Two experiments were performed in order to determine the extent to which children in the age range of five to 10 years can discriminate between (a) necessarily true vs. contingent propositions and (b) necessarily false vs. contingent propositions. In each study five types of sentence pairs (30 examples of each) were presented and success was assessed by the number of errors made. Children were either taught the basis of the discrimination by the experimenter trial-by-trial p o s t hoc or received no such instruction. Children in the five to six age range appreciated the necessarily true/contingent distinction with teaching, and the necessarily false/contingent distinction without teaching, although ease of discrimination varied strongly with sentence type. The data were interpreted as being contradictory to the Piagetian account of the development of necessity, and as suggesting that linguistic experience may play a more central role in the acquisition of logical reasoning than Piaget proposed.
Necessity in propositions 255[ 991 (intermediate age children) 7.8 years and 8.3 years; (oldest children) 9.5 years and 9.9 years. Each of these six subgroups was divided further into five subgroups which each received a different kind of necessary/contingent discrimination.Assignment to groups was random apart from the constraint that for teaching vs. no teaching comparisons at each age level the two groups of children tested on each of the five types of discrimination were age-matched on the criterion of a child having an equivalent in the other group within at least six months of his or her own age. Frequently age differences were less. Sex differences were balanced across groups.Material. There were 30 examples of each of the five kinds of necessary/contingent distinction employed here. Some examples of each of these five sentence types are presented in Table 1. A basic syntactic equivalence was maintained between the sentences of each pair as well as an equal number of words in almost every case. The five sentence types will now be described.Definitional sentences were those in which one of the sentences in the pair was true by definition (e.g. 'John's dad is a man' vs. 'John's dad is a farmer').Situational sentences were those whose description of a situation had to be true (e.g. 'There are people in the crowd' vs. 'There are children in the crowd').Relational sentences contained two terms which necessarily had to be related in the way expressed (e.g. 'My green shirt is the same colour as my green tie' vs. 'My green shirt is the same material as my green tie').Empty sentences were similar to the definitional ones except that the necessity was less strong because it was logically possible, though unlikely, for the negation to hold (e.g. 'That zebra has stripes' vs. 'That zebra has babies').Negated sentences were those where the predicate was the equivalent in negative form of the adjective qualifying the subject (e.g. 'The fat man is not thin' vs. 'The fat man is not well').The apparatus for the presentation of the se...