Search

This study investigates the influence of ( a ) the surface form of complex sentences and ( b ) the appropriateness of the nouns which are subjects of the various clauses in these sentences for their respective verbs upon the comprehensibility of sentences. Base sentences cast as right-branching sentences are much more comprehensible than the same sentences cast in a center-embedded form. Furthermore, an intermediate case is intermediate in comprehensibility, and both the center-embedded and intermediate forms are sensitive to added clauses, while the right-branching form is not. Appropriateness of subject-noun and verb is a poor predictor of comprehensibility. These and earlier results lead to the conclusion that grammatical structure and not semantic relations is the major determiner of comprehensibility.

This study investigates the psychological correlates of linguistic marking in nouns, verbs, and adjectives covertly marked. Marked and unmarked forms are reliably differentiated by Ss naive with respect to the concept of markedness. The criteria used by Ss in such sorts appeared to be based upon an evaluative feature. Unmarked adjectives are reliably classed in a positive category and marked adjectives into a negative category. Unmarked (present tense) verbs and marked (past tense) verbs are reliably classed in positive and negative categories, respectively. Unmarked (singular) nouns are classed into positive categories and marked (plural) nouns into negative categories if the noun is semantically negative in meaning while the reverse is the case for semantically positive nouns. In addition, oppositional verbs not distinguished by criteria of marking were reliably categorized into positive and negative categories.

Most of the single-word responses to twoor three-word compound stimuli were nouns, whether the compounds were of adjectives or nouns. Content coding revealed that the responses to compound adjectives were positive or negative on the semantic feature 'evaluation' depending on whether the stimulus words had been unmarked or marked ones respectively. It is argued that marking and evaluation are correlated because marked words usually describe less usual states and that all associations have as their basis a semantically deep propositional structure in which the stimulus is the predicate and the response the subject.