The language of possibility and probability: Effects of negation on meaning

Abstract

Modal adjectives (e.g., "possible" and "probable") have been the subject of much discussion by linguists and logicians. For ordinary speakers, it was found that an important aspect of the meaning of modal adjectives is that they can all be used to qualify the truth of a statement: Subjects sorted modal adjectives according to similarity of meaning and then ordered the same adjectives solely according to their degree of qualification. The sorting data yielded a one-dimensional scaling solution of low stress that correlated highly with the results from the ordering task. Using the same techniques. negation was found to translate a modal adjective down the scale of qualification (e.g., "improbable" is more qualifying than "probable") so that the order of affirmative adjectives with respect to each other is preserved for the corresponding negated adjectives. Negation in this domain is more analogous to a subtractive, rather than multiplicative, process. Also, affixal negation, as in "improbable," was consistently found to be more qualifying than lexical negation, as in "not probable."

Self-assessment of second language ability: the role of response effects

“…As DeLameter (1982) points out, it may be the case that any alternative wording affects meaning, perhaps in unexpected ways. Studies of negation in mirrorimage phrases such as 'quite likely' and 'quite unlikely' (Lichtenstein and Newman, 1967) and in modal adjectives such as 'definite' versus 'indefinite' (Reyna, 1981 ), indicate that negation does not necessarily produce perceived scales where 'quite likely' and 'quite unlikely' are found at opposite ends. Instead, negations seem to act to translate negative items down the scale in a manner that preserves the relationship of the adjectives/expressions in relation to each other.…”

Quantifying Probabilistic Expressions

“…Reyna [14] also studied the effects of negation on probabilities of things happening. Like ours, many of her subjects did not assign the extreme probabilities to expressions implying certainty.…”

“…Most of these 52 expressions have been studied by other investigators. Table 1 displays the average probability to the nearest whole percent for these expressions given in 19 other studies [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] and the current study [20] of science writers. Our references at the end of this paper also include information about sample sizes and kinds of respondents participating in each study.…”

The Effect of Catastrophe Potential on the Interpretation of Numerical Probabilities of the Occurrence of Hazards1

“…These studies suggest that individuals use some unidimensional scale of qualification when asked to interpret probability phrases (cf. Reyna, 1981), which seems to be relatively stable over time (e.g., Beyth-Marom, 1982;Budescu & Wallsten, 1985). On the other hand, individual differences exist, which may stem from the inherent vagueness of verbal probability expressions (Wallsten, Budescu, et al, 1986).…”