A Reexamination of Some Semantic Feature Hypothesis Predictions About Dimensional Adjective Acquisition

Deseret Language and Linguistic Society Symposium Volume 11 Issue 1 Article 18 2-15-1985 A Reexamination of Some Semantic Feature Hypothesis Predictions About Dimensional Adjective Acquisition Carolyn Matthews Squires Follow this and additional works at: https://scholarsarchive.byu.edu/dlls BYU ScholarsArchive Citation Squires, Carolyn Matthews (1985) "A Reexamination of Some Semantic Feature Hypothesis Predictions About Dimensional Adjective Acquisition," Deseret Language and Linguistic Society Symposium: Vol. 11 : Iss. 1 , Article 18. Available at: https://scholarsarchive.byu.edu/dlls/vol11/iss1/18 This Article is brought to you for free and open access by the Journals at BYU ScholarsArchive. It has been accepted for inclusion in Deseret Language and Linguistic Society Symposium by an authorized editor of BYU ScholarsArchive. For more information, please contact , . 178 A Reexamination of Some Semantic Feature Jlypothesis Predictions f\bout Dimensional f\djective f\cquisition Carolyn Matthews Squires Brigham Young University \Vithin the last fifteen years, much interest and attention have centered around the concept that children develop an adult lexicon by acquiring in their definitions of words semantic features or components progressively more similar to the adult lexicon. E. Clark (1973) synthesized various aspects of lexical development into her Semantic Feature Hypothesis, tracing the development of lexical meaning through stages of overgeneralization, starting when the child's perception of objects may be limited to one or only a few of their semantic features, through a gradual adding of features more and more specific to the meaning of a word. This finally results in the adult lexicon. As part of }ler hypothesis, E. Clark predicted the confusion of antonyms and words that overlap in meaning within a particular semantic field. Based on her own research and others' (Donaldson and Balfour, 1968; Donaldson and Wales, 1970; 11. Clark, 1970; E. Clark, 1971, 1972, 1973), she postulated that relational concepts which are expressed in binary pairs, such as more and less and dimensional adjective pairs, are overgeneralized by the young child to the effective early exclusion of appropriate meaning for the negative or marked memher of the pair. (For discussion of positive/ unmarked and negative/marked pairs, see fl. Clark, 1970; E. Clark, 1973; KIa tzky, et al., 1973.) For example, children between the ages of 3: 0 and 5:0 are reported to fail to differentiate between the word less and the word more. In the studies cited, less was treated as if it were synonymous with more (Donaldson and Balfour, 1968; E. Clark, 1973). It was concluded that the feature of 'amount' was the salient feature, adopted first by children. In other words, some comparative adjectives can be characterized as consisting of a dimensional feature and a polarity feature, and the dimensional features are acquired before the polarity features. lVhen the dimensional feature has been acquired but not the polarity feature, children interpret both members of a pair to refer to the positive pole. (See Clark, 1973; Townsend, 1976.) Clark, based on the evidence for more/less and her own studies of before/after (1971), postulates that children can be expected to have similar difficulty distinguishing between the polar adjectives in such pairs as big-little, high-low, long-short, tall-short, fat-thin, thick-thin and wide-narrow. Children are expected to interpret big-little as 'size,' high-low as 'height,' long-short as 'length' and so on, with the marked or negative member of the pair swallowed up in the unmarked member, at least in younger children, before complete proficiency is achieved. Clark reported that in this semantic field, children demonstrated understanding 179 of the positive-pole uimensional adjectives when they responueu incorrectly to stimuli relating to the negative member of the set. (E. Clark, 1973.) Donaldson and Wales (1970) reported a similar amhiguity, citing a 'consistent anu obviously statistically significant finding that responses to the positive pole are superior to responses to the negative pole (p. 259).' They, therefore, concluued that children clearly operate in terms of the polarities. H. Clark (1970) reporteu simHar results in his stuuies, wi th subjects comprehenuing positive adjectives in comparative sentences more quickly than negative adjectives. He further stated that the child first learns the nominal use of polar adjectives, so that both members of a pair mean 'having extent.' He then acquires the subordinate properties of the 8ntonyms, their contrastive use. The other aspect considered here of the Semantic Feature Hypothesis predicted by E. Clark relates to the order of acquisition of the complete dimensional adjective pairs, from most general to more specific. This consists of an ordering, basically, from big-little through the other pairs. Big-little, she suggests, serves as a cover for the other adjective pairs, particularly in the marked member of a dimensional pair, and is, therefore, acquired early. Her order of acquisition predicts learning of the biglittle pair first, concurrently with the acquisition of the pair's polarity feature, followed by tall-short, high-low and long-short, and finally by wide-Illirrow, thick-thin and deep-shallow, the most difficult to acquire (E. Clark, 1972, 1973). Her predictions were based upon percentages of correct responses to stimuli in each of the aujective pair categories. Support for E. Clark's hypothesis, or parts of it, and reiteration of her predecessors' results were reported by Klatzky, et al.(1973), Layton and Stick (1979), Richards (1979), Bartlett (1976), Eilers, et ale (1974) and Ehri (1976). However, most of these observers could confirm only part of Clark's hypothesis. Bartlett (1976) and Eilers, et a1. (1974), for example, both confirmed Semantic Feature Hypothesis predictions about the order of acquisition for dimensional features: more general features are acquired first. However, neither found evidence to support the notion that marked memhers of dimensional pairs are interpreted as having the meaning of the unmarked members. Richards (1979) also found no asymmetry in comprehension of the polar adjectives used in his 1979 study. Similarly, Townsend (1976) found no evidence for a marking explanation of children's acquisition of dimensional adjectives. He rejects a 'strong form of the marking theory, in which the polarity feature is acquired only once and then applied to all marked/unmarked pairs.' He also rejects a weak form of the marking theory 'in which the polarity feature is acquired separately for each marked/ unmarked adjective pair (p. 392).' The Current Study As certain aspects of Clark's Semantic Feature Hypothesis are shown to be less than universally accepted, the purposes of the current study emerge. This study was designed to determine whether children interpret negative or marked (...truncated)