 | |  |
Notes
Presenting the work of twelve authors: Cooper, Stephen Crane, John Dewey, Emerson, Hawthorne, Howells, Irving, Melville, Simms, Thoreau, Mark Twain, and Whitman.
A more detailed account of the history of the CEAA can be found in William M. Gibson's "The Center for Editions of American Authors" and in John H. Fisher's "The MLA Editions of Major American Authors" (and the chronology which follows it) in Professional Standards and American Editions: A Response to Edmund Wilson (Modern Language Association, 1969), pp. 1-6, 20-28. A survey of editions in progress and proposed editions at the time of the inception of the CEAA (and based on the 1962 discussions) is provided by William M. Gibson and Edwin H. Cady in "Editions of American Writers, 1963: A Preliminary Survey," PMLA, 78 (September 1963, part 2), 1-8 (reprinted in an MLA pamphlet, The Situation of English, 1963); another essay useful for background relating to the inception of the CEAA is Willard Thorp's "Exodus: Four Decades of American Literary Scholarship," MLQ, 26 (1965), 40-61.
A convenient record of the books and articles related to CEAA editorial theory and practice is provided as an appendix (entitled "Relevant Textual Scholarship") to the CEAA Statement (2nd ed., 1972), pp. 17-25.
The added footnote, enclosed in brackets on p. 382 of The Collected Papers, is the one which attributes the phrase "the tyranny of the copy-text" to Paul Maas. (He used it in his review of Greg's The Editorial Problem in Shakespeare, RES, 20 [1944], 76; Greg's reaction appears on pp. 159-160 of the same volume.) The 1966 text also adds the second "are" to "what readings are possible and what are not" at 381.31; it inserts the comma in "In the folio, revision and reproduction are so blended" at 390.32; and it adds the clause set off by commas in the statement that "the quartos contain, it is generally assumed, only reported texts" at 391.19. Another difference in the 1966 text is an error: in the quotation from McKerrow's Prolegomena at 380.32, "what we call inner harmony" ought to read "what we may call inner harmony," as it did in the original SB printing. As one would expect, the essay has been included in anthologies: Bibliography and Textual Criticism, ed. O M Brack, Jr., and Warner Barnes (1969), pp. 41-58; Art and Error, ed. Ronald Gottesman and Scott Bennett (1970), pp. 17-36. Both of these anthologies reprint the SB text rather than the text from The Collected Papers.
Greg himself calls attention in a footnote (p. 378) to the fact that McKerrow used the word "substantive" to refer to "an edition that is not a reprint of any other," and he adds, "I do not think that there should be any danger of confusion between 'substantive editions' and 'substantive readings.'" Nevertheless, it is unfortunate that the word should be given two special meanings in editorial discourse. The awkwardness of "accidental" (when it is not used as a plural noun, "accidentals") is obvious and is in fact demonstrated by Greg's own prose a few lines after his introduction of the term: "As regards substantive readings . . . they will doubtless sometimes depart from them accidentally and may even . . . do so intentionally: as regards accidentals, they will normally follow their own habits . . ." (pp. 376-377). Furthermore, since both terms are used by grammarians, one might at first suppose that editors, also concerned with language, might use them in the same way; but "substantive" in grammar means "noun" (which is a less inclusive concept than Greg's "substantive"), while "accident" (or "accidence") refers to inflection for case, number, tense, and so on (which is not "accidental" alteration in Greg's sense but rather falls within the scope of "substantive" changes). Perhaps the closest parallel is the use of "substance" and "accident" in philosophy to signify the essential and the nonessential; yet Greg insists, rightly, that his concern is not with a philosophical distinction. (Greg had used the word "accidents" in 1942 in The Editorial Problem in Shakespeare, where one section is entitled, "Note on Accidental Characteristics of the Text," pp. l-lv; but instead of "substantives" the term "essentials of reading" is employed.)
After defining the two terms (p. 376), he says in the body of his text, "The distinction is not arbitrary or theoretical, but has an immediate bearing on textual criticism"; and in a footnote to the definitions he emphasizes, "The distinction I am trying to draw is practical, not philosophic." See also footnote 12 below.
Although Greg does not say so here, McKerrow's attitude doubtless sprang from his overreacting against the abuses of some nineteenth-century editors, who felt free to choose among variant readings without adequate study of the nature and origin of the editions in which those variants appeared. Later Greg does make a similar point in general terms: "The attitude may be explained historically as a natural and largely salutary reaction against the methods of earlier editors. Dissatisfied with the results of eclectic freedom and reliance on personal taste, critics sought to establish some sort of mechanical apparatus for dealing with textual problems . . ." (p. 383). For a development of this point, see Fredson Bowers, "McKerrow's Editorial Principles for Shakespeare Reconsidered," SQ, 6 (1955), 309-324; and "Multiple Authority: New Problems and Concepts of Copy-Text," Library, 5th ser., 27 (1972), esp. 90-91.
Greg did not move all at once to his final position. In "McKerrow's Prolegomena Reconsidered," RES, 17 (1941), 139-149, and, more fully, in the Prolegomena to The Editorial Problem in Shakespeare (1942), he recognized the unnecessary rigidity of McKerrow's insistence on adopting all the substantive variants from an edition which contains some authorial revisions, but he followed McKerrow's inclusion of "wording" as one of the criteria for choosing a copy-text. By the end of the decade, however, in this "Rationale," he had developed his distinction between substantives and accidentals and therefore admitted in a footnote, "There is a good deal in my Prolegomena that I should now express differently, and on this particular point I have definitely changed my opinion. I should now say that the choice of the copy-text depends solely on its formal features (accidentals)" (p. 386). In making revisions in 1950 of The Editorial Problem, he added a new preface repeating this point and referring to the "Rationale" essay.
More than once Greg calls attention to McKerrow's use of the word "reprint" for "critical edition" (e.g., pp. 379, 380). "Reprint," of course, implies complete absence of editorial interference; but while McKerrow expects an editor to use critical judgment in correcting obviously impossible readings he does not conceive of the result as a "critical edition." Greg calls this confusion "symptomatic"—that is, of McKerrow's pervasive reluctance to give rein to individual judgment.
The examples are effective in demonstrating not only undue reliance on the copy-text but also the self-confidence required to alter the copy-text, and in both instances Greg perhaps overstates the obviousness of the emendations he proposes. His arguments for emending the copy-text in each case are persuasive; but the larger argument of his essay does not require him to assert that these emendations are in fact correct but only to show that an editor ought not to be prevented from seriously considering them by too rigid an adherence to the copy-text. It is one thing to say that Boas and Simpson might have adopted his emendations if they had not been under the tyranny of the copy-text, but quite another to imply that they certainly would have done so.
It should be clear that there is no contradiction involved between the assertion that Greg is dealing with a concept and his own repeated emphasis on the practical rather than the theoretical. Obviously, as the word "rationale" in his title suggests, his argument is conceptual and theoretical, since it attempts to formulate a general statement which can be illustrated by reference to specific situations. But the theory itself is proposed as a matter of expediency, as a workable practical solution to a problem, rather than as a philosophic truth. One can say that it is a theory suggesting how best to accommodate one's ignorance but not that it is a theory leading to a reduction of that ignorance.
Library, 5th ser., 27 (1972), 81-115. His later article, "Remarks on Eclectic Texts," in Proof, 4 (1974), furthers the discussion by elaborating upon and providing numerous examples to illustrate the distinction between single-authority and multiple-authority situations. There are, of course, many instances of multiple authority in which the earliest surviving texts (earliest in each line) are not equidistant from the lost common ancestor. But Greg's theory operates in such cases, because—in the absence of contrary evidence—one can presume the text nearest the lost ancestor to be the most reliable in accidentals. At times, as in any other copy-text situation, an editor may have reason to believe that some text other than the nearest one is the most reliable, and he would then select it as copy-text; otherwise he would select the nearest one. Unlike a situation of equidistant radiating texts, the editor in these instances has a presumptive authority to fall back on when there is no other means for reaching a decision. In other words, it is the existence of authoritative texts that are equidistant from a lost ancestor, not simply the existence of texts representing independent lines of descent from that ancestor, which poses a problem for the application of Greg's approach.
Bowers calls it "one of Greg's three criteria for determining the authority of variants that when a choice seems indifferent, the odds are in favor of the specific authority of the original reading" (p. 15). Actually Greg does not make this one of the three criteria (as stated on p. 385 of Collected Papers and summarized above, p. 177, in the sentence beginning "That judgment depends partly") but rather a procedure to follow when use of these criteria proves inconclusive (for if they were not inconclusive, the choice would not be indifferent).
This streamlining of the apparatus extends also to the simplification of the symbols employed, utilizing considerably fewer than were envisaged by McKerrow in his Prolegomena.
Such as that in On Editing Shakespeare and the Elizabethan Dramatists (1955), pp. 71-83; "Old-Spelling Editions of Dramatic Texts," in Studies in Honor of T. W. Baldwin, ed. D. C. Allen (1958), pp. 11-12; and Textual and Literary Criticism (1959), pp. 141-142.
Read before the American Literature section of the South Atlantic Modern Language Association on 22 November 1962 and published in SB, 17 (1964), 223-228.
The term often used in connection with Renaissance texts is "old-spelling"; but "unmodernized" is probably better, since it more clearly suggests that the modernizing of accidentals in general, not just spelling, is the point at issue.
The fact that nineteenth-century accidentals are nearer our own than those of the sixteenth century means that the general reader or the classroom student has less difficulty in using an unmodernized text for this period; but it has nothing to do with the fact that a scholar requires a text representing as accurately as possible the author's own accidentals, regardless of the ease or difficulty with which that text can now be read. There are, in fact, a considerable number of differences between nineteenth- and twentieth-century accidentals, particularly punctuation; but, as Bowers says, "one may flatly assert that any text that is modernized can never pretend to be scholarly, no matter at what audience it is aimed" (p. 223). Obviously, if accidentals form part of an author's expression of meaning, one cannot modernize and still have what the author wrote and meant. One can always argue that the authors themselves would not want their punctuation and spelling to be preserved at the cost of not being read; but such an argument has no bearing on the needs of scholars to have before them, insofar as it is possible, exactly what the author wrote. Modernized editions can then follow, when they seem necessary, though they must inevitably be a compromise. Some of the issues involved in the question of modernizing are discussed in John Russell Brown, "The Rationale of Old-Spelling Editions of the Plays of Shakespeare and His Contemporaries," SB, 13 (1960), 49-67; Arthur Brown, "The Rationale of Old-Spelling Editions . . . A Rejoinder," SB, 13 (1960), 69-76; and Jürgen Schäfer, "The Orthography of Proper Names in Modern-Spelling Editions of Shakespeare," SB, 23 (1970), 1-19. See also footnotes 93, 98, and 99 below.
That is, it must result from editorial decisions and not be simply a reprint of one particular text. It could, of course, in rare instances be such a reprint—but only because the editor judged no emendations to be necessary, not because he was committed to reproducing a single document without alteration.
He continued to demonstrate the bibliographical and textual problems raised by nineteenth-century works in "Old Wine in New Bottles: Problems of Machine Printing," in Editing Nineteenth-Century Texts, ed. John M. Robson (1967), pp. 9-36.
But one which, it seems reasonable to assume, is basic, since it can provide the details necessary for use in preparing other kinds of editions.
The use of the Hinman Collator, developed by Charlton Hinman for the detection of variant formes in the Shakespeare first folio, to make collations of copies of machine-printed books from the same type-setting or plates is another instance of the application to later books of methods conceived in connection with earlier ones. For a list of articles dealing with mechanized collation, see the CEAA Statement (2nd ed., 1972), pp. 19-20.
He also emphasizes, like Greg, the practical side of the point: "This distinction," he says, "is not theory, but fact." Obviously the "fact" is not that the editor can always distinguish correctly between the author's and other's changes but that Hawthorne's manuscripts do exhibit many differences from their first printings and that unauthorized changes are likely to enter a text every time it is set in type.
Bowers, using the example of Hawthorne, speaks of reporting all the editions during the author's lifetime. But since the reason for choosing that period is to cover any editions which might incorporate authorial changes, one can infer that any pirated edition—which can be established as pirated and thus as having no connection with the author—can be excluded from the listing. A collation of such an edition, however, is naturally still called for, so that the editor can be sure that the variant readings in it do not suggest authorial revision in spite of the external evidence.
This point again reflects Greg's rationale in its focus on the accidentals of the copy-text. Once the importance of preserving the accidentals of the copy-text is established, the importance of deciding when to retain, and when to omit, copy-text line-end hyphens becomes evident—as well as the importance of informing the reader in this respect, as in others, of exactly what occurs in the copy-text.
The history of the pamphlet is sketched by Thorpe in his introduction. The MLA's Committee on Research Activities had earlier presented a report (edited by Helmut Rehder) entitled "The Aims, Methods, and Materials of Research in the Modern Languages and Literatures," published in PMLA (67, no. 6 [October 1952], 3-37) and as a pamphlet. The section of that report on "Editing and Textual Criticism" (pp. 15-19), written by Lawton P. G. Peckham, does not mention Greg's rationale and sets forth the idea that "the last edition revised by an author, or published in his lifetime with his consent, is most likely to satisfy literary needs" (p. 16).
The essay appears on pp. 23-42 of the original edition. In 1970 a revised edition was published; Bowers's essay, on pp. 29-54, was enlarged chiefly by the insertion of illustrations from the writings of Stephen Crane (the longest such insertion occurs on pp. 51-52).
In addition to covering, as one would expect, such matters as whether a text is to be critical, whether it is to be modernized, and what role analytical bibliography plays in editing.
Under the revised title Statement of Editorial Principles and Procedures. The revised discussion of proofreading adds a further requirement, based on the experience of several editions: that a final check be made to determine whether printers' errors have entered after the final proofs, by performing a machine collation of the unbound printed gatherings against the last set of proofs.
The seal itself was of course devised as a shorthand way of informing the reader or buyer that certain standards had been met and of promoting a broader general awareness of the need for reliable texts—with the potential result that readers would begin to demand, and publishers to seek, texts which qualified for the seal.
Except for one footnote, which suggests that the "graphic peculiarities of particular texts" should probably not be recorded in the "general apparatus" but "may appropriately form the subject of an appendix" (p. 386). Apparently the "general apparatus" Greg has in mind consists of footnotes; and, since he believes that "in this respect the copy-text is only one among others," he is expressing the view that emendations of accidentals in the copy-text should "probably" not be listed at the foot of the page but rather at the end of the volume—the practice which Bowers adopts in the Dekker. As early as 1760 Edward Capell, in his Prolusions, employed a similar system, in which one category of readings is listed at the foot of the page, and at the end come "all the other rejected readings of the editions made use of" (p. iii). Capell saw the value of making a specific text the "ground-work" of his own and of recording all departures from it as well as variant readings from other editions, so that the reader would have "all the materials that can be procur'd for him," in order to re-examine the editor's decisions.
The only way in which adherence to Greg's theory affects the content of the usual CEAA apparatus is that the historical record of variants normally lists only substantives, not accidentals. There would obviously be no objection to the inclusion of the accidentals as well, but in most cases the number involved would be so great that the effort and expense of listing them would not seem to be justified, in view of the lack of importance attaching to accidentals in later editions under Greg's theory. If someone disagrees with the editor's choice of copy-text, therefore, and wishes to re-edit the work using a different copy-text, he cannot reconstruct the accidentals of that text from the usual CEAA historical collation. Nevertheless, the editor of a CEAA edition explains his choice of copy-text in his textual essay, citing not only external evidence but illustrative readings from the texts; he generally provides enough evidence so that a reader will have an adequate basis for agreeing or disagreeing with his choice. Naturally a person who decides to re-edit the text employing a different copy-text will have to turn to a copy of the edition containing that copy-text; but he should be able to rethink the question of copy-text in the first place on the basis of what is included within the CEAA volume.
Obviously it makes sense to follow established forms whenever there is no particular reason for not following them, so that readers will have fewer adjustments to make as they turn from one edition to another.
For an examination of the variations in apparatus among CEAA volumes and of certain considerations to keep in mind in choosing among them, see G. T. Tanselle, "Some Principles for Editorial Apparatus," SB, 25 (1972), 41-88; and "Editorial Apparatus for Radiating Texts," forthcoming in the Library.
For some discussion, see William H. Gilman, "How Should Journals Be Edited?", Early American Literature, 6 (1971), 73-83.
The CEAA has recently begun to offer a seal for "An Approved Facsimile." The first facsimile to be published under this plan is that of the manuscript of The Red Badge of Courage, edited by Fredson Bowers with extensive introductory material and appendixes and published in two volumes in 1973 by Microcard Editions Books. The next two such facsimiles are to be those of F. Scott Fitzgerald's Ledger and of the manuscript of The Great Gatsby, edited by Matthew J. Bruccoli.
The CEAA seal, reading "An Approved Text," is awarded only to editions which contain these two elements. However, when a CEAA text is leased by a reprinting publisher, the "An Approved Text" seal remains on the new volume, so long as the text is faithfully reproduced, even if the apparatus is not also reprinted. Perhaps logically the original CEAA edition should contain a seal for "An Approved Apparatus" as well as for "An Approved Text," since the "Text" seal, when it appears in the originating edition, covers more than the text. Such a seal—reading "An Approved Apparatus"—already exists for a different purpose: it is available to editors who have gone through precisely the same CEAA editorial process but find that publication of the actual text is not feasible (because of copyright restrictions, for example, or lack of interest on the part of a publisher). In these cases the apparatus is keyed by page-line references to the copy-text edition, and a reader, entering the listed emendations on a copy of that edition, can bring its text into conformity with the critical text established (but not published) by the editor. The first apparatus of the CEAA pattern to be published separately from a text was Matthew J. Bruccoli's "Material for a Centenary Edition of Tender is the Night," SB, 17 (1964), 177-193; the advantages of and appropriate occasions for such an approach are discussed in James B. Meriwether's "A Proposal for a CEAA Edition of William Faulkner," in Editing Twentieth-Century Texts, ed. Francess G. Halpenny (1972), pp. 12-27. Bruccoli has recently prepared the first separate apparatus to receive the CEAA seal and to be issued as an independent publication (The Great Gatsby, University of South Carolina Press); see his discussion of it and the proposed series of which it is to be a part, in "The SCADE Series: Apparatus for Definitive Editions," PBSA, 67 (1973), 431-435.
"Practical Texts and Definitive Editions," delivered on 16 February 1968 and published, along with a paper by Charlton Hinman, in Two Lectures on Editing: Shakespeare and Hawthorne (1969), pp. 21-70. Bowers's essay includes (footnotes 8, 9, 11, 16, 17, on pp. 36-39, 42, 46-48) some comments on Richard H. Fogle's unfavorable review of the Hawthorne edition in American Literary Scholarship: An Annual, 1965, ed. James Woodress (1967), pp. 21-27; Fogle's comparison of the Ohio State edition of The House of the Seven Gables with a classroom edition (Riverside) prompted Bowers's decision to elaborate on the differences between the two kinds of editions: "without the stimulation of his confusion of the true issues it is unlikely that this paper would ever have been written" (p. 38). Bowers has continued to discuss the differences in later articles, such as "The New Look in Editing," South Atlantic Bulletin, 35 (1970), 3-10 (which also comments on Jesse H. Shera's review of the Hawthorne edition in American Notes & Queries, 1 [1962-63], 159-260); and "The Ecology of American Literary Texts," Scholarly Publishing, 4 (1972-73), 133-140. Joseph Katz discusses the shortcomings of certain practical editions in "Practical Editions: A Bad Resource for American Literary Study," Resources for American Literary Study, 3 (1973), 221-229 (which includes references to the Proof articles surveying the practical editions of individual works).
This point has been well put by Leon Howard in his review of the Hawthorne edition in NCF, 22 (1967-68), 191-195, when he remarks that, "even though textual theory might change, the work has been done and the information made available for every serious student of American literature" (p. 193). Of course, new information can turn up later, as Bowers recognizes when he says that "definitive" is "only a comparative term, since we must always believe that from time to time the accumulation of scholarship will enable an editor to improve on the work of his predecessors" ("Old-Spelling Editions of Dramatic Texts" [cited in footnote 18 above], p. 13). Similarly, he begins his discussion of "Established Texts and Definitive Editions," PQ, 41 (1962), 1-17, by noting, "Nothing but confusion can result from the popular assumption that only one form of an established text can ever exist, and hence that a definitive edition of a single form of a literary work is invariably possible."
Although I shall be commenting here principally on essays which take an adverse view of Greg's theory and the CEAA (since they naturally bring into sharpest focus the issues involved), it is clear that some of the reviewers who have written favorable, but often perfunctory, notices of CEAA volumes have no real conception of the aims of those editions. References to some of the more significant reviews are made in the CEAA Statement (2nd ed., 1972), p. 23; a number of other reviews, principally from 1969, are listed (along with other CEAA publicity) in CEAA Newsletter, No. 3 (June 1970), pp. 36-38. Although a few of these reviews, and some others, are referred to in these pages, I do not take up individual reviews in any detail, since the questions they raise usually involve judgment of particular cases rather than general principles and procedures.
In the 26 September installment (pp. 7-10), headed "Their Wedding Journey," he comments not only on this volume of the Indiana Howells edition but also on Typee in the Northwestern-Newberry Melville and The Marble Faun in the Ohio State Hawthorne; on 10 October (pp. 6, 8, 10, 12, 14) he limits himself to "Mark Twain," taking up three volumes of the California edition of Mark Twain papers—Satires and Burlesques, Letters to His Publishers, and Which Was the Dream?
The New York Review of Books for 19 December 1968, pp. 36-38, contained letters from William H. Y. Hackett, Jr., and Theodore Besterman, which Wilson prints in his postscript; letters from George B. Alexander, Ronald Gottesman, and Paul Baender, which Wilson refers to; and letters from Frederick Buechner and Frank J. Donner, which Wilson does not mention (but whose correction of "Albert Payson Terhune's" to "Albert Bigelow Paine's" is incorporated in the text at 189.6).
As Wilson notes in the 1968 pamphlet, the article had originally appeared "in a slightly different form." For pamphlet publication a number of stylistic revisions were made (e.g., "persistent" for "acute" at 179.21 [all page references are to the 1973 volume, as the most accessible text]), some errors were corrected (e.g., the comments on the Constable Melville and the Russell & Russell reprint at 191.6-9), some additions were incorporated (e.g., the parenthetical sentence at 172.8-11), and five footnotes were added (those on pp. 164, 166, 182, and 186, and one not retained in the 1973 book: attached to the sentence ending at 156.4, it read, "These volumes now range in price here from $10 to $14"). Three of those footnotes cite information supplied by correspondents (those on pp. 164 and 166 based on Gottesman's published letter, and that on p. 186 credited to Alexander's published letter). As Wilson points out in his postscript, other revisions were based on comments in letters, particularly Gottesman's (though in describing one of the corrections—"Reedy" to "Rudy" at 164.5—he reverses the two words); but some corrections available to him in letters were not in fact utilized (see footnote 49 below). The 1973 volume incorporates a few more corrections (e.g., "Newberry" for "Newbury" at 163.22) and omits one footnote (as noted above).
Epstein and John Thompson submitted a proposal—for a series of editions of the kind Wilson desired—to various foundations and eventually to the National Endowment; in 1966, at the time of the initial award of $300,000 to the MLA, $50,000 was to be made available for the Wilson plan whenever facilities for administering it were developed (they never were). The New York Review of Books, edited by Epstein's wife, had included, eight months before Wilson's article, another review critical of a CEAA edition—Lewis Mumford's "Emerson Behind Barbed Wire" (18 January 1968, pp. 3-5). Mumford objected to the Harvard edition of The Journals and Miscellaneous Notebooks of Ralph Waldo Emerson because of its inclusion of material discarded by Emerson and its use of editorial symbols within the text to record Emerson's revisions. The Mumford piece provoked considerable correspondence, including a letter from Wilson criticizing the MLA "stupid academic editions" and describing his Pléiade idea. The issue of 14 March 1968 (pp. 35-36) contained letters from (besides Wilson) Lewis Leary, William M. Gibson, and G. S. Rousseau, and a reply from Mumford; another letter, from M. H. Abrams and Morton W. Bloomfield, along with another reply from Mumford, appeared in the issue of 23 May (p. 43). Mumford's review, of course, does not touch on the subject of Greg's rationale, since the choice of copy-text is not an issue in connection with Emerson's manuscript journals.
For example, he approves the inclusion in the Pléiade Proust of "an omitted episode" and the restoration in the Soviet editions of Pushkin and Tolstoy of "cut or altered passages," and he looks forward to a complete edition of Mark Twain's Autobiography, since "we have never had the whole of this work"; yet he believes that one of the pieces included in Which Was the Dream? ("Three Thousand Years among the Microbes") "might well be omitted from the canon" because "it turns out to be disappointing" (p. 178).
Gibson's essay, "The Center for Editions of American Authors" (pp. 1-6), was reprinted from Scholarly Books in America, 10 (January 1969), 7-11. Two of the letters (pp. 7-12), by Ronald Gottesman and Paul Baender, had previously been published in the New York Review of Books (19 December 1968); two others (pp. 13, 17-19), by Frederick Anderson and Oscar Cargill, also addressed to the New York Review, had not been published before; and a fifth (pp. 14-16), by John C. Gerber, had been sent directly to Wilson and had not been published. (A footnote to Gerber's letter points out that Wilson did not correct in the pamphlet his misstatements about the Mark Twain Papers and Works noted here.) Fisher's essay, "The MLA Editions of Major American Authors" (pp. 20-26), besides providing a historical account which includes information about Wilson's "Pléiade" plan, makes some response to Wilson's articles. The pamphlet ends with "A Calendar" (pp. 27-28), listing relevant events back to 1947. Bowers's comments on Wilson's articles appear in Two Lectures on Editing, pp. 23-25 (footnote 2) and p. 70 (footnote 30). Benjamin DeMott's "The Battle of the Books," in the New York Times Book Review, 17 October 1971, pp. 70-72, offers a journalistic account of the controversy.
It is symptomatic that Wilson, and Mumford before him, both try to divorce these editions from humanistic learning. Mumford says that the culprit behind the Emerson edition is the "Academic Establishment," fostering "the preconceptions and the mock-scientific assumptions governing the pursuit of the humanities today" (p. 4). Wilson makes disparaging remarks about editors who are not interested in literature (p. 170), exaggerates the technical language employed (p. 169), and prints with obvious delight a letter from W. H. Y. Hackett, Jr., ridiculing, among other things, the Hinman Collator (pp. 198-99). Ray's comment, though printed without a citation of source, is taken from his "Foreword" to The American Writer in England: An Exhibition Arranged in Honor of the Sesquicentennial of the University of Virginia (1969), p. viii.
The other of his principal illustrations deals with collateral texts, deriving independently from a lost common ancestor; in these situations he believes that it would be "misleading" to denominate one of the collateral texts a "copy-text." An editor's statement, however, ought to make clear the reasons for selecting a particular text as the basic one, so that the reader will not find the label "misleading." If the collateral texts are equidistant in descent from the lost common ancestor, of course, it is true that there may be no basis for selecting one over another, and the editor must then construct a copy-text on the basis of all these texts, as Bowers has explained in his discussion of "radiating texts," referred to above (footnote 13); but from that point on Greg's theory of copy-text applies as usual. And if the collateral texts are not equidistant from the lost common ancestor, the editor is able to follow Greg's rationale directly, by selecting the one nearest the ancestor unless he has strong evidence pointing toward another choice.
My views on Baender's argument are set forth in greater detail in "The Meaning of Copy-Text: A Further Note," SB, 23 (1970), 191-196. See also Bowers's comment in "Multiple Authority" (cited in footnote 13 above), p. 82 (footnote 1).
Although Pizer continually refers to "the" theory of copy-text, it is obviously Greg's theory which he is discussing. One should understand, however, that an editor necessarily has a "copy-text," whatever he may call it, and that Greg's is only one among many conceivable rationales for selecting a copy-text.
That the result might be "complex and bulky" is undeniable; that it is therefore "all but unusable" does not follow. Anyone who wishes to comprehend a complex textual history would presumably not expect to find the evidence as easy to follow as it might be in less complicated situations.
Pizer is particularly worried about the future republication of CEAA clear texts without accompanying apparatus: "It is a nice point," he believes, "whether a clear-text critical edition sans apparatus is any different from an unedited text"—because "in either case the reader must go to considerable effort to check the evidence" (p. 152). A critical text exists to present an informed reconstruction of an author's intended text, based on an examination of all known evidence and on critical insight into the author's aims and methods; a text which is "unedited" (presumably edited only by the original publisher's editor or reproduced from a contemporary printing by a later editor) does not purport to serve this purpose. There is nothing similar about the two except that they are texts of the same piece of writing and that they are texts as opposed to apparatuses. If a reader wishes to consult the documentation which an apparatus provides and finds no apparatus accompanying his text, he may be somewhat inconvenienced by having to go to a library to examine a CEAA edition (text plus apparatus), but surely less so than if he had to collect the evidence himself with which to judge an "unedited" text.
If Pizer is concerned (as he seems to be in his proposal of "textual organicism," pp. 152-153) to preserve as a respectable possibility the idea of a collected set which includes some "definitive editions" along with some "practical editions" (to use Bowers's terms), all one can say is that there is no theoretical objection to it, so long as each text is clearly labeled for what it is.
Pizer, in his footnote 5 (p. 148), gives the impression that he has not fully grasped Greg's central insight: that there is no reason to expect authority in substantives and authority in accidentals to reside in the same text. What Pizer says is that the "suspicion of later texts . . . affects the entire matter of the choice of copy-text while receiving explicit expression primarily in relation to accidentals." That is of course just the point: the potential authority of a later text in respect to substantives is in no way affected by the choice of an early text as the authority in accidentals.
Pizer's sixth footnote (p. 148) recognizes this fact and is a more trenchant discussion of the issue than what appears in the body of his article: "I should note my awareness of the great range of variation possible within the publisher-author relationship and of the consequent need for editorial knowledge and discretion in determining the degree of authorial acceptance of a publisher's changes."
While it is possible to argue that the change of even a single word in a text produces a new work, critical discrimination has not advanced very far which makes no attempt to locate that point along the spectrum of revision where alterations to improve the expression of one conception give way to alterations that shift the conception itself.
Grabo, "Pizer vs Copy-Text," BNYPL, 75 (1971), 171-173; Parker, "In Defense of 'Copy-Text Editing,'" 337-344. Bowers makes a few remarks on Pizer's article in "Multiple Authority" (cited in footnote 13 above), pp. 86-87 (footnote 11).
"How Not to Edit American Authors: Some Shortcomings of the CEAA Editions," BNYPL, 75 (1971), 419-423.
Because Freehafer cites Hershel Parker's reply to Pizer in connection with this discussion, Parker makes a further brief comment in "Historical Introductions vs Personal Interpretations," BNYPL, 76 (1972), 19. Freehafer's statement that "those who cannot successfully criticize an author ought not to edit him" (p. 420) seems not to recognize editing as a critical activity itself; what constitutes "successful" criticism is of course an open question.
As for explanatory notes, the CEAA editions have not by any means uniformly excluded them, and the CEAA Statement encourages them for certain kinds of works. In any case, Freehafer's belief that an editor who does not provide explanatory notes will be less likely to detect, for example, errors in the spelling of proper names is merely questioning editorial competence in general; any responsible editor investigates the spelling of names of persons and places as a routine part of his job, and, if he makes a mistake in a given instance, the fault surely cannot be traced to the fact that he was not obliged to write explanatory notes.
The view that the CEAA editions "will probably be looked upon in the future as a monument to a temporary overemphasis on an imperfectly borrowed and excessively bibliographical style of editing" (p. 421) is puzzling. The "bibliographical" emphasis is an emphasis on establishing the history of each text, and the CEAA editors could be charged with "excess" in this regard if they claimed that all future editions of these works should have the same emphasis; but it is odd to regard as "temporary overemphasis" the effort to put on record information which will be useful in the future for producing different kinds of editions with texts based on other principles.
"The Marble Faun and the Editing of Nineteenth-Century Texts," Studies in the Novel, 2 (1970), 487-503. This article is a detailed review of the Ohio State edition of The Marble Faun and makes fewer general observations than the title might suggest. It represents the kind of close examination of a CEAA volume which has been all too infrequent, but for the most part it is concerned with the evaluation of particular emendations rather than with questions involving the use of Greg's rationale. One paragraph (pp. 498-499), however, does say that, because Greg's theory seems to work in connection with The Marble Faun manuscript, it does not follow that the theory can be applied "to all fair-copy manuscripts of the nineteenth century," since certain authors (several are cited) are known not to have punctuated their manuscripts for publication and others (Henry James) are known to have carefully revised their punctuation for later editions. It should not be necessary to repeat that Greg's theory does not demand the use of manuscript as copy-text when there is convincing evidence favoring another course. Freehafer makes some of the same criticisms of the Hawthorne edition, especially in regard to emendations resulting from a policy of "normalization," in his reviews of Hawthorne scholarship in Jahrbuch für Amerikastudien, 15 (1970), 293-294, and 16 (1971), 268-269.
"'On the Editing of Modern American Texts': A Final Comment," BNYPL, 75 (1971), 504-505. Pizer has published remarks on CEAA editions or on Greg's theory in a number of other places. For example, he has commented unfavorably on the Crane edition in a review in MP, 68 (1970-71), 212-214, and in his survey of Crane scholarship in Fifteen American Authors before 1900, ed. R. A. Rees and E. N. Harbert (1971), p. 100 (the edition reflects "the present emphasis on critical texts and common sense be damned"). And his Rosenbach lecture, "Dreiser's Novels: The Editorial Problem" (published in Theodore Dreiser Centenary, 1971), asserts that Greg's theory, by causing editors to focus on that prepublication state which is "at once chronologically closest to the printed book and still completely sanctioned by the author" (p. 10), results in neglect of earlier prepublication states. A twentieth-century author, he says, "was more apt than his fellow novelist of a hundred years earlier to find that what appeared in a first edition was indeed what he wanted to appear in that edition" (p. 11); as a result, the real editorial problem becomes—for Dreiser, at any rate—"not to determine his final intention but to use the material at hand to demonstrate how he reached that intention" (p. 12). Most editors, however, unless they have decided to edit a particular early version of a work, are inevitably concerned with "final intentions"; but that concern does not mean that their editions cannot include data relevant to a genetic study of the prepublication stages of the work, and indeed the CEAA Statement urges editors to include a record of at least the substantive pre-copy-text variants. (The principal difference between the two approaches is that Pizer prefers printing selected passages from earlier states as footnotes to the main text, whereas the CEAA Statement recommends a discussion, illustrated by quotations, of the nature of the various stages of prepublication revision.)
Proof, 1 (1971), 122-155. Peckham's ideas had earlier appeared, in compressed form, in the discussion of "General Textual Principles" in the first and second volumes of the Ohio University edition of Robert Browning (1969, 1970), pp. vii-ix; beginning with the third volume (1971) this section is somewhat expanded (pp. vii-xiii) and contains a reference to this Proof essay.
It is not clear, however, why "sparse and inconsistent" punctuation in itself justifies the separate treatment of substantives and accidentals. Similarly, Peckham states in the next paragraph, "Greg's distinction rests upon the fact . . . that nobody but Ben Jonson [among Renaissance dramatists] took writing for the public theater seriously" (p. 125). Does this imply that later writers of "serious" literature normally managed to exert careful control over the printed forms of their work? And does it imply that for works of later periods editing usually involves only the correction of obvious errors in a text which otherwise embodies the author's final wishes in every respect?
When Peckham says that "what to do about punctuation is an empirical matter, not a theoretical matter, not a matter of editorial principles or rules" (p. 126), he is actually agreeing with Greg's position that each situation must be examined on its own terms; Greg was providing a "rule" only for those situations in which empirical evidence does not convincingly settle the question.
Of course, errors are made in the transmission of substantives as well as accidentals. David J. Nordloh, in "Substantives and Accidentals vs. New Evidence: Another Strike in the Game of Distinctions," CEAA Newsletter, No. 3 (June 1970), pp. 12-13, cites an instance in which a substantive variant between a manuscript and a magazine text can be shown to have resulted from a typist's error in the intervening typescript. It may be that without the typescript an editor would have taken the magazine reading as the one intended by the author; but the existence of such instances does not affect the general proposition that substantives as a class have normally received more careful treatment in transmission than accidentals.
Peckham's recollection of having heard, as a boy, stories about the serious consequences of incorrect punctuation in government documents (p. 125) does not really illustrate any common awareness "that punctuation cannot be separated from words," for the point of telling such stories is that the situations involved are exceptional and contradict everyday experience. Similarly, the "shift from rhetorical to syntactical punctuation in the first half of the nineteenth century" is not convincing evidence of such an awareness; both approaches support Peckham's view that punctuation consists of written signs for juncture, pitch, and stress and thus is part of the meaning, but he does not make clear why the shift from one to the other reflects a general awareness of this point.
One further observation of Peckham's on accidentals deserves notice. A logical consequence of his view that accidentals and substantives are inseparable is to deplore the absence of accidentals variants in the historical collations in CEAA volumes. The CEAA Statement naturally does not prohibit their inclusion; but it is undeniable that their absence springs not merely from the great expense that would be incurred in most cases by listing them but also from the emphasis of Greg's theory itself on the lack of significance of post-copy-text accidentals. As Peckham points out, a record of variants in accidentals would be important for the historical study of punctuation, and in addition, of course, it would give the reader a still fuller picture of the evidence which the editor had at his disposal. My own view is that variants in accidentals ought to be included whenever feasible (and particularly when the copy-text is a manuscript); but the time and money involved may in many—perhaps most—instances seem out of proportion to the amount of use that would be made of such information. See also footnote 35 above.
Peckham uses the word "change" as more neutral than "revise," for he is under the impression that "'revise' now generally means to change for the better" (pp. 138-139).
As I called it in "Textual Study and Literary Judgment," PBSA, 65 (1971), esp. 113-114. I do not mean that "scholarly" editing is not also creative in a general sense but am using "creative editing" as a shorthand way of referring to editing which has a different aim from that of historical reconstruction. Lewis Leary, who uses the term in "Troubles with Mark Twain," Studies in American Fiction, 2 (1974), does not, in my view, sufficiently distinguish between the editor who adopts a reading which seems "best" to him and the editor who selects what he thinks would have been regarded as "best" by the author.
In the section of his essay on analytical bibliography and science (pp. 129-136), Peckham recognizes that the analytical bibliographer is a historian; see Tanselle, "Bibliography and Science" (cited in footnote 70 above), pp. 83-87.
Whether one is attempting to reconstruct his first, or last, or some other, intention is an important matter but is beside the point until one grants the goal of reconstructing authorial intention of some sort. The question of "original" versus "final" intentions is helpfully illustrated by examples from Melville in Hershel Parker's "Melville and the Concept of 'Author's Final Intentions,'" Proof, 1 (1971), 156-168.
When Peckham says that "textual editing" is "logically independent of problems of aesthetics" (p. 136), he means that the artistic status of a work (whether or not it is generally considered to be an effective work of art) has nothing to do with the process of editing the work. But that does not mean that critical or aesthetic judgment is not involved in the editor's assessment of the evidence. At another point (p. 151) Peckham states, "The notions of text and author have been responsible for the fact that a discipline which came into existence as a reaction against textual eclecticism has returned to textual eclecticism"; but the more likely explanation would seem to be simply the growing recognition that it is foolish to attempt to eliminate critical judgment from historical research.
Let me repeat: I recognize that Peckham is asserting the essential identity of all "editorial" actions and that the scholarly editor I am speaking of cannot avoid, in Peckham's terms, producing his own version of a work. I am not disagreeing with Peckham on this point but am trying to show that there are valid discriminations to be made nevertheless among the versions produced and that it is not meaningless to regard some as approaching more closely than others to an "authorial" version, even if what is "authorial" must be to some extent a subjective judgment. (A "critical" edition, of course, by definition involves an editor's inferences about authorial intention, as Bowers makes clear by using the word "inferential" in his description of an editor's aim as "an attempt to approximate as nearly as possible an inferential authorial fair copy"—in "Textual Criticism" [cited in footnote 30 above], p. 33.)
Their whole approach shows their recognition of the impossibility of a definitive text, even though they, too, sometimes contribute to the confusion by an imprecise use of the words "text" and "edition."
I have commented in somewhat more detail on these parts of Thorpe's argument in "Bibliography and Science" (cited in footnote 70 above), pp. 78-80. Thorpe devotes an entire chapter, "The Province of Textual Criticism" (pp. 80-104), to setting forth the view that the "bibliographical orientation" of textual criticism is excessive; and his discussion of textual criticism as a "science" occupies the second section (pp. 57-68) of his chapter on "The Ideal of Textual Criticism" (this chapter was originally read on 8 February 1969 at a Clark Library Seminar and published that same year, along with a paper by Claude M. Simpson, Jr., in a pamphlet entitled The Task of the Editor, pp. 1-32).
Thorpe believes that the title Statement of Editorial Principles should be "Statement of Editorial Methodology" (p. 73), and he takes the title to be indicative of a confusion between aims and techniques. Actually the CEAA pamphlet deals with both, as the title of the 1972 revised edition (Statement of Editorial Principles and Procedures) attempts to indicate.
Thorpe is correct in saying that "actual collations never provide more than some facts on which the trained intelligence can work" (p. 73); but the point he misses in the Statement's remark that relevance is decided by collation is simply the fact that external evidence (e.g., an author's statement that a particular edition is a piracy) must be tested by what appears in the text itself (e.g., the presence or absence of differences that could reasonably be regarded as authorial revisions). See footnote 27 above.
If Peckham needs evidence that people still do react differently to accidentals than to substantives, Thorpe's statement here (and elsewhere, as on p. 74) provides a good illustration. Philip Young, like Thorpe, seems to judge the worth of editorial labor by how dramatically the text is altered, when he remarks that he "cannot find a single really significant difference between the new text [Ohio State Scarlet Letter] and that of the Riverside Edition (1883), regularly referred to in the bibliographies as Standard" ("Hawthorne and 100 Years: A Report from the Academy," Kenyon Review, 27 [1965], 215-232; reprinted as "Centennial, or the Hawthorne Caper," in Three Bags Full [1972], pp. 79-98). Young's discussion fails to recognize that what he regards as insignificant may appear significant to another critic and that the evidence, whatever it is, should be available in print for all to consult.
This statement occurs in a paragraph which was not present in the 1969 published version of this chapter.
The CEAA Statement does not take a great deal of space to make this point, nor does Thorpe, who says, "The sources of such information are so various that it is hardly worth mentioning any, except as examples" (p. 181).
This chapter is an expanded version of a paper read at the University of Kansas on 30 April 1971 and published later that year as a pamphlet entitled Watching the Ps & Qs: Editorial Treatment of Accidentals.
Additional evidence of the freedom with which nineteenth-century compositors altered accidentals is offered in two of James B. Meriwether's contributions to the CEAA Newsletter: "House-Styling, Vintage 1856," No. 3 (June 1970), pp. 11-12; "'On Careless Punctuation,'" No. 5 (December 1972), p. 3.
Thorpe's position on this question is criticized by Peckham, who points out (p. 152) that "intention" about past events must inevitably be a reconstruction, for which an author's statement is only one piece of evidence. In an essay called "The Intentional? Fallacy?", included in The Triumph of Romanticism (1970), Peckham states the point more fully: "Briefly, an inference of intention is a way of accounting for or explaining the generation of an utterance; it can never be a report. The speaker of an utterance has greater authority than anybody else in his so-called intentional inference only because he is likely to have more information for framing his historical construct, not because he generated the utterance" (p. 441). See also footnote 81 above.
Although Thorpe says, "The basic principle is that the author's intentions with respect to accidentals should be carried out" (p. 198), he also asserts, at the end of the same paragraph, "Whether the text should be presented in old-spelling or in modernized accidentals is mainly a matter of convenience for the intended audience"—as if there is no contradiction involved. He surveys the arguments for and against modernizing accidentals on pp. 134-140 and pp. 169-170 and concludes, "I can say that the losses from modernization seem to me less than most textual scholars assume" (p. 170). Nevertheless, despite his reluctance to distinguish clearly the purposes and implications of the two kinds of texts, the primary focus is on the author's own wishes: "Our task is, I believe, to fulfill the intentions of the writer in these small details [accidentals] as well as in greater matters" (p. 165).
For Bowers's criticism of Thorpe's position, see footnote 6 in "The New Look in Editing," South Atlantic Bulletin, 35 (1970), 8; for Peckham's, see Proof, 1 (1971), 122, 135-138, 152. See also the reviews by John Feather, in MLR, 68 (1973), 381-382, David J. Nordloh, in Resources for American Literary Study, 3 (1973), 254-257, and G. R. Proudfoot, in Library, 5th ser., 28 (1973), 77-78.
Bowers, in his review of Gaskell, "McKerrow Revisited," PBSA, 67 (1973), 109-124, speaks of Gaskell's "rejection of Greg's classic theory of copy-text" (p. 122). But Gaskell does not reject it totally: his explanation of the difference between substantives and accidentals and of the reason for choosing an early text as authority for accidentals is obviously derived from Greg. What Gaskell rejects is the logical extension back to the manuscript of the steps that led to the choice of the first edition over a later edition as copy-text. Cf. my review in Costerus, n.s. 1 (1974).
The separate question of whether different versions of a work exist, each deserving to be edited separately from different copy-texts, is touched on by Gaskell in a somewhat confusing way. Near the beginning of the chapter he calls it "an anomaly of bibliographical scholarship today" that "much effort is expended" on editing works "of which the early texts differ from each other only in minor and frequently trivial ways," while "books of which we have texts in several widely different forms are either avoided by editors or edited in a single version" (p. 337). If "single version" means one of the author's versions, the process would seem to be what the situation probably calls for; if it means (as the context suggests) an eclectic or "critical" text, one would have to say that such a text might be, but would not necessarily be, inappropriate—depending on the way in which the forms of the work are "widely different," whether as a result of a large number of changes or as a result of the nature of what changes there are. That Gaskell is thinking primarily in quantitative terms is shown later in the chapter. In discussing authorial revision of printed texts (p. 341), he states that the first edition remains the copy-text, provided that the author did not revise the punctuation and provided that "the revision [of substantives] is not extensive (say no more than a word or two in each paragraph)." Similarly, after describing the extensive revisions of Pamela (citing 8400 changes in the first two volumes of the last version), he asserts, "Here it would obviously be impossible for an editor to incorporate the first, the intermediate, and the final versions of the novel in a single critical text" (p. 342). Maybe so, but the reason is not the sheer number of changes; for only the nature of the differences, and not merely their quantity, can justify regarding two versions of a work as, in effect, separate works. I discuss this point further in my forthcoming article on "The Editorial Problem of Final Authorial Intention."
Of course, "rigidly" prejudices the case, since an editor who "rigidly" followed the manuscript, or any other text, without regard for the specific evidence involved would plainly be in the wrong.
That regularizing or normalizing amounts to modernizing has been made clear by Hershel Parker in "Regularizing Accidentals: The Latest Form of Infidelity," Proof, 3 (1973), 1-20. Furthermore, as he says in his cogent concluding section on the dangers of modernizing, "Normalizing to satisfy an editor's instinct for tidiness or to make smooth the way of a reader is ultimately demeaning for the editor and insulting to the reader." I have made some further comments on regularizing in "Bibliographical Problems in Melville," Studies in American Fiction, 2 (1974), 57-74, and in "The New Editions of Hawthorne and Crane," Book Collector, 23 (1974), 214-229.
A third book of a general and introductory nature which appeared in 1972 is F. W. Bateson's The Scholar-Critic: An Introduction to Literary Research. Although not limited to bibliographical and textual matters, as Thorpe and Gaskell are, it contains a chapter on "Textual Criticism" (pp. 126-146) which belittles the "'biblio-textual' school" of Greg and his followers (for supposedly attempting to eliminate literary judgment from editing) and endorses modernized texts. His argument for modernization seems strangely inconsistent with his own general position. He begins by labeling as a fallacy the view that "the ability to compose great literature necessarily carries with it the ability to spell and punctuate it correctly" (p. 139). But he later makes the sensible distinction between "good English" and "correct English," and it would seem that the attempt to enforce "correct" punctuation (by whatever standard) on an author's work would reveal as petty an attitude as to wish that he had been more "correct," and perhaps less effective, in his wording. And if printed literature is to be considered a recorded form of oral language—as Bateson describes it in "Modern Bibliography and the Literary Artifact," in English Studies Today, 2nd ser. (1961), pp. 67-77, from which part of this chapter is drawn—it would seem that punctuation would have to be regarded as an inextricable part of the effort to convey nuances of meaning in print. When he recommends that accidentals "should always be modernized" except when "such a process affects the meaning" (p. 142), it does not seem unreasonable to conclude that his statement amounts to saying that accidentals should never be modernized. That we can understand Shakespeare without reading him in his own pronunciation—a point cited by Bateson in support of his position—is irrelevant; for, while the pronunciation of words does not, within limits, seriously affect their meaning, the way in which we are directed to speak them by the punctuation does frequently affect it. (Bateson has also expressed his view of modernizing in a letter to the TLS, 1 January 1971, pp. 14-15.)
While the present paper was in proof, another essay critical of Greg's theory appeared. Vinton A. Dearing, in "Concepts of Copy-Text Old and New," Library, 5th ser., 28 (1973), 281-293, argues that Greg's procedure, by emphasizing an early text rather than a later one, "implies that a scribe or compositor regularly puts more errors into a text than the author takes out of his copy" (p. 293). Actually, of course, it implies no more than that errors do creep into a text as it is transmitted; each variant must still be given careful individual attention. Dearing's proposed solution raises many more questions than it answers: "Count the changes certainly made by the author and those certainly made by the scribe or compositor, and assign the rest to the cause with the greater total." I trust that it is unnecessary to enumerate the difficulties which such a statement involves. Still another relevant essay which appeared too late to be cited above is "The CEAA: An Interim Assessment," by Hershel Parker with Bruce Bebb, PBSA, 68 (1974), 129-148, which offers succinct evaluations of the CEAA editions in respect to design, arrangement of material, textual policies, and the provision of historical essays.
| | |