 | |  |
Notes
A version of this paper was given at the Medieval Colloquium of the University of Washington, Seattle, in 1979.
Eugene Vinaver, "Principles of Textual Emendation" in Studies in French Language and Mediaeval Literature Presented . . . to M. K. Pope (1939), pp. 351-369.
Eugene Vinaver (ed.), The Works of Sir Thomas Malory (2nd. ed. 1967), 3v., esp. v. 1, chapter IV, "The Method of Editing", pp. c-cxxvi. The point is that the principles described in Vinaver's "Textual Emendation" article—while they draw their examples from Malory—could be applied to virtually any scribal copying, whereas the discussion in the Malory volume has few implications for the editing of other texts.
David C. Fowler, "A New Edition of the B Text of Piers Plowman", The Yearbook of English Studies 7 (1977), 23-42, a review of Piers Plowman: The B Version. . . . An Edition in the Form of Trinity College MS B. 15. 17. . . . ed. George Kane and E. Talbot Donaldson (1975). See esp. Fowler, p. 32, for a defence of "He is worse þan Iudas þat Ʒiueth a iaper siluer" (based on Latin proditor est prelatus cum Iuda) over Kane-Donaldson's harmonising of the alliterative stave "He is Iugged wiþ Iudas þat Ʒyueþ a Iaper siluer". The translation issue is, of course, only one small part of Fowler's case against the Kane-Donaldson editorial method.
Vinaver, Malory, esp. pp. cxiv-cxviii. See, for example, the defence of "holes" (Caxton) over "towyrs" (Winchester) by reference to French "fenestres"; "hors" (Caxton) over "swerde" (Winchester) by reference to French "cheval"; "bed" (Winchester) over "hede" (Caxton) by reference to French "lit", all supported by the stemma:
Kane-Donaldson, Chapter III, "The Archetypal B Manuscript", pp. 70-97. "The strong presumption of the corruptness of the archetypal text of the B version of Piers Plowman is a main instrument for its editors" (p. 97), a position which allows the editors to restore the "compression, pregnancy, technical excellence, and in the end . . . the poetry" which they find lacking in the scribal versions.
This is not to say that useful textual work has not been done on mediaeval translation. For example, Anthony J. Cárdenas has, in a series of recent conference papers, specialised in a bibliographical/codicological investigation of translation, producing much convincing evidence: "Variant Codices and the Reconstruction of Lost Text in the Libro del Saber de Astrologia of Alfonso X, el Sabio", Annual Meeting of the Medieval Institute of the University of Western Michigan, Kalamazoo, 1982; "The Florentine Version of Alfonso X's Libro del Saber de Astrologia: The Case of the Tell-Tale Lacunae". Society for Textual Scholarship Conference, New York, 1983. Similarly, David Yerkes has done much useful work on the textual implications of Old English translation from Latin— see especially his An Old English Thesaurus (1979). Most of the standard editions of major Middle English authors (Chaucer, Gower, Langland, Lydgate etc.) do, of course, contain textual notes where appropriate, citing readings in sources translated by the Middle English text, but as in the case of Vinaver's Malory mentioned above, these rarely have theoretical value and are normally limited to demonstrating the textual transmission of specific readings. See note (17) below on Chaucer's Troilus and Criseyde as an example. Kenneth E. Carpenter's "The Bibliographical Description of Translation", Papers of the Bibliographical Society of America, 76 (1982), 253-271 focuses on the Eighteenth and Nineteenth Centuries and has no clearly textual implications (unlike the bibliographical work of Cárdenas).
See Paul Maas, Textual Criticism, translated by Barbara Flower (1958), Giorgio Pasquali, Storia della tradizione e critica del testo (2nd. ed. 1952). Henri Quentin, Essais de critique textuelle (1926), Vinton A. Dearing, Principles and Practice of Textual Analysis (1974), W. W. Greg, The Calculus of Variants (1927). All of these studies provide theoretical models (especially stemmatic models) for representing the transmission of a text, but while it might be possible to modify, say, Maas' Lachmannian system to respond to the special problems created by translation, such a modification does not seem to have interested these theorists.
All page numbers refer to the critical edition of Trevisa's De Proprietatibus Rerum (1975), for which I edited Book XV. The text is based on BL MS. Add. 27944. Latin readings cited here are from Bodley 749 and the Vatican copy of Georg Hausner's 1485 Strassburg edition. Because there is no critical edition of the Latin text—nor even if there were, could we be sure that it reflected the manuscript Trevisa used for his translation—there is no attempt to 'correct' the Latin readings except where noted.
i.e. collation of all extant manuscripts in the translating and translated languages reveals no variant within each separate language, but the context does perhaps suggest some problem in the textual tradition. In the examples cited in support of MODEL A, there is unanimity in the readings of extant manuscripts of both languages (although not with each other), but since we are postulating a non-extant reading in the textual tradition of the translating language, the evidence derived from context would be available only within that language. As the discussion demonstrates, the degree of doubt created by the context will probably vary greatly.
R. B. McKerrow, Prolegomena for the Oxford Shakespeare. A Study in Editorial Method (1939). McKerrow's rule is that the editor should depart from "the originals . . . only where they appear to be certainly corrupt" (p. 20). This corruption is defined as "any form which, in the light of our knowledge of the language at the time when the text in question was written, was 'impossible', that is, would not have been, in its context, an intelligible word or phrase" (p. 21, italics mine). My point is that none of the forms listed as illustrations of MODEL A could fall clearly under McKerrow's rule (which is questioned by W. W. Greg, The Editorial Problem in Shakespeare, 2nd ed. [1951], p. xxvii) and that even collation—which McKerrow does allow in a later part of the Prolegomena —would fail to identify the dubious reading. Translation, in the case of MODEL A, becomes virtually the sole ground for the "judgment" which McKerrow falls back upon: "Ultimately, however, the decision whether a reading is sound, and therefore to be allowed to stand, or an error of transmission, and therefore, if possible, to be set right, is a matter of the editor's personal judgment" (p. 35).
A. E. Housman, "Preface to Manilius I 1903" in Selected Prose, ed. J. Carter (1961), p. 50. "a conjecture which alters only a single letter may be more improbable palaeographically than one which leaves no letter unaltered". Housman argues that "closeness to a MS" is not one of the "merits essential to a correction". While I would agree with Housman's general principle that the emendation must not depend solely upon palaeographic similarities, it must surely be only logical that, given two parallel textual traditions, an emendation must satisfy palaeographic probability more securely in one tradition than the other if it is to be shown that the error in copying occurred in only one language.
See G. Thomas Tanselle, "External Fact as an Editorial Problem", Studies in Bibliography 32 (1979), 1-47 for a discussion of this problem.
Isidorii Etymologiarvm, ed. W. M. Lindsay (2v. 1911). Bartholomaeus' De Proprietatibus Rerum is to be edited by J. D. Pheifer of Trinity College, Dublin. Books III and IV of the Latin DPR were edited by James M. Long (1979).
Subsequent editors/publishers of Bartholomaeus/Trevisa claimed to have done just that, but the evidence is ambiguous at best. See Berthelet (1535), Bertholomevs de Proprietatibvs Rervm. "This booke . . . is newely printed with many places therein amended by the latyne examplare" and Thomas East (1582) Batman vppon Bartholome, where the changes introduced by Batman are mostly additions (marked as such) prompted by recent scientific and geographical discoveries. See V. M. Parrish, "Batman's Additions from Elyot and Boorde to His English Edition of Bartholomaeus Anglicus" in Studies in Language, Literature and Culture of the Middle Ages and Later, ed. E. B. Atwood and A. A. Hill (1969), pp. 337-346 and D. C. Greetham, "On the Properties of Things: From Patristic Repository to 'Shakespeare's Encyclopaedia'", CUNY English Forum II (1980) forthcoming.
This is not meant to suggest that a work may not be translated twice (Trevisa himself translated works which were then retranslated during the Fifteenth Century— De Regimine Principum, Bodleian Library MS Digby 233, ff. 1-182v, transcript by David C. Fowler, concordance by Kenneth C. Conroy, no edition; retranslated by Thomas Hoccleve, ed. Thomas Wright (Roxburgh Club, 1860), F. J. Furnivall (EETS E.S. lxxii, 1897), and Polychronicon, printed by Caxton (1482), De Worde (1495), Treveris (1527), C. Babington and J. R. Lumby (Rerum Britannicarum Medii Aevi Scriptores vol. 41, Polychronicon Ranulphi Higden Monachi Cestrensis, 1865-86), 9v. Rolls Series, which also contains the anonymous Fifteenth-Century translation), but rather that the two translations are unlikely to share the same bibliographical/textual traditions (i.e. to be regarded as a single composite work, exemplified in the same manuscripts). A. S. G. Edwards, in his chapter "John Trevisa" (Middle English Prose: A Critical Guide to Major Authors and Genres, forthcoming) suggests the possibility of a two-stage translation for the DPR, but this two-fold translation would not necessarily represent MODELS D1 or D2, unless that is, it could be shown that the translator (or subsequent scribes) had translated from two different manuscripts belonging to two different parts of the Latin textual tradition.
Valerie I. J. Flint, "The Historia Regum Britanniae of Geoffrey of Monmouth: Parody and Its Purpose. A Suggestion" Speculum 54 (1979), 447-468.
The text of Chaucer's major source, the Filostrato of Boccaccio, has been printed as an aid to critical commentary on Chaucer's "translation" (e.g. Chaucer Society, ed. W. M. Rosetti, 1873-83, parallel texts), but this and other sources are usually cited in critical editions more in general explanatory notes and only rarely in textual apparatus (see e.g. IV. 57-59 in F. N. Robinson, The Complete Works of Geoffrey Chaucer (2nd. ed. 1957), p. 910).
Representative opinions include M. C. Seymour, "A Note on the Text" in On the Properties of Things. John Trevisa's Translation of Bartholomaeus Anglicus' De Proprietatibus Rerum (2v. 1975), "Trevisa was an intelligent and competent Latinist who generally (and sometimes literally rather than idiomatically) followed the Latin text before him without question" (p. xv); Babington and Lumby, "Trevisa appears to have been puzzled with the Latinity of Higden . . . It must be owned that Trevisa has occasionally fallen into the most ludicrous errors, which a very little 'avisement' might have avoided . . . It is impossible not to perceive that Higden's scholarship is very far superior to that of his translator" (p. lxi); A. C. Cawley, however, finds at most an "occasional" mistranslation ("Relationships of the Trevisa Manuscripts and Caxton's Polychronicon", London Mediaeval Studies i, 1937-48, pp. 464-65); and Kinkade's study of the same work turns up only three errors in translation (B. L. Kinkade, The English Translations of Higden's Polychronicon, Urbana, Illinois, 1934, p. 8). A number of mistranslations are found in H. K. Kim's edition of Trevisa's Gospel of Nichodemus (Ph.D. Diss. University of Washington 1963). See also D. C. Greetham, The Fabulous Geography of John Trevisa's Translation of Bartholomaeus Anglicus' De Proprietatibus Rerum (Ph.D. Diss. CUNY 1976, pp. 175-183) and T. Lawler, "The Properties of John Trevisa's Major Translations", Viater 14 (1983), which appeared while this article was in proof.
| | |