CURRENT pedagogical theory holds that language students have most difficulty learning those foreign language elements which differ most from their native language. Contrastive analysis (or contrastive grammar) seeks to catalogue, through the comparative analysis of the native and foreign language systems, the points of difference, so that more effective language-learning materials, based precisely on these learning problems, can be developed. The validity of this procedure is especially evident at the levels of pronunciation, where student mistakes are consistently encountered as predicted, and syntax, where second language grammatical errors often follow the pattern of native language analogies. On the vocabulary level, however, it is at least possible to suggest alternatives to the contrastive procedure. One of the most frequently applied techniques of contrastive analysis is the search for second-language distinctions which are lacking in the native language. On the phonological level the investigation reveals points in the system at which new sound distinctions must be learned. For example, Spanish speakers must learn to distinguish /s/ and /z/ in English, where Spanish has only /s/. It identifies learning problems on the morphosyntactic level by indicating of sentence patterns which correspond to single patterns in the native language. The indicative/subjunctive contrast, vestigial in English but vital in Spanish, is an example. On the lexical level contrastive analysis brings to light what can be labelled pairs in the second language: of words which are (usually) represented by only one word in the mother tongue. Spanish examples are salir/dejar (English leave), ser/estar (be), conocer/saber (know), por/para (for). Teachers of other languages will easily recognize parallel cases, for such present well-known problems-students in practice certainly do confuse these words. Teaching experience and tradition support contrastive analysis for vocabulary learning. The results obtained at the University of Michigan during the experimental development of elementary audio-lingual materials for Spanish1 suggest a break with contrastive vocabulary procedure. No effort was made, in the elaboration of these materials, to apply the contrastive analytic technique on the vocabulary level. First appearance and subsequent drilling of new words depended primarily on structural and frequency (or utility) criteria. All of the more important problem occurred in the materials, but each member of the pair appeared separately, at a point determined by other criteria to be most appropriate. These words, traditionally presented in pairs, were never contrasted, no effort was made to point out the relation between them, and there were no drills designed to prevent their confusion. The fate of these (specifically salir/dejar, conocer/saber, and ser/estar) was in surprising contrast to expectations: Our students' control of the was markedly better than that of the usual first year Spanish students. No confusions were made; the students we questioned were not aware of any problem; they were even surprised to find later that, in translating sentences containing these words, two different words in Spanish were represented by only one in English. Thus, paradoxically, when of words which are known traditionally and shown analytically to be a problem are placed in juxtaposition, explained, contrasted and drilled, students tend to continue confusing them; when they are presented as if no problem existed students have little or no difficulty. How to view the paradox? One solution is to conclude that problem are not a problem after all and therefore contrastive analytic procedure applied in this way to vocabulary learning is incorrect, even harmful. It is possible to marshall certain arguments in support of such a conclusion. Consider first of all the following Spanish sentences:
Read full abstract