Fortescue, Michael. 2007. The typological position and theoretical status of polysynthesis. Tidsskrift for Sprogforskning 5 (1): 1-27

[read: 2014-01-18]

複統合的言語を一つのタイプとしてまとめるような構造的特性のセットを特定することができるかについて考察している。「複統合的言語」と呼ばれている言語の間にもかなり大きな差異があることを指摘している。

NOTES:

Goal of the paper (p. 2)

In search of a better way to characterize features that are common to languages displaying the overlapping cluster of polysynthesis.

Subtypes of polysynthetic morphology (p. 2)

a) Pure incorporating type: Chukchi
b) Field-affixing (Lexical affixing) type: Nuuchahnulth (Nootka)
c) Recursive suffixing type: Eskimo

* Recursive suffixing type “is not generally regarded as instantiating canonical incorporation since words in such languages, however long, may as a rule only contain one lexical morpheme”

Trait cluster that create the appearance of a distinct polysynthetic type (p. 2)

(a) Noun/adjective incorporation.
(b) A large inventory of bound morphemes (but restricted number of stems).
(c) The verb a minimal clause.
(d) Pronominal markers on verbs (subject/object) and nouns (possessor).
(e) Adverbial elements integrated into verbs.
(f) Numerous morphological ‘slots’.
(g) Productive morphophonemics and resultant complex allomorphy of bound and free morphemes.
(h) Non-configurational syntax.
(i) Head-marking (or double marking) type of inflection. “

Baker’s definition ‘real’ incorporation (p. 17)

“according to his definition ‘real’ incorporation is syntactic and solely concerns the incorporation of direct object nouns into their verbal heads (Baker 1996: 295)”

Features that suggest newer development of polysynthesis (p. 21)

(a) Lexical sources of derivational affixes transparent.
(b) Residual stress on incorporated or serialized stems.
(c) Strict adhesion to Bybee’s morpheme-ordering generalizations (derivation affixes closer to stem than inflection).
(d) Productivity of incorporation or morphological verb serialization.”

Features that suggest older polysynthesis (p. 21)

(a) Few if any lexical sources of derivational affixes to be found.
(b) No independent stress (or other individualizing prosodic marking) on incorporated morphemes.
(c) Entangled ordering of derivational and inflectional morphemes.
(d) Evidence of successive historical layering of affixes, with fossilization. 

Key characteristic of polysynthetic languages (p. 21)

“polysynthetic languages are different from other languages in the degree to which their predicate-formation rules may extend into and interact with the layered structure of the clause as a whole”

Common motivation for highly synthetic structure (p. 22)

“elaborate derivational potential of their verbal morphology. Derivational processes in these languages may apply not only to word stems but to whole phrases treated as stems despite any external manifestations of their own syntactic dependencies. Inflected verbs are minimal sentences in these languages, thus it is no surprise that the watertight distinction between the syntactic domains of words and sentences is blurred in them. “