed above will not be familiar to many readers but typologically, it is not as unusual as it might seem at first glance.

          g) My friend, the late Klaus Baer of the Oriental Institute, once reminded me during an early discussion of the Proto-Language, that the correct identification of the earliest phonemes was secondary since they could be regarded as "tokens" around which generally consistent reflexes in various derived languages could be grouped.



return to HOME PAGE ?


the latest revision of this document can be found at

Patrick C. Ryan * 9115 West 34th Street - Little Rock, AR 72204-4441 * (501)227-9947


p. xvi, Descartes' Error — Emotion, Reason, and the Human Brain, Antonio R. Damasio, A Grosset/Putnam Book, 1994, New York, G. P. Putnam's Sons