Edwin den Boer IGEL Conference 1998, Utrecht



THE FREQUENCY OF ORIGINAL METAPHORS IN LITERARY AND NONLITERARY TEXTS



The number of metaphors was measured in 30 short texts, selected at random, representing four genres:

P 10 (lyrical) poems F 8 fragments of prose fiction (short stories) N 5 frontpage newspaper reports S 7 fragments of scientific (psychological) articles


METHOD: the texts were analyzed by myself; originality was measured on the basis of three criteria: imagery value, novelty, and complexity (see handout).



Significant differences between genres: Results of Mann-Whitney-Wilcoxon tests Correlation with Correlation [N > S = more metaphors in news than in science, p < .05] 'Literariness' Significance [P >> F = more metaphors in poetry than in fiction, p < .01] [Spearman rank- order correlation]



VERY CONVENTIONAL METAPHORS (e.g., A full investigation): NOT COUNTED.

MODERATELY CONVENTIONAL METAPHORS (e.g., His scholarship's been cut):

Types: [As predicted, -.07 p = .694

Tokens: no significant differences] -.28 p = .130

MODERATELY ORIGINAL METAPHORS (e.g., The known, blazing like a headlamp):

Types: P >> F P >> S F > S N > S .61 p = .000

Tokens: P >> F P > N P >> S F > S N > S .62 p = .000

VERY ORIGINAL METAPHORS (e.g., Life bleeds language):

Types: P > N P > S F > N F > S .50 p = .002

Tokens: P > N P > S F > N F > S .48 p = .003

TOTAL NUMBER OF METAPHORS:

Types: P > F P >> S .52 p = .001

Tokens: P > F .21 p = .133