1. The construction of words by a mixed system of derivation and new formation.
2. The objective relation is treated as a species of possession; and
3. The possessive relation is regarded as the leading and substantial one, and controls the form of expression.
The first of these corresponds to what I should call _polysynthesis_; the others to _incorporation_ in the limited sense of the term.
Some special studies on this subject have been published by M. Lucien Adam, and he claims for them that they have refuted and overturned the thesis of Duponceau, Humboldt, and Steinthal, to the effect that there is a process called _incorporative_ or _polysynthetic_, which can be traced in all American languages, and though not in all points confined to them, may fairly and profitably be taken as characteristic of them, and indicative of the psychological processes which underlie them. This opinion M. Adam speaks of as a “stereotyped phrase which is absolutely false.”[291]
So rude an iconoclasm as this must attract our careful consideration. Let us ask what M. Adam understands by the terms _polysynthesis_ and _incorporation_. To our surprise, we shall find that in two works published in the same year, he advances definitions by no means identical. Thus, in his “Examination of Sixteen American Languages,” he says, “_Polysynthesis_ consists essentially in the affixing of subordinate personal pronouns to the noun, the preposition and the verb.” In his “Study of Six Languages,” he writes: “By _polysynthesis_ I understand the expression in one word of the relations of cause and effect, or of subject and object.”[292]
Certainly these two definitions are not convertible, and we are almost constrained to suspect that the writer who gives them was not clear in his own mind as to the nature of the process. At any rate, they differ widely from the plan or method set forth by Humboldt and Steinthal as characteristic of American languages. M. Adam in showing that polysynthesis in his understanding of the term is not confined to or characteristic of American tongues, missed the point, and fell into an _ignoratio elenchi_.
Equally narrow is his definition of incorporation. He writes, “When the object is intercalated between the subject and the verbal theme, there is _incorporation_.” If this is to be understood as an explanation of the German expression, _Einverleibung_, then it has been pared down until nothing but the stem is left.