Polysynthetic Languages

Polysynthetic languages are languages that allow the formation of extremely long and complex words that are built up spontaneously out of many smaller parts. One such word can typically be the functional equivalent of an entire sentence in a language like English. For example, a speaker of the Mohawk language might make up the word wahonwatia'tawitsherahetkenhten', and this would immediately be understood by other Mohawk speakers as meaning "She made the thing that one puts on one's body ugly for him." The term polysynthesis was coined in the late 1800s, when linguists began to develop typologies of natural languages based on knowledge of languages from outside Europe and the Middle East. For these early typologists, a synthetic language was one like Latin and Greek, which use affixes to express the structural and meaning relationships among the words in a sentence. A polysynthetic language, then, is one that carries this method of expression to an extreme (Boas 1911; Sapir 1921; the first important discussion of the concept is Humboldt 1836, although he doesn't use this term). Polysynthesis is particularly associated with the languages of North America, Inuit and Aztec being two paradigm-defining cases. Nevertheless, it refers to a structural type of language, not a linguistic area: there are polysynthetic languages spoken in Australia, New Guinea, Siberia, and India, whereas many native American languages are not polysynthetic. The polysynthetic languages probably do not constitute a discrete type; rather there seems to be a continuum of languages determined by how much they rely on complex words to express various linguistic relationships.

The study of polysynthetic languages has been important for several reasons. First, they present an excellent way of exploring the relationships between the different branches of linguistics. In particular, ideas about the connections between SYNTAX and MORPHOLOGY are well studied by looking at these languages, because they seem to use a different division of labor from languages like English, with more burden on morphology and less on syntax. Thus, the study of such languages has led to new proposals about the relationship between these components (e.g., Sadock 1980, 1985; Baker 1988). These languages also raise interesting questions about the LEXICON and its relationship to both syntax and morphology, because it is clear that speakers of a polysynthetic language cannot possibly learn more than a tiny fraction of the expressions that count as words in their language.

However, the first and most important reason for studying polysynthetic languages is that they constitute one of the most extreme and "exotic" classes of language in a linguistic TYPOLOGY. As such, they provide one of the strongest testing grounds for the validity of proposed LINGUISTIC UNIVERSALS. In this way, they become indirectly relevant to questions about the INNATENESS OF LANGUAGE, because that idea implies that there must be many substantive features of natural language that are attested across the whole human species (see also RATIONALISM VS. EMPIRICISM). Finally, polysynthetic languages raise questions about the EVOLUTION OF LANGUAGE: their existence forces one to ask how it is that human linguistic capacities are articulated enough to account for the ease of language acquisition and yet flexible enough to generate languages that are superficially so different.

What linguists think they have learned about such matters from studying polysynthetic languages has varied over time. These languages contributed greatly to the impression of Boas and Sapir that "language is a human activity that varies without assignable limit" (Sapir 1921), leading away from linguistic universals, innateness, and rationalism. However, more recent research has uncovered facts that point toward the opposite conclusion. For example, one common aspect of polysynthesis is noun incorporation, whereby the noun referring to the thing affected by an action is expressed inside the verb, rather than as a separate direct object (see THEMATIC ROLES and GRAMMATICAL RELATIONS). Thus, in Mohawk one can say wa'kana'tara kwetare' 'It cut the bread,' a single word that contains both na'tar 'bread' and kwetar 'cut'. This is unlike English, where one cannot naturally say It bread-cut. However, there is a point of similarity as well. English does allow affected objects and verbs to be compounded in other environments: one can refer to a long, serrated knife as a bread-cutter, for example. Significantly, neither language allows a noun that refers to the cause of the event to be inside the verb. Thus, in Mohawk one cannot say wawasharakwetare' (containing ashar 'knife') for 'The knife cut it'; neither in English could one call a sliced-up loaf a knife-cuttee. Moreover, there is evidence that the verb and the affected object form a relatively tight unit in the syntax of English (the verb phrase) to the exclusion of the causer of the event (see X-BAR THEORY). Collating facts like these, one finds a true universal: affected objects form tighter constructions with verbs than causers do. This universal property then manifests itself in different ways in English syntax, English morphology, and Mohawk syntax, because of differences in whether HEAD MOVEMENT takes place (Baker 1988).

Another property of polysynthetic languages is that their verbs contain elements that indicate the person, number, and gender of both the subject and object. As a result, the verb, subject, and object can appear in any imaginable word order, in addition to the subject-verb-object order that is required in English. The subject and the object can also be left out entirely. (Languages with these properties are called nonconfigurational; Hale 1983.) All this contributes to the impression that these languages have no syntactic structure to speak of -- in contrast to English. The view changes, however, once one realizes that the elements in the verb are really the equivalent of English pronouns (Jelinek 1984; Van Valin 1985; Bresnan and Mchombo 1987; Mithun 1987). Thus, the Mohawk sentence Sak wahahninu' atyatawi is not best compared to English "Sak bought a dress," but rather to a colloquial dislocated structure "Sak, he finally bought it, the dress." Such structures allow some freedom of word order ("That dress, he finally bought it, Sak") and the omission of noun phrases ("He bought it") even in English. Baker (1996) argues that there are in fact many such abstract similarities between polysynthetic languages and more familiar ones. If this is correct, then polysynthetic languages could actually give some of the most striking evidence in favor of linguistic universals, the innateness of language, and a rationalist view.

Additional links

-- Mark Baker

References

Baker, M. (1988). Incorporation: A Theory of Grammatical Function Changing. Chicago: University of Chicago Press.

Baker, M. (1996). The Polysynthesis Parameter. New York: Oxford University Press.

Boas, F. (1911). Introduction. In F. Boas, Ed., Handbook of American Indian Languages pp. 1-83.

Bresnan, J., and S. Mchombo. (1987). Topic, pronoun, and agreement in Chichewa. Language 63:741-782.

Hale, K. (1983). Warlpiri and the grammar of nonconfigurational languages. Natural Language and Linguistic Theory 1:5-49.

Humboldt, W. von (1836). Über die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluss auf die geistige Entwicklung des Menschengeschlechts. Berlin: Königliche Akademie der Wissenschaften.

Jelinek, E. (1984). Empty categories, case, and configurationality. Natural Language and Linguistic Theory 2:39-76.

Mithun, M. (1987). Is basic word order universal? In R. Tomlin, Ed., Coherence and Grounding in Discourse. Amsterdam: Benjamins, pp. 281-328.

Sadock, J. (1980). Noun incorporation in Greenlandic. Language 56:300-319.

Sadock, J. (1985). Autolexical syntax: A proposal for the treatment of noun incorporation and similar phenomena. Natural Language and Linguistic Theory 3:379-440.

Sapir, E. (1921). Language. New York: Harcourt Brace Jovanovich.

Van Valin, R. (1985). Case marking and the structure of the Lakhota clause. In J. Nichols and A. Woodbury, Eds., Grammar Inside and Outside the Clause. Cambridge: Cambridge University Press, pp. 363-413.

Polysynthetic Languages

Additional links

References

Further Readings