¹ See Charles F. Hockett, Two models of grammatical description, Word 10.210-34 (1954). This model and all other ‘discovery procedures’ have been criticized recently by Noam Chomsky, Syntactic structures (‘s-Gravenhage, 1957); but see W. Haas’ review of Chomsky's book in Archivum linguisticum 10.50-4 (1958), where it is suggested that ‘Evaluation is clearly relevant to “constructing a grammar”; those procedures of analysis which linguistic theory accredits as yielding the “better” description, will be adopted.‘ A procedure for evaluating grammars may serve to evaluate indirectly the procedures used in constructing those grammars. Furthermore, even an evaluation procedure requires the validation of individual grammars. This validation will presumably require precise definitions of terms like grammatical or contrast, based on some operational test of an informant's behavior. These procedures (like the pair test) may turn out to be convertible into techniques which linguists will use in constructing grammars; see Heles Contreras and Sol Saporta, The validation of a phonological grammar (in preparation). For studies of other portions of Spanish verb morphology, see Donald Leuschel, Towards a procedure for evaluating alternative segmentations into morphs, and Marlene K. Smith, Morpheme alternants in Spanish verb forms (both in preparation).

The author's research has been supported by grants from the National Science Foundation and the Institute of Mental Health. An earlier version of this paper was presented at the meeting of the Linguistic Society of America at Ann Arbor in July 1957.

² The imperative forms, the infinitive, and the present and past participles are excluded as not being relevant to the problem. The abbreviations of the tenses are Pres (present indicative), Prt (preterit), Ftr (future), PSub (present subjunctive), Impf (imperfect), ISub-r (-ra form of the imperfect subjunctive), ISub-s (-se form of the imperfect subjunctive), Cond (conditional).

³ Note the dialectal form /amástes/ with the allomorph /-s/ corresponding to standard Spanish zero. In the text, but not in the tables or in the statements of morphemic structure, braces are used to enclose morphemes; the hyphens in the phonemic transcriptions indicate morph boundaries; single quotes are used to indicate the meanings of morph(eme)s, or semantic components.

⁴ It is irrelevant for the purpose of this paper that some of the tense morphemes may be further segmented. Similarly, a complete semantic analysis might result in the identification of semantic components beyond those suggested here.

⁵ Problems of morphemic analysis, Lg. 23.321-43 (1947).

⁶ In the segmentation proposed here, the stresses are part of the same morphs as the simultaneous vowels. Thus, {am} ‘love’ has alternants /ám-/ and /am-/, {PSub} ‘present subjunctive’ has alternants /-é-/ and /-e-/.

⁷ Eugene A. Nida, The identification of morphemes, Lg. 24.414-41 (1948).