THE CHART METHOD IN DEFINITOLOGY Alfred E. Karpovich Born in symbolic logic1 and developed by the American engineer E. Veitch,2 the chart method has subsequently been employed in automatic relay theory3 and eventually extrapolated to linguistics. In the 1960s and 1970s, when the era of structuralism reached Russia—although in the Soviet context it was not destined to develop —the chart method was used in the works of several Russian scholars (e.g., I. Arnold, L. Malakhovsky, N. Romanova and D. Doktorevich4). All of them seem to have followed the same direction in their endeavors to combine some element of symbolic logic with opposition theory as articulated for phonology by N. Trubetskoy and other Prague School structuralists.5 The Russian linguistic analysts of the Leningrad State University were commonly concerned with lexical semantics, the domain of linguistics unrelated to phonology. It was, however, a time of imitation and semasiologists spared no effort to become as rigorous in their methodology as phonologists had become. Some ofthem have obtained interesting data, particularly in the semantic exploration of the English noun, despite hostility of their more conservative colleagues.6 My first attempt to use some elements of symbolic logic and the opposition theory in my monolingual lexicographic research was made in the mid-70s when I was seeking an effective approach to the problem of dictionary definitions. I was interested in knowing what specific factors might determine and motivate the lexicographer 's choice of definition types or, in other words, if the choice is assumed as intuitive, what the items were that constituted and realized lexicographic intuition. In particular, I was concerned with the lexicographer's expe58 Alfred E. Karpovich59 rience, language skills, and linguistic expertise. Although intuition most likely begins in a mysterious way, its explanation must not rest upon mysticism but on the lexicographic matter itself. The following procedure has been constructed upon the afore-mentioned assumptions. For simplification, I worked with such English dictionaries for learners as M. West's or A. S. Hornby's from Great Britain and C. Barnhart's (with Thorndike or Lorge) in the United States.7 In the dictionaries ofthis type, definitions are actually clear-cut sentences containing brief and essential information about entry-words. The results of the research might also hold true for comprehensive dictionaries if certain modifications are introduced. Regardless of diverse individual systems, depending generally on the researcher's approach to the subject, the number of basic techniques involved in the defining operations almost never exceeds four or five. These are (a) transformations ofkernel patterns, viz. ofentry-words, e.g., "understandable—which (that) can be understood" or "comprehensible—that (which) can be understood or comprehended" (with the replacement of one of the structural components); (b) a synonym or a synonymous group, e.g., opulent explained through rich, or magnificent through splendid, excellent; (c) an antonym supplied with the NOT-particle or other negation, e.g., stupid—not clever; and (d) varied combinations of transformational depictions with synonyms, or a synonym with a notantonym , or a transform with a ???-antonym. Intuition seems to be almost the only tool in working out dictionary definitions. At the beginning of any research, intuition is an indispensable tool but later, during the operational procedure, it has to be replaced by some more objective means of a scientific value. There is another goal in view, too: I want to trace the operations of lexicographers in order to check their intuitive solutions. Although only adjectives were chosen for the procedure , I have some grounds to assume that it is quite pos- 60The Chart Method in Definitology sible to perform the same kind of operations on other classes (e.g., nouns and verbs). With the other classes, there would only be a different set ofdifferential features. The English adjective has no number-case-gender paradigms, and it is changed by the degrees of comparison only. Nonetheless, adjectives vary in their stylistic values , their morphological structures, and in their semantic characteristics (qualitative, relative, numerical). As their usage frequencies come into effect, their quantitative differences also become apparent. Thus, despite their poor explicit properties, the adjectives do alter, at least, by four variable characteristics. All operations, provided by the chart...