¹ The Dragon in the Gate: Studies in the Poetry of G.M. Hopkins (Berkeley: Univ. of California Press, 1968), pp. 11–12.

² A Companion to Wittgenstein's Tractatus (Ithaca: Cornell Univ. Press, 1964), p. 144.

³ The peculiarities of Hopkins' style are thoroughly examined by W.H. Gardner in Gerard Manley Hopkins: A Study of Poetic Idiosyncrasy in Relation to Poetic Tradition, 2 vols. (London: Oxford Univ. Press, 1944; rev. ed., 1948) and by W.A.M. Peters, S.J., in Gerard Manley Hopkins: A Critical Essay towards the Understanding of His Poetry (London: Oxford Univ. Press, 1948). Chapter iv of Gardner's first volume enumerates and illustrates such departures from convention as coinings, compounds, phonetic correspondences, and syntactical shifts. Linguists have encountered some difficulty in formulating a precise definition of deviation in poetry, but there is no doubt that deviations do occur and that they are important stylistic components. The concept of deviation employed in this paper is an intuitive one that corresponds to the description offered by Riffaterre: an unexpected construction that disrupts a linguistic pattern, projecting its meaning through an effect of contrast with the expected pattern. Most of Hopkins' deviations belong to Riffaterre's class of “convergences”—conscious, concentrated accumulations of unpredictable constructions deliberately intended to attract attention and representing “extreme awareness in the use of language.” For discussions of deviation, see Michael Riffaterre, “Criteria for Style Analysis,” Word, 15 (1959), 154–74, and “Stylistic Context,” Word, 16 (1960), 207–18; and Samuel R. Levin, “Deviation—Statistical and Determinate—in Poetic Language,” Lingua, 12 (1963), 276–90.

⁴ The Journals and Papers of Gerard Manley Hopkins, ed. Humphry House and Graham Storey (London: Oxford Univ. Press, 1959), p. 129. In this passage Hopkins is translating, and agreeing with, a line from Parmenides that is translated by John Burnet in Early Greek Philosophy as “for it is the same thing that can be thought and that can be.”

⁵ J. Hillis Miller, The Disappearance of God (1963; rpt. New York: Schocken, 1965), pp. 324–57.

⁶ For an earlier brief sketch of this shift, oriented toward English literature, see Earl R. Wasserman, The Subtler Language: Critical Readings of Neoclassic and Romantic Poems (Baltimore: Johns Hopkins Univ. Press, 1959), pp. 10–11.

⁷ Journals and Papers, pp. 5, 11. Similar word sequences and related comments are scattered through the diary entries of 1863 and 1864.

⁸ Stéphane Mallarmé, Œuvres complètes (Paris: Pléiade, 1945), pp. 918–66. Mallarmé also recognized the existence of “mots isolés,” words that share the phonetic characteristics of a word family but do not belong to it.

⁹ See L.S. Dembo, Conceptions of Reality in Modern American Poetry (Berkeley: Univ. of California Press, 1966) for a partial history of this view, and a discussion of its difficulties. Dembo formulates W.C. Williams' view in this way: “If the objectivist poet sees ‘the thing itself with ‘great intensity of perception’ and devoid of all preconception, the poem he writes must be the verbal reflection of the object perceived and hence in itself a kind of object” (p. 45). Further, such poets believed that the poem's “authenticity to the object, perception or experience is proved when it becomes a thing-in-itself with its own life” (p. 49). But, as Dembo observed earlier, such theories threaten to end in subjectivism, for “Anticonceptualism … resolves itself into a philosophical idealism in which the perceiver, through his consciousness, becomes the determining element of the world he perceives” (p. 6). One might add that, if the aspect of the object that the poem shares is no more than its mode of being as a thing-in-itself, this is not a radical variety of mimesis, but a very general one.

¹⁰ The Poems of Gerard Manley Hopkins, ed. W.H. Gardner and N.H. MacKenzie, 4th ed. (London: Oxford Univ. Press, 1970). Subsequent quotations from this volume are documented by page numbers in the text.

¹¹ Modern Poetry and the Idea of Language (New Haven: Yale Univ. Press, 1974), p. 197.

¹² “Lyric and Modernity,” Selected Papers from the English Institute (New York: Columbia Univ. Press, 1970), p. 175.

¹³ The Letters of Gerard Manley Hopkins to Robert Bridges, ed. Claude Colleer Abbott (London: Oxford Univ. Press, 1935), p. 66.

¹⁴ Further Letters of Gerard Manley Hopkins, ed. Claude Colleer Abbott, 2nd ed. (London: Oxford Univ. Press, 1956), p. 373.

¹⁵ Journals and Papers, p. 130. The mingling of Being and not-Being is from Parmenides; the application of it to inscape is, of course, Hopkins' own idea.

¹⁶ “The Dehumanization of Art,” The Dehumanization of Art and Other Writings on Art and Culture (1948; rpt. Garden City, N.Y.: Doubleday, 1956), pp. 9–10.

¹⁷ Journals and Papers, p. 126. Hopkins himself did not attach any esthetic value to deviation or foregrounding. He spoke of his queerness as a “vice” and said that he expected his deviations to become naturalized as the reader became familiar with his poems. In one passage he compares the resistance the critic feels on reading his poems to “bilge-water,” which grows clear after it stands for a long time. See Letters to Bridges, pp. 50–51.

¹⁸ See Roman Jakobson, “Closing Statement: Linguistics and Poetics,” Style in Language, ed. Thomas A. Sebeok (New York: MIT Press, 1960), and “Poetry of Grammar and Grammar of Poetry,” Lingua, 21 (1968), 597–609. I am indebted to J. Hillis Miller for calling my attention to Jakobson's comments on Hopkins.

¹⁹ Gerard Manley Hopkins and the Victorian Temper (New York: Columbia Univ. Press, 1972), pp. 67–75.

²⁰ The Philosophy of Symbolic Forms, trans. Ralph Manheim (New Haven: Yale Univ. Press, 1961), i, 111. I am indebted to Bruns's Modern Poetry and the Idea of Language for a formulation of this role of poetry and for an account of its history. Bruns calls this function of poetry “Orphic” and observes that mimesis is regarded as a lesser function, which is adopted when Orphic beliefs decline.