¹ Noah Webster, A grammatical institute of the English language (Hartford, 1784): ‘It is very common to hear these phrases “it was me”, “it is him”.’ Joseph Priestley, Rudiments of English grammar² 143 (London, 1798): ‘When the pronoun precedes the verb … it is very common, especially in conversation, to use the nominative case … as, “who is this for”.’ Priestley also discusses the prevalence of It is me. See also Robert Lowth, A short introduction to English grammar²133–6 (London, 1775). For a general discussion see Charles C. Fries, American English grammar 88–96 (New York, 1940), and for references to the literature consult H. L. Mencken, The American language⁴ 447 (New York, 1949). The current situation among educated speakers in England is represented by Randolph Quirk, ‘Relative clauses in educated spoken English’, English studies 38 (1957).

² This notion of grammar is developed by Noam Chomsky, Syntactic structures (‘s-Gravenhage, 1957). A detailed treatment of specific grammatical problems can be found in Robert B. Lees, A grammar of English nominalizations (Bloomington, 1960). For a similar notion of simplicity in description, see Morris Halle, ‘Phonology in generative grammar’, Word 23.55 (1962).

³ The usage in L₁ is that of general school grammar as described for formal writing by H. W. Fowler and F. G. Fowler, The King's English 60–1 (Oxford, 1922). L₂ and L₃ present clear-cut divergencies from the former. Styles that are intermediate between those presented here are also possible. At least one such style does in fact exist between L₁ and L₂. In that intermediate style, the interrogative pronoun does not have case marking, except after prepositions, whereas the relative pronoun has case marking under the same conditions as in L₁. A style is also conceivable which contains case-marked pronouns after the verb be but maintains the case marking of the relative and interrogative pronouns in object function. Such a system, when formulated, would be more complicated than those presented and would necessitate one set of case-marking rules before Wh-attachment (case marking for objects of verbs and prepositions) and another set of case-marking rules after Wh-attachment (case marking for pronouns after be), such that there would be Whom could she see? like She could see him but not *Whom was it? like It was me. The usage in L₄ agrees entirely with that of the dialogue in Nelson Algren, The man with the golden arm (New York, 1949) and The neon wilderness (New York, 1960). This corpus generally supports observations made for vulgar or substandard English by Mencken in The American language and by Albert H. Marckwardt, American English 148 (New York, 1958). Of particular significance for the present study is the absolute consistency, in the dialogues of Algren, of the usage characterized as L₄, to the complete exclusion of features from other styles.

⁴ For the symbolism, see Appendix C.

⁵ The use of THAT as marker of subordination has historical justification. In Middle English, all subordinating conjunctions, relative pronouns, and interrogative pronouns (but only in indirect questions, that is, in subordination) were followed optionally by THAT. Til it was noon they stoden for to see who that ther com—Chaucer, Troilus and Cressyda 5.1114; Oonly the sighte of hire, whom that I serve, … wold han suffised right ynough for me—Knight's Tale 1231; This is the worchynge of the conclusion to knowe yf that eny planete be directe or retrograde—Astrolab 330–7. (The works of Geoffry Chaucer, edited by A. W. Pollard, H. F. Heath, M. H. Liddell, and W. S. McCormick; London, 1913.) See also George O. Curme, Grammar of the English language: 3. Syntax 218 (Boston, 1931), and Otto Jespersen, Modern English grammar 3.116 (Copenhagen, 1954).

⁶ This follows the analysis suggested by Chomsky, Syntactic structures² 69. Elsewhere (69 and 112) he loosens the formulation, attaching Wh to the more general constituent Noun Phrase. (See also Lees, A grammar of English nominalizations 36). Harris treats the attachments of Wh as pro-morphemes, whose full forms appear in the corresponding assertion; see Zellig Harris, ‘Co-occurrence and transformation’, Lg 32.3 (1956). While for the purposes of the present study I allow Wh questions to be based on the personal pronouns, as a more precise analysis I would propose deriving questions (but not relative clauses) from sentences in which the pronoun to be attached to Wh is one of the indefinites: someone, anyone, somebody, something, etc. In this way, certain features of questions, direct and indirect, containing interrogative words (though not of relative clauses) would be accounted for; e.g. Who else left? like Someone else left, Whom did he see of importance like He saw someone of importance. If we consider CASE applicable also to the indefinites (CASE has no phonological form with someone, just as it has none with it), the rest of the analysis need not be changed essentially. Similarly, the relative which, derived from Wh+it(+CASE)-Number, will have a natural basis of differentiation from the interrogative what, which, according to the improved analysis, is derived from Wh+something(+CASE)-Sg.

⁷ Algren, The neon wilderness 153.

⁸ Algren, The man with the golden arm 14, 84.

⁹ Karl Brunner, Die englische Sprache 2.104. (Halle, 1950–1951).