University of Virginia Library

II

Bowers was forty at the end of the war, and was almost at the midpoint of his life. The four preceding decades had been preparatory, on both a personal and a professional level, for the extraordinary fulfillment—on both levels—that came to him in the following four decades. Once back from Washington, he wasted no time in making a new beginning, pushing ahead in the bibliographical studies that had captured his attention before the war; and in 1948 and 1949 he was responsible for two publications that permanently altered the course of bibliography. The first was a volume of bibliographical essays and notes, edited by him, called Papers of the Bibliographical Society of the University of Virginia. The Bibliographical Society of the University of Virginia had been formed in February 1947, with the collector Chalmers L. Gemmill as

president, Bowers as vice-president, and the University's rare-book librarian, John Cook Wyllie, as secretary-treasurer, in order "to foster [in the words of its constitution] an interest in books (including books in manuscript), maps, printing, and bibliography." Bowers had himself read the first paper before the Society at its inaugural meeting on 26 February 1947, a paper entitled "Some Problems and Practices in Bibliographical Descriptions of Modern Authors"; in it he stressed the need for a more scholarly approach in this area and praised the work of Michael Sadleir, Percy Muir, and Jacob Blanck. By late 1948 a dozen papers had been read by speakers from Baltimore, Philadelphia, New York, and Washington, as well as from Charlottesville (and many of the papers had been published in mimeographed form). Among the speakers were Charlton Hinman (on why the Folger Shakespeare Library has so many copies of the Shakespeare First Folio), Harold Tribolet (the Chicago bookbinder), C. William Miller (later the bibliographer of Benjamin Franklin), and Giles Dawson and James G. McManaway of the Folger.

The Society had also published eight numbers of a newsletter (called Secretary's News Sheet), and in the seventh (dated June 1948) this announcement appeared:

At about the same time a flyer was widely distributed, advertising the new publication: a two-leaf fold of Strathmore Pastelle, it showed the title page on its first page and a sample page of text on its third (a handsome layout in Goudy's Monotype Garamont), with a listing of the contents on its second page and an order blank on its fourth.

The volume that appeared at the end of the year (two hundred pages, printed in December at the William Byrd Press, Richmond, and bound in heavy white paper) contained eleven articles and six "notes" and set the pattern for the volumes that were to follow. The pieces ranged widely (from the Middle Ages to the nineteenth century) and were contributed

both by established scholars (such as Curt F. Bühler, the incunabulist of the Pierpont Morgan Library) and by young assistant professors (like C. William Miller and Allan H. Stevenson) and even graduate students, including several from Virginia (such as James S. Steck, George Walton Williams, and Philip Williams). Bowers was clearly not concerned with the stature of the contributors, as long as their articles met his standards, and through this group of articles he was making the statement that he was particularly interested in work that explored new areas of analytical bibliography. The volume contained the first of Allan Stevenson's brilliant expositions of the uses of paper as bibliographical evidence (two years Bowers's senior, Stevenson had published four previous articles, all on James Shirley); Curt Bühler's piece was an experiment in applying Bowers's and Hinman's recent work on headlines to incunabula; and there were three forward-looking analytical studies dealing with the Renaissance—Philip Williams's examination of spelling to identify compositors, James Steck's statement of the usefulness of center rules as bibliographical evidence, and Bowers's own explanation of the role of headline evidence in determining half-sheet imposition. The volume was enthusiastically reviewed (in Modern Language Quarterly, 1951) by R. C. Bald, who accurately assessed the significance of its publication: He added—also correctly as it turned out—that the volume "is almost certainly the forerunner of a distinguished series."

Although The Library had already shown its hospitality to the kind of analytical article that characterized Bowers's publication, there was no doubt that Bowers was staking out this area as a particular concern, and succeeding volumes confirmed the fact. The second one, bearing the now familiar title Studies in Bibliography (the original title remaining as a subtitle), contained as its lead article an investigation by William

B. Todd of the early editions and issues of Matthew Lewis's The Monk. This article (deriving from a talk before the Society on 17 December 1948, while he was writing a pioneering University of Chicago dissertation) was Todd's first publication, the first of his many new applications of analytical bibliography to eighteenth-century English books. The volume also included another analytical piece by Philip Williams (who had just received his Ph.D. under Bowers's supervision) and an essay by Bowers himself on the use of type-page width as a reflection of the compositor's stick and therefore as an initial test for identifying the stints of different compositors. The Times Literary Supplement (London) declared (on 14 July 1950) that, with this volume, the Society "firmly establishes its sponsorship of a journal which should be on the shelves of every great reference library."

It was the next volume, however, that really made Studies (or "SB," as it was also called) a force in the scholarly world. The third volume, dated 1950 and labeled as for "1950-1951" (and consisting of over three hundred pages), presented a dozen major articles by a roster of contributors that in retrospect appears an honor roll: R. C. Bald (then at Cornell), W. W. Greg (just knighted the previous June), Archibald A. Hill (who had been at Virginia during Bowers's early years there), Curt F. Bühler, Ernst Kyriss (the great German scholar of bookbindings), Philip Williams, Charlton Hinman, C. William Miller, William B. Todd, Rollo G. Silver (who was to become the most distinguished historian of American printing), J. Albert Robbins, and Bowers himself (with an important essay on "Some Relations of Bibliography to Editorial Problems"). Bowers's energetic and astute solicitation of articles, demonstrated so often in the ensuing years, was illustrated here by the presence of four significant papers from the English Institute of 1949—those of Bald, Greg, and Hill, as well as his own. Contemporary readers could not have known that Greg's essay, "The Rationale of Copy-Text," would become one of the most seminal papers in the history of English scholarship, the point of departure, and also sometimes of contention, for textual critics throughout the next half-century. But surely those early readers shared with the anonymous reviewer in the Times Literary Supplement for 23 March 1951 (in fact, John Carter) the view that this volume "must take its place in the first rank of such publications anywhere in the world."

With the fourth volume (1951), appearing in hard cover (paper-covered boards) with paper labels on the spine and front cover, Studies assumed the elegant look that was to continue through 1966 (when stamping replaced the labels). With more work by Stevenson, Todd,

Bühler, and Bowers himself, and with articles by G. I. Duthie, James G. McManaway, and Giles E. Dawson, this volume continued to feature bibliographical analysis and textual criticism, especially as applied to Renaissance books; and the reputation of Studies in this area was reinforced by the repeated appearances, over the next fifteen years, of John Russell Brown, I. B. Cauthen, Jr., Charlton Hinman, Cyrus Hoy, Harold Jenkins, Robert K. Turner, Jr., Alice Walker, and George Walton Williams, among others—though there were also distinguished contributions in other fields. The sense one had in examining these early volumes was effectively caught by Philip Edwards in his review of the fifth volume: "Professor Fredson Bowers has created a new aristocracy in the world of bibliography, and the works he has here collected (so often revealing his own inspiration) have that kind of energy and sense of discovery which moved the Bibliographical Society in the days of Pollard, Greg, and McKerrow" (Shakespeare Quarterly, 1953).

Allusion to illustrious English precedents was again present when Studies reached its twentieth volume (dated 1967). Frank Francis, director of the British Museum and immediate past president of the Bibliographical Society in London, contributed a foreword that recognized the inspiration drawn by Bowers from the work of the English society. The Virginia organization, he said, had been formed by "a group of dedicated individuals banded together"—wording reminiscent of his phrase "a sense of common adventure," which he had used (in the London society's 1945 jubilee volume) to describe the early days of the older society. But "in this case," Francis added, a journal "sprang into being full-grown and fully armed, like Minerva from the head of Jupiter, whose thunderbolts she wielded from time to time." The identity of Jupiter was obvious: "the Society with the prestige it now enjoys and the great success it has had is substantially the image of its Editor, whose industry and monumental achievements in analytical bibliography are the admiration of bibliographers." Another foreword by Edgar F. Shannon, Jr., president of the University of Virginia, showed that the renown of Studies was celebrated at home as well as abroad. And the fame of Studies prompted a full-page anniversary notice in the Times Literary Supplement headed "Bibliography and Dr. Bowers" (27 April 1967), recalling the time when the first volume "came forth to astonish all the bibliographers of Europe."

The centrality of Studies, evident from its continued publication of pioneering work, was further suggested by the inclusion, beginning with the third volume, of a series of annual checklists of bibliographical scholarship; and, in turn, those lists—with major sections entitled "Printing,

Publishing, Bibliographical and Textual Scholarship"—helped to promote one of Bowers's goals for the journal by taking for granted the integral relationship among conventional studies of printing and publishing history, the newer and rapidly developing approaches of analytical bibliography, and investigations of textual problems. (Bowers discontinued these checklists with the one covering 1972 because of the duplication that would theoretically have resulted from the expansion of the Dutch publication ABHB: The Annual Bibliography of the History of the Printed Book and Libraries to include American scholarship as of 1973; but ironically much of the analytical and textual work of the kind that had been listed in Studies was not regarded by the editors of ABHB as within its scope—an indication that Bowers's view of bibliography as a field was not yet widely understood.)

One glimpse behind the scenes in the early years of Studies is afforded by Bowers's reply to the Times Literary Supplement's review of the fourth volume (on 7 March 1952): the reviewer lamented the paucity of contributions dealing with the post-1800 period in all bibliographical journals, and Bowers replied (on 9 May 1952) that he was conscious of this "imbalance," that the five articles and ten notes on this period thus far published in Studies "represent the total number submitted to us," and that he had not been able to secure more "by personal solicitation." The solicitation of material for Studies was one of Bowers's dominant preoccupations for the rest of his life: he never missed an opportunity to ask the authors of papers he heard and admired at conferences if they would let Studies print their work, and when he met or learned of scholars whose research sounded promising, he not only encouraged them to think of Studies but also often gave them detailed suggestions. His continuous solicitation of articles is amusingly epitomized by a note in his 1975 volume of collected essays: in a new footnote appended to a sentence (written in 1941) on the possibility of using headlines to detect page-by-page (rather than forme-by-forme) printing, he wrote, "For reasons that escape me, I never have pursued this promising lead. Perhaps some interested bibliographer will have a look at the problem (and let me publish his results in SB . . .)" (p. 207). Even in his collected essays, he could not resist the urge to stir up contributions. For good reason, Studies was close to his heart, and he knew that the maintenance of a first-rate journal requires the editor's unceasing attention. As one thinks about the progress of his life, and the many projects that he was involved in at any given time, one should remember that a constant in the background, from 1948 on, was all the activity connected with the annual production of Studies.

The other major bibliographical event of the late 1940s, besides the inauguration of Studies in Bibliography, was the publication by Princeton University Press in 1949 of Bowers's Principles of Bibliographical Description, one of the great books in the history of bibliography (indeed, of scholarship generally). Many years later he wrote (in a letter on 22 October 1985) that "the Principles was, or now seems, almost an accident." The reason was that it emerged as a by-product of his major research endeavor at this time (one that he continued to work on, whenever time allowed, for the rest of his life)—a descriptive bibliography of the English printed drama from 1660 to 1700. The first volume of W. W. Greg's magisterial A Bibliography of the English Printed Drama to the Restoration had been published in 1939; whether Bowers before the war conceived the idea of extending Greg's work through 1700 is not clear, but after the war (at least by early 1946) he began research for the project. He identified himself in the first volume of Studies in Bibliography as "writing a descriptive bibliography of the post-Restoration English Drama 1660-1700" (one should note that at this early date he was speaking of "writing" a bibliography, for he continued to emphasize that the kind of bibliography he was concerned with was a genre of writing, not something that one "compiles"—as he specifically pointed out in a footnote to his 1948 article in the Papers of the Bibliographical Society of America [PBSA]). His pieces in the first two volumes of Studies (1948-49) drew their examples from Restoration drama, and in a footnote in the second volume he acknowledged grants for this undertaking from the Research Council of the Richmond Area University Center and the Research Committee of the University of Virginia. One of the first results of his work on Restoration drama was A Supplement to the Woodward and McManaway Check List of English Plays 1641-1700, which he published as a 22-page pamphlet through the Bibliographical Society of the University of Virginia in 1949. If he was to write a descriptive bibliography, he had to decide on the principles and procedures to be followed; as he said in 1985, "The state of affairs was a mess, and from starting the play examinations and recording, I began to see that I could learn enough, with some special reading, to begin to put some order into a mushy situation."

The required reading at that time, besides R. B. McKerrow's Introduction to Bibliography for Literary Students (1927), was Greg's "A Formulary of Collation" in The Library for March 1934, slightly supplemented by his "Provisional Memoranda" in the 1939 volume of his Bibliography. The first published results of Bowers's thought about the situation were published in PBSA in 1947 and 1948: the first essay

(in the Fourth Quarter 1947 number), "Criteria for Classifying Hand-Printed Books as Issues and Variant States," attempted to clarify McKerrow's discussion by emphasizing the importance of cancel title-leaves as determinants of "issue"; in the second (Third Quarter 1948), called "Certain Basic Problems in Descriptive Bibliography" (read before a meeting of the Bibliographical Society of America on 5 June 1948 in Philadelphia), he talked about the necessity for examining multiple copies of every edition to be described and suggested the standards and procedures that he believed were entailed in such examination (as they had evolved in his own research for the Restoration bibliography). The depth of Bowers's feeling about the kind of work he was embarked on is suggested by the final words of this second article: It was the previous article, however, that aroused more comment. In the Times Literary Supplement for 4 September 1948, John Carter discussed it with his characteristic sympathy for any advance in bibliographical rigor, but he also shrewdly noted that "Professor Bowers will seem to withdraw too far into bibliographical purism for some readers."

One such reader was Paul Dunkin, senior cataloguer at the Folger Shakespeare Library, who responded with an article entitled "The State of the Issue" (PBSA, Third Quarter 1948), largely a defense of the more pragmatic approach of library cataloguers and a plea that the practice of bibliographers should not drift away from that of cataloguers through the adoption of "intricate philosophical distinctions." (Dunkin was to be a persistent critic of Bowers over the next two decades.) Bowers's reply, a three-page letter in the next number of PBSA, is notable, first, for its statement of a principle that he generally adhered to: "one has one's original say in print and should not feel that a formal reply is necessary for every disagreement with one's views which thereupon develops." But he did wish to use the occasion to emphasize the difference between a catalogue and a bibliography, and the letter is a good early example of his firmness in controversy: "I cannot as a bibliographer," he said, "accept the complacent notion that issue is something everybody knows but nobody can define. . . . I feel myself obliged to disagree with any professional cataloguer who argues that descriptive bibliography must

conform to cataloguing standards. . . . such standards are wholly inadequate for the purposes of rigorous bibliographical research." There can be no doubt that Bowers's argument was far more logical than Dunkin's, but this kind of blunt statement was not calculated to promote mutual understanding.

That the practice of bibliographical description was a topic much in the air is further suggested by the fact that the Rosenbach Lectures at the University of Pennsylvania in the spring of 1947 took up the subject, with Curt F. Bühler speaking on incunabula, James G. McManaway on early English literature, and Lawrence C. Wroth on early Americana. Although the published volume of these lectures (entitled Standards of Bibliographical Description) did not appear until about the same time in 1949 as Bowers's Principles, there was some interchange of ideas among the four scholars. Bowers noted in his foreword that he had been given typescripts of the three lectures; McManaway, in turn, read parts of the Principles in typescript, and Bühler acknowledged in print his "deep gratitude" to Bowers, whose "kindness in making his notes readily available" was "an example of scholarly coöperation all too rarely encountered" (p. 45). Bowers and Bühler did not agree on the description of incunables, however. Bühler thought that the Greg-Bowers style of signature collation was not suitable for incunabula; as a result, he said, "the most recent exposition of the Gregian bibliographical shorthand, which it was my privilege to read in manuscript, has left me very dissatisfied" (p. 26). Bowers's third appendix to the Principles specifically set out to respond to Bühler's objections and to show how the formulary could be applied to fifteenth-century books. Bowers was also in touch with Greg at this time. Greg's full statement of the method he employed in his Bibliography had been drafted in 1942, even though it was not published until 1959 (in the fourth volume of his Bibliography), and he was therefore able to read Bowers's typescript in the light of what he had already written. He chose not to insert references to the Principles in his final version, however, and simply noted that "Professor Bowers and I do not always see eye to eye" (p. iv). If his brief public acknowledgment of "helpful criticism and correction" from Bowers was not particularly magnanimous, he was nevertheless privately generous to Bowers in making detailed suggestions; and Bowers handsomely recorded his gratitude in his foreword:

Bowers had clearly by this time come to regard Greg as a model, and he dedicated the book to Greg.

The Principles, as it finally emerged out of these discussions at the end of 1949 (officially published on 30 December), was a 500-page book that concentrated for its first three hundred pages on sixteenth- and seventeenth-century books and then, after relatively short sections on the application of the same procedures to fifteenth- and eighteenth-century books, devoted about a hundred pages to the nineteenth and twentieth centuries. In the major sections, a thorough discussion of classification (into editions, impressions, issues, and states) was followed by recommendations for title-page transcription, signature collation, statements of pagination and contents, and descriptions of typography, illustrations, and (for the later books) paper and publishers' bindings. The book was by far the most detailed treatment the subject had ever received, serving in matters of rationale as well as of procedure to draw together the developments of the previous hundred years and to provide a coherent framework within which to proceed. Any careful reader of this book must be struck by the rigorously logical and powerfully synthesizing mind that it reveals. It takes for granted that descriptive bibliography is a branch of historical scholarship, and it works through in great detail the various situations that bibliographers may have to deal with, illustrating in the process the relevance of artifactual evidence to intellectual history.

In the foreword, Bowers stated one of the aims of descriptive bibliography as providing "an analytical investigation and an ordered arrangement of these physical facts which would serve as the prerequisite for textual criticism"; and in the first section, he explained the relation of physical details to textual and literary criticism, pointing out that the "investigation of the text of a book as it passed through the printer's hands may produce facts of considerable critical importance, particularly as the evidence may turn on the relation of the printed text to the manuscript copy" (p. 15). The interrelations of descriptive bibliography with analytical bibliography and textual criticism underlie everything that is said (and are implied by the statement on the dust jacket that the book "is intended to supplement and act as a companion piece to McKerrow's well-known Introduction to Bibliography"). The book fully deserves to be called magisterial, but its author took a more limited satisfaction in it than one might imagine. "It was done," he said (in that 22 October

1985 letter), "under intense time-pressure from Princeton and could have been better if I had not been so pushed. At the time I was reasonably satisfied with most of it as the best I could do at that moment. It is only of late that I have been truly surprised that it turned out to be something of a seminal work." There is no question about its stature: despite a few deficiencies, which Bowers knew were present and which have become more evident in hindsight, it is the basic book in its field, and it is not likely ever to be supplanted. It may have been a by-product of his work on Restoration drama, but it was the book that made him famous in the scholarly world, and deservedly so.

One might expect such a work to have been eagerly welcomed, as it indeed was by some reviewers, but others were repelled by its quantity of technical detail and its unrelenting emphasis on extensive investigation. The reviewer for the Times Literary Supplement (29 September 1950) said concisely what many people apparently believed—that Bowers "seems inclined at times to make a comparatively easy matter difficult by over-elaboration." Herbert J. Davis, not an unsympathetic reader by any means, raised a similar point much more thoughtfully (in Year's Work in English Studies for 1949): Bowers, he said, "bravely" attempted to answer all questions, but the "danger of such elaboration is that it may so widen the division between the bibliographer and the student of literature that it will be impossible even for the textual critic to use without difficulty some of the technical devices now employed in the analysis of the physical make-up of a book." On his third page, Bowers acknowledged, "I am conscious of attempting to set a standard for descriptive bibliography which is not customarily thought to be necessary and hence has been seldom observed." The system he set forth was actually not very complex, not nearly as formidable as people often seemed to think; but the sheer size of his book, the result of his thoroughness, unfortunately encouraged the perception of descriptive bibliography as an arcane specialty.

When Bowers wrote, descriptive bibliography was at least taken seriously for books from the first few centuries of printing, and the Principles was the culmination of the English tradition of studying book structure, a tradition that stemmed from Henry Bradshaw, librarian of Cambridge University from 1867 to 1886. But another English tradition had grown up in the late nineteenth century, as collectors began to consider the literature of their own century worth collecting: the production of "author bibliographies" that were actually simple listings intended to enable collectors to identify the "first editions" of the authors they wished to pursue. Although by 1949 several of the bibliographies of

nineteenth- and twentieth-century authors were somewhat more elaborate, there was still a prevalent feeling that the detail required for older books was not necessary for machine-printed books: there was still a split between the scholarly description of the structure of the artifacts called books and the identification of "modern first editions" for collectors. Bowers addressed this problem in the introductory chapter to his discussion of nineteenth- and twentieth-century books: in no uncertain terms, he pointed out that for books of these centuries "little has been done of a quality which can bear comparison with work devoted to books printed on the hand press" (p. 356). "Bibliographical study," he said, "is essentially a continuous operation, and truly scholarly investigation of modern books can be engaged in only by those writers who are conversant with the methods of research developed for earlier books" (p. 365). He went on to say, "The requirements of collectors cannot be neglected, but these are fully met as by-products of the work and should not be its sole reason for existence" (p. 366). A few earlier bibliographers, notably Michael Sadleir, had understood this point, but in general the treatment of modern books was dominated by a tradition that did not recognize the relations between collecting and scholarship.

It is not surprising, therefore, that Bowers's approach and style provoked a considerable amount of resistance. John Carter, generally sympathetic with Bowers's aims, put the matter astutely in his 1951 Windsor Lecture at the University of Illinois (printed in Nineteenth-Century English Books, 1952): "here is a powerful—indeed an autocratic—intelligence, addicted to formula and impatient of imprecision, at large in an area whose bibliography is still in a highly fluid, and therefore vulnerable, state" (p. 80). (Eighteen years later, when presenting Bowers with the Bibliographical Society's Gold Medal, Carter called him "the first man to adapt McKerrovian principles to those bibliographically complex centuries which McKerrow himself never tackled.") The most famous episode illustrating this situation is Geoffrey Keynes's delivery of his Presidential Address before the Bibliographical Society in London on 17 March 1953 (printed in The Library the same year). At that time Keynes had published the first eight of his well-known bibliographies, but he did not greet the Principles as a helpful guide; rather, he announced that the book was responsible for "the shadow which seems in recent years to have descended over our amiable bibliographical discipline as we have conceived it to be." The publication of the Principles (a "remarkable and, indeed, splendid book") was, he said, "an event of shattering importance to the little world of bibliography, because it brought home to our consciousness the fact that what we had thought

in our innocence was a pleasant, if sometimes exacting, pastime, was in fact a prime example of 'pure scholarship', to be pursued with the mind of a detective, the spiritual temperature of an iceberg, and the precision of a machine."

Nothing could have revealed more dramatically the state of descriptive bibliography than for the president of the society that had been responsible for establishing bibliography as a scholarly field to complain about the Principles for being too scholarly. In his autobiography, The Gates of Memory (1981), Keynes described the event:

What Keynes did not say is that the discussion period had begun with the silence of embarrassment and that Bowers (who was sitting between John Crow and William B. Todd) spoke only after Crow had whispered to him that everyone was waiting for him to break the silence. Bowers's own version of this episode (in a letter to his former student Matthew J. Bruccoli on 7 February 1970) was that John Hayward gave Keynes "an almost embarrassing going over at his Presidential Address . . ., at which I was present smiling like a Cheshire cat, but keeping my mouth shut except to compliment Keynes on the felicity of his title." (Bowers, who was in England the entire year of 1952-53 on a Fulbright Fellowship, had actually presented his views from the same platform four months earlier, on 18 November 1952, when he addressed the Society on "Purposes of Descriptive Bibliography, with Some Remarks on Methods," elaborating the implications of his thesis that "bibliography is properly an advanced form of independent scholarship.") The graciousness of Bowers's remark following Keynes's address was typical of his courtesy in social situations; but he expressed his real opinion on other occasions, as when, nearly six years later, he called the attitude represented by that address "the very reverse of scholarly" (in his Kansas lecture "The Bibliographical Way").

Keynes's odd belief that an emphasis on rigor and precision was somehow anti-humanistic and his fear that bibliography was tending to become an end in itself are two concerns that, illogical as they are, Bowers's book has continued to provoke, for the reluctance to see descriptive bibliography as a branch of historical study has been curiously persistent. Thus the Principles, though now an established classic to which many people turn with some knowledge of what they will find there, still has the power to startle other people, revealing to the open-minded the relevance for cultural history of examining books as artifacts. Within a year of its publication, according to Lewis Leary (in his review in South Atlantic Quarterly in 1951), the book was "an inspired revelation for a crusading band of young converts," and this process continues. Leary presciently stated that the book was "one of the most important and, implicitly, one of the most provocative volumes produced by literary scholarship within our generation."

During the writing of the Principles, Bowers was working on a descriptive bibliography of George Sandys, the seventeenth-century English verse-translator and travel-writer who was an official of the Virginia colony in the early 1620s. He had been drawn to Sandys by Richard Beale Davis, who was preparing to write a biography of Sandys (Davis had received a doctorate from Virginia before Bowers's arrival but had taught there in the summers from 1938 through 1942). Bibliographical details from Sandys's books are cited as illustrations about a dozen times in the Principles, and full descriptions of two editions of Sandys's translation of Ovid's Metamorphoses are included among the sample descriptions at the end of the book. The complete Sandys bibliography appeared in print, as a collaboration between Bowers and Davis (although Bowers wrote the descriptions), just a few months after the Principles,

first in the April, May, and June 1950 numbers of the Bulletin of the New York Public Library and then as a separate pamphlet of fifty-four pages.

Bowers thus put his ideas about descriptive bibliography into practice in a full-scale bibliography almost simultaneously with the Principles; but this fact has often gone unnoted because of the modesty of the title, George Sandys: A Bibliographical Catalogue of Printed Editions in England to 1700. In the introduction Bowers explained that "bibliographical catalogue" was chosen for the title rather than "descriptive bibliography" ("which the fullness of the descriptions would ordinarily warrant") because "certain editions were available in too few copies for anything approaching a definitive examination which would reveal all possible variants" and because "a control copy of each edition was not used for direct page for page comparison of the typesetting in each copy examined." Bowers was obviously obliged to follow the stringent standards set forth in the Principles, where he recommended the term "bibliographical catalogue" for "elaborately conceived and executed descriptions . . . when the high requirements of bibliographies have not been completely met either in the number of copies compared or in the method of examination" (p. 5). In the Sandys he said, "It is our hope that the descriptions in this present work are actually definitive, but we cannot speak with the absolute authority which would justify the title of a descriptive bibliography." Stressing this point was a natural element in Bowers's campaign to elevate the status of bibliographical research; but the resulting title has, I think, caused many people seeking a model of bibliographical description not to look into the work. What we actually have in it is Bowers's explicit use of "the full system" set forth in the Principles. As Herbert Davis said in the Year's Work in English Studies (1950), the Sandys exemplified the "fullest requirements of a modern bibliography." It is not only a companion-piece to the Principles but the most extensive work in the genre of bibliographical description that Bowers published in his lifetime (though he did include descriptions in some of his critical editions).

With the Principles off his hands, Bowers was able to concentrate in the early 1950s on two editorial projects—a long-standing one, on Thomas Dekker, and a new one, on Walt Whitman—that were important in their own right and significant in their anticipation of his later work. During 1949, before the Principles was published, he had become convinced of the value of a new approach to Elizabethan editing set forth in an essay Greg was writing on "The Rationale of Copy-Text."

Bowers had solicited the paper for the English Institute conference to be held in September; and, as revised on the basis of Bowers's suggestions, it was read for Greg on that occasion by J. M. Osborn. Besides securing this paper for Studies in Bibliography, Bowers wrote a commentary on it, expanding and clarifying certain points and showing the relevance of the approach to Restoration literature as well. (This piece, "Current Theories of Copy-Text, with an Illustration from Dryden," in fact came out in Modern Philology for August 1950, before Greg's essay had appeared.) From then on, Bowers was the champion of Greg's approach, publicizing it in his lectures and using it as the basis for his own editorial practice.

The essence of Greg's position was that, when there are two or more potentially authoritative editions of a work, the authority for variants in wording ("substantive variants") may well lie in a later edition (if it incorporated revisions by the author), whereas the authority for variants in punctuation and spelling ("accidentals") may well lie in the earliest edition (which is the closest one to the author's manuscript and thus may preserve more of these features of it than any later edition does). Greg was therefore encouraging editors to use their judgment (informed by bibliographical analysis as well as literary sensitivity) in combining features from different documents to produce texts that could be defended as closer to what the authors intended than were the texts of any of the surviving documents. In emphasizing the exercise of editorial judgment, Greg was coming round to an attitude that had been elegantly advocated for classical literature by A. E. Housman half a century earlier. Editors of the modern languages had been more insistent on limiting the role of judgment, though by 1939 McKerrow, in his Prolegomena for the Oxford Shakespeare, had taken a tentative step in the direction that Greg finally moved with decisiveness. (In a rare excursion into bibliographical history, Bowers published in Shakespeare Quarterly in 1955 an essay on "McKerrow's Editorial Principles for Shakespeare Reconsidered," in which he placed McKerrow's position in its historical context and showed how McKerrow was moving to the position that Greg enunciated.) The time was ripe for this kind of move, and conceivably Bowers would have made it without Greg's lead; there is no question, however, that Greg's paper came at the right moment to help Bowers formulate his own position, and there is little doubt that Bowers, in turn, caused Greg's ideas to spread faster than they would have without such energetic support.

By June of 1952 Bowers had completed his work on the first volume of The Dramatic Works of Thomas Dekker, and it was published in

1953 by Cambridge University Press. Its historic status as the first edition to employ Greg's "Rationale" is not proclaimed in the volume itself, and Bowers's quiet and succinct statement of his procedure (ten pages on "The Text of This Edition") does not suggest how momentous it was. For the two Dekker plays that have authoritative second editions, Bowers stated, "I use the methods of recent textual theorists" (citing in a footnote Greg's essay and McKerrow's Prolegomena, along with his own Modern Philology article): In line with this policy, he indicated that he would follow his judgment, and no mechanical rule, in the choice between variants from so-called corrected and uncorrected states of sheets, preferring the "uncorrected" states when the "corrected" variants appeared to be alterations made by a printing-house reader without recourse to the manuscript; and he would alter the accidentals of the first-edition copy-texts sparingly, allowing inconsistencies to remain. (This general statement of editorial rationale was supplemented by textual introductions to the individual plays, setting forth their textual and printing histories, with the help of bibliographical analysis of multiple copies of the original editions.)

The other notable aspect of this edition, besides its use of Greg, was its presentation of the apparatus. Traditionally in scholarly editions a single record of variants and emendations had been placed at the foot of the page, with discursive notes sometimes appended to that listing and sometimes placed at the end. Bowers's thinking about the implications of Greg's rationale and about the convenience of the reader led him to alter this plan: at the foot of the page he noted only his substantive departures from the copy-text, moving the rest of the apparatus to the end of the text and arranging it in several sections (discursive notes, press-variants, alterations of copy-text accidentals, and a "historical collation" of substantive variants in pre-1700 editions—employing throughout two symbols, the wavy dash and the caret, that had been suggested by McKerrow to stand for a repeated word and the absence of punctuation). The importance of the copy-text, and its accidentals, in Greg's

rationale is reflected in this arrangement by the segregation of the two listings of copy-text emendations from the record of other variants, thus making it easier for the reader to focus on the editor's judgments, and indeed to concentrate at will on either the substantives or the accidentals. As Arthur Brown later said of the Dekker, in surveying the research on Renaissance drama since 1900 (Shakespeare Survey, 1961), "Bowers's methods differ in almost every important respect from those of previous editors. He is quite deliberately putting into practice the accumulated theory of fifty years of textual scholarship in this field."

This innovative plan was followed in three more stately volumes, appearing in 1955, 1958, and 1961. In the 2100 pages of the Dekker edition, Bowers provided a model for the twentieth-century editing of Elizabethan drama, illustrating the application of the latest developments in analytical bibliography and editorial thinking. This achievement naturally brought him praise; but there was criticism, too, largely occasioned by his decision to confine his introductions and annotations strictly to textual matters. In the first volume, he announced, "The explanatory notes in these volumes are not intended as elucidations of the text but only as discussions of specific emendation, or of refusal to emend." Many readers may not have grasped the full import of this statement at first, accustomed as they were to find historical and literary annotation in editions. The absence of such notes led a number of people (including Greg in Review of English Studies in 1959) to describe the edition as "austere." Arthur Brown, reviewing the first volume for The Library (1954), said, "Austerity is perhaps the keynote of the work." Recognizing that "this is really the first time that a major Elizabethan dramatist other than Shakespeare or Jonson has been edited with present-day disciplines," Brown felt it "desirable to consider what interests are best served by these methods." He questioned whether bibliographically minded scholars would not wish to have more details than were provided, while general readers might prefer to exchange some of those details for historical commentary.

Bowers replied in a lengthy letter published the following year, making clear that from the beginning Cambridge had planned to publish one or two volumes of critical introductions and commentary by another scholar. (Although he did not say so here, that other scholar was John Crow. Robert K. Turner, Jr., later conjectured that the Syndics of Cambridge University Press may have been reluctant to publish an edition by "a relatively unknown American" without a British scholar involved. In any case, the decision—whoever made it—to include only textual commentary in the text volumes set a pattern for Bowers's later

editions. Four volumes of commentary for the Dekker were eventually published in 1980 by Bowers's former student Cyrus Hoy.) Bowers also took the occasion to restate the functions of a critical edition (which, as he pointed out, do not include serving as a substitute for original editions or facsimiles), and in the process he showed his skill in debate: Seven years later, after publication of the fourth volume, Brown praised the whole edition for having set "a new and very high standard for future work of this kind"—but added, "however much an editor may dislike the principles upon which Professor Bowers has worked" (The Library, 1962). A bibliographer such as Allan Stevenson, on the other hand, naturally liked the approach and found that the "total effect" of Bowers's textual introductions was "a fuller view of printing-house practice" (Shakespeare Quarterly, 1956). And John Crow (remarking that Bowers's "practical work is as good as his theoretical") felt that the textual statement could be "taken as model by all future editors of early plays" (Year's Work in English Studies, 1953). In retrospect it is surprising that so little immediate attention was paid to the implications of Greg's rationale (which favored dramatists' private intentions over the texts of performances) and that the absence of explanatory commentary proved such a stumbling-block. But it is easy to see why the Times Literary Supplement (27 October 1961) called the edition "a major achievement."

Bowers's other editorial project of the fifties began in early September 1951 when C. Waller Barrett, the great collector of American literature who lived in the country not far from Charlottesville, showed him "a thick leather case packed full with autograph manuscripts by Walt Whitman" (as Bowers put it in his second Sandars lecture). These papers (the largest surviving block of Whitman manuscripts) consisted of the final handwritten drafts of some eighty poems prepared for the third edition of Leaves of Grass in 1860, including a large proportion of the new poems to be added to that edition. Bowers admitted the "drudgery" involved in working through the physical evidence and transcribing the texts, but he found it "fascinating" because he was able—through an analysis of the paper and its treatment—to shed light on the dating of these poems. He called his analysis "chiefly 'bibliographical,' with only

occasional notice of literary and critical matters outside of textual considerations"; the purpose of the volume, published by the University of Chicago Press in 1955 and dedicated to James Southall Wilson, was to offer "full texts," leaving to others the pursuit of "the critical possibilities which these manuscripts open up for the study of Whitman as a conscious artist" (although he explored a few of them himself two years later in his second Sandars lecture). After an extensive analytical and historical introduction (called "a masterpiece of its kind" by H. Lüdeke in English Studies, 1958), the edition consisted of parallel texts: the left-hand page of each opening presented a "diplomatic" (i.e., literal and unaltered) transcription of the final manuscript texts (with footnotes recording the manuscript revisions that preceded those texts), and the right-hand page printed the corresponding texts as they appeared in the 1860 edition of Leaves of Grass. Bowers's description of the manuscript revisions, consisting of straightforward statements that do not entail symbols, follow essentially the same pattern as those he had used in The Fary Knight and was to employ in the Hawthorne edition. He kept thinking about the problem of recording manuscript alterations throughout his career and later developed a different plan; but the Whitman volume represents the full working-out of his original approach and foreshadows—both in this respect and in its nineteenth-century American subject—his extensive editing of American authors in the last three decades of his life. It was as important in its way as the Dekker edition, for it brought a new level of editorial scholarship to the study of American authors—and at a time when there were stirrings of interest among scholars of American literature in the organizing of full-scale editions of major American authors. Gay Wilson Allen, whose biography of Whitman appeared the same year as Bowers's edition and with whom Bowers had a mutually beneficial correspondence, was correct to say (in Modern Language Quarterly, 1956) that Bowers's work "initiates a new stage of Whitman scholarship and is likely to have far-reaching consequences."

By the early 1950s, with the Principles published and with Studies in Bibliography and the Dekker edition well launched, Bowers was recognized—in the words of John Carter's 1951 Windsor Lecture—as "the most energetic and wide-ranging of contemporary bibliographical theorists" (p. 79). He was accordingly invited, during that decade, to deliver the three most prestigious series of bibliographical lectures in the English-speaking world. First came the lectures presented under the A. S. W. Rosenbach Fellowship in Bibliography (founded 1930) at the

University of Pennsylvania on 21 and 28 April and 5 May 1954, entitled "On Editing Shakespeare and the Elizabethan Dramatists." Then the two English series, the Sandars (founded 1894) and the Lyell (founded 1952) came in successive years at the end of the decade: he was Samuel Sandars Reader in Bibliography at Cambridge University in 1957-58 and delivered three lectures under the title "Textual and Literary Criticism" on 20-22 January 1958 (he was abroad for only two weeks, 18 January to 1 February 1958, between semesters); and in Trinity Term 1959 (he was in England from 5 March to 17 July) he delivered six lectures (on six consecutive Tuesdays: 28 April; 5, 12, 19, 26 May; 2 June) on "Bibliography and Textual Criticism" as the fourth James P. R. Lyell Reader in Bibliography at Oxford (following Neil Ker, W. W. Greg, and Stanley Morison). (One intriguing souvenir of that occasion was a piece of printing produced in Bowers's honor by Herbert Davis's bibliography class on 12-13 March 1959, using the Daniel press at the Bodleian; the text chosen was "The Apologie of William Iaggard," as published in Augustine Vincent's A Discoverie of Errours, 1622, and the class put together two copies of a volume showing various stages in the production of this piece, with a title page reading The Transmission of a Text: An Example Prepared by the Members of the Bibliography Class at the New Bodleian for Fredson Bowers, James P. R. Lyell Reader in Bibliography. This item of Bodleian printing was included in the 1969-70 traveling exhibition "The Bodleian Library and Its Friends" and was described as entry 106 in the published catalogue.) The texts of these three series of lectures resulted in three major books—the last books (other than editions and a students' guide to Hamlet) that he was to publish until he collected his essays in 1975 and 1989. These books provided the theoretical underpinning for his editing, and once he had them in print, making their rationale available for others to use, he concentrated on applying the principles and procedures they set forth to specific editorial tasks.

On Editing Shakespeare and the Elizabethan Dramatists (1955) began with a survey of "The Texts and Their Manuscripts," and the opening paragraph of this first lecture made clear Bowers's emphasis on a reasoned approach to editing. What he opposed was "thoughtless imitation" of method, which "marks the road to sterility"; instead, editors should understand the reasons underlying any prescribed principles and then develop a procedure from "the particular kind of problem" confronting them. Although he specifically dealt with the routes by which Elizabethan plays reached print, his clear insistence on the need for independent thinking (epitomized by his example of Greg, "the

greatest textual critic of our time" [p. 32], being led into "logical difficulties" by conventional assumptions about the role of Shakespeare's foul papers) had applications for all editors. The second lecture, on "The Function of Textual Criticism and Bibliography," dealt with "the kinds of evidence that may be granered from the printing process" and "the problems when multiple documents are preserved for the same text" (p. 34). Textual criticism and bibliography must be joined together in editing, producing what "may be called the biblio-textual method": bibliography, which attacks textual problems "from the mechanical point of view," and nonbibliographical textual criticism, which "works with meanings and literary values," must operate as counter-checks. "Bibliography endeavors to take as much guesswork as possible out of textual criticism, and the literary method endeavors to inform bibliography with value judgments as a check on mechanical probability" (p. 35). The bulk of the lecture, following several pages of these general observations, offered supporting examples from Shakespeare; but the aim of the lecture, like the first one, seems as much to set forth a sound approach for all editorial tasks as to instruct Shakespearean editors in particular.

In the final lecture, "The Method for a Critical Edition," Bowers sketched the recent history of attempts both to restrict and to liberate editorial judgment and took his stand beside Greg:

This passage not only expresses a principle in which Bowers deeply believed but also illustrates his promotional efforts on behalf of the status of editorial work. He went on to elaborate the ways in which his and Greg's emphasis on critical judgment was different from the unprincipled eclecticism of eighteenth-century editors. The demanding new requirements, he realistically observed, were time-consuming enough that "what may be called a definitive text of Shakespeare" (p. 100) was not likely to appear in his lifetime; but he was confident that the piling up of bibliographical facts and the incorporating of them into critical evaluation would bring the editor "as close as mortal man can come to the original truth" (p. 101).

These lectures, though early in his editorial career, foreshadowed the essence of his mature position, and they contained his characteristic mixture of scholarly rigor and promotional enthusiasm. There had been no previous general book-length introduction to the postwar bibliographical approach to editing, and some reviewers welcomed the book as such. J. C. Maxwell, for instance, reviewing it in the Review of English Studies (1957), said, "Few recent books give the reader a comparable sense of being right on the frontier of knowledge." Herbert Davis's review, in Modern Language Notes (1956), was typical of many that Bowers was to receive over the years in that it mentioned "his usual vigor and thoroughness," found his prose style difficult (the lectures are "written by an expert who does not always follow the more plain and easy way"), and combined admiration for Bowers with a trace of suspicion about bibliography ("There is a dangerous fascination in the very pursuit of these elaborate investigations and he is to be congratulated in the way he preserves his critical sense and good judgment"). A similar hint of disapproval is present in C. J. Sisson's generally respectful treatment of Bowers as "the central figure in the present energetic school of research which is represented in the admirable series of Studies in Bibliography" (Modern Language Review, 1956). The most severe criticisms of the book were made by Greg in Shakespeare Quarterly (1956). He objected to Bowers's minimizing of the importance of "foul papers," but he reserved his strongest invective for an attack on Bowers's style, which he regarded as "a parade of technical and pseudo-scientific language" and "a fog of verbiage." After giving numerous examples of "pretentious writing" and of the "shaky syntax" that results from a "lack of precise thinking," he noted that clear writing comes at "the price of constant vigilance and revision" and suggested that Bowers "has taken no less pains to write obscurely." (These comments were no doubt still in Bowers's mind four years later when he wrote, in his obituary of Greg for the American Philosophical Society, that Greg's "sense of outrage at violations of proper standards, and of clear English, could result in devastating criticisms.") When the book was republished in a new printing eleven years after the first (in 1966), with two subsequent lectures

(from the 1962 Shakespeare Jahrbuch and the 1966 SB) added, T. H. Howard-Hill wondered (in Shakespeare Studies, 1968) why some revision to accommodate Greg's strictures had not been undertaken. And E. A. J. Honigmann used the occasion to lament (in The Library, 1968) what seemed to him Bowers's excessive attention to popularizing: "In so far as he writes as a salesman, or a prophet, to convert the unfaithful to 'the bibliographical way', one can applaud. But, since he also happens to be the most influential man in his field, I regret that he is more interested in the hard sell than in the harder reading." Bowers himself saw no difficulty in being both a bibliographical evangelist and a productive scholar, but his reviewers frequently did question the combination. This issue, having surfaced with his first book on editing, refused to go away.

It may in fact have been reinforced by Bowers's choice of subject for his next series of lectures, because the gist of his Sandars Lectures in 1958 was substantially the same as that of the earlier series. The principal differences were the greater range of illustrative examples (since the nominal topic was no longer exclusively Shakespeare) and the greater attention to the role of textual concerns in literary criticism. The first lecture, "Textual Criticism and the Literary Critic," is a most effective polemic, an expert weaving together of instances—from the literature of four centuries—that show how literary critics (and editors) have been misled by failing to investigate the textual history of the works they were concerned with. It deserves to be considered a classic essay: it remains an ideal introduction to the need for textual criticism in the study of literature, if one is approaching literature from the point of view of authorial intention. Bowers's own sense of its effectiveness may be reflected in the fact that he chose it (but not the other two Sandars lectures) for his 1975 volume of collected essays. When the Times Literary Supplement reported on the Sandars series two weeks later (on 7 February 1958), this was the lecture that seems to have influenced most of the comments—such as the title "The Critics Criticized" and the statement that "in spite of the urbanity of approach which one would expect from a visitor from Charlottesville the critics came in for some rough handling." A further sense of the occasion can be gleaned from another comment that remarks on Bowers's "formidable artillery" in criticizing critics:

Bowers's own version of the latter point appeared in his preface to the published volume, Textual and Literary Criticism, which Cambridge University Press published on 28 February 1959, scarcely more than a year after the lectures were delivered: "A few excisions made in deference to the patience of my hearers have been restored for the more leisurely reader."

In that preface he described the contents of the volume as "something of a mixed bag" but stated a unifying "rationale": "Literary criticism is viewed as directly dependent upon expert textual criticism." The second lecture was devoted to Whitman's revisions for the 1860 edition of Leaves of Grass (and was based on Bowers's research for his 1955 edition of Whitman's manuscripts); this lecture was called by John Carter, in the Gold Medal citation mentioned earlier, a "fascinating detective story" that "no one who heard it or has read it will forget." And the third lecture, "The New Textual Criticism of Shakespeare," traversed the same ground as the Rosenbach Lectures, though with more detailed coverage of the techniques of compositor analysis (by this time there were more to be covered). The salutary insistence that editing is a form of criticism again had pride of place, as in a notable sentence near the end, in which he aligned himself with his old Harvard teacher: "I should prefer the taste and judgement of a Kittredge (wrong as he sometimes was), and of an Alexander, to the unskilled and therefore unscientific operation of a scientific method as if it were the whole answer to the problem and automatically relieved an editor of the necessity to use his critical judgement in any way" (p. 116). The published volume included a fourth lecture, delivered before the Bibliographical Society in London on 23 January 1958, the day after the last of the Sandars Lectures in Cambridge. "Principle and Practice in the Editing of Early Dramatic Texts" explained the rationale for critical old-spelling editions (that is, those with eclectic unmodernized texts), offering a spirited argument for regarding old-spelling texts not as "specialised scholarship" over the heads of students, but rather as "the normal means of reading the literature of an earlier period." The lecture, and thus the book, ended on this note:

This sentence is a fair example of the Bowers style—impassioned, colorful, blunt, flowing, even breathless.

The polemical tone repeatedly troubled reviewers, as it had earlier bothered some members of the audience. Alice Walker (whose work Bowers respected, as he indicated by publishing it in Studies and by conducting extensive correspondence with her in the 1950s) said in the Review of English Studies (1960), "This book . . . is not, as a whole, agreeable work. The status of bibliography is not increased by belittling other studies which play an essential part in textual criticism and it will fall into disrepute if the claims made for it are exaggerated or ill-founded. . . . it is because the latter part of this book seems to jeopardize the status of bibliography that I must join issue with it." She was herself a skillful user of analytical bibliography, but she felt that it "should not be so much all over the place as in these chapters." Similarly, D. F. McKenzie, who was later to express in detail his distrust of analytical bibliography, complained about the tone: "the manner here adopted towards critics seems more calculated to arouse resentment than to encourage patronage" (The Library, 1959). And Joseph S. M. J. Chang, reviewing the 1966 paperback reprint of the book, believed that "One might find his withering contempt for critics who are not particular about the editions they use less objectionable if it was not so singularly aimed at the students of imagery and ambiguity" (Modern Language Journal, 1968). Chang's own critical predilections made him particularly alert to slights directed at formalist critics, but he was right to call attention to a key passage that revealed Bowers's seeming eagerness to attack the New Criticism:

Although these lines are indeed weakened by the bias they reflect and by the superficiality of the contrast between the literary "act of love" and the New Criticism, they are notable for showing Bowers's sense of literary values; his emphasis on physical evidence sprang from—and was not divorced from—his love of literature and his understanding of the place of literature in human life. Several reviewers emphasized such strengths in the lectures while glancing at what they regarded as shortcomings: Herbert Davis, who thought it "a little odd" that Bowers would need to defend textual criticism at Cambridge, nevertheless considered the defense "splendid and impassioned" (Book Collector, 1959); R. A. Foakes, who quarrelled with Bowers's double advocacy of old-spelling editions for scholars and modernized editions for a general audience, believed that Bowers here "puts compendiously into perspective the significance of his own work, and that of the school of bibliographers he has fostered through his editorship of Studies in Bibliography" (Shakespeare Quarterly, 1960); and Bowers's former student Cyrus Hoy, who found the lectures "polemical in tone," also regarded them "as the definitive account of the function of textual bibliography at the present time" (PBSA, 1959)—a judgment that in retrospect seems correct.

Bowers's ability to disturb his audiences continued unabated in his third series of lectures, the Lyell in 1959—judging from John Crow's comments (in the Spring 1965 Book Collector) on the book that derived from them, Bibliography and Textual Criticism (1964). Crow was an ardent admirer of Bowers's approach, and he therefore was disappointed by adverse audience reaction: "I was mystified," he said, "when, at times during the giving of the lectures, I saw eminent heads being shaken." Even Crow, like almost everyone else, felt that Bowers "seems at times to argue too forcibly" and "to be so anxious to prove the value and virtue of Bibliography that he claims too largely for the good woman"; but he nevertheless found the book "admirable and enthralling," "often difficult" but "never dull." Bowers had worked on these lectures—before and after delivery—as one of his activities during his Guggenheim Fellowship of 1958-59. (He later reported to the Foundation that, in addition to making substantial progress on his major project, his Restoration drama bibliography, he had gone "much deeper" into the text of Dryden, used a fortnight to go over galley proof for the last volume of his Dekker edition, and written the "first draft of a book . . . based on a series of lectures at Oxford." He added, "I think I can say that I was not idle.") But he took a considerable time to complete his preparation of the lectures for publication: the foreword is dated 23 February 1963,

four years after they were delivered (and the book did not come out for another year). He made clear, however, that he did not substantially revise them but added nineteen pages of discursive notes as well as many footnotes referring to recent scholarship. One of the reasons for the delay may have been a desire to make specific page citations of Charlton Hinman's The Printing and Proof-Reading of the First Folio of Shakespeare (1963), but in the end he let his book go to press before published copies of Hinman's book were available. He referred to Hinman's work frequently, however, having read the typescript of it in 1957-58 and, indeed, having followed Hinman's exciting research since the days when Hinman was his student. (He recalled "the various discussions we had . . . about the import of the evidence that he was to consider with such superb originality and clarity"; and thirteen years later, in his obituary of Hinman for the Book Collector, he called Hinman's book "the supreme example in this century of analytical bibliography.") The appearance of Hinman's and Bowers's books in two consecutive years symbolizes their status as the linked monuments of the flourishing postwar period of analytical bibliography.

If Bibliography and Textual Criticism is not the masterpiece that Hinman's book (or Bowers's own earlier Principles) is, it nevertheless has a significant place in bibliographical history. That Bowers himself thought highly of it is suggested by his selecting it (rather than either of the two preceding books) to dedicate to his wife. (The Rosenbach volume had been dedicated to Philip Williams, his former student, who had died on 15 March 1955, while the book was in press; and the Sandars volume carried no dedication.) He later referred to the book as "misunderstood" (in a letter to me on 12 December 1972), and the reason is perhaps not far to seek. Those who wished that Bowers would spend more time on original research were naturally not receptive to what may have seemed a rehashing of points made twice before in book form. Bowers acknowledged in the foreword, "I have made no effort to offer fresh original research except in the final lecture," and he stated explicitly that the lectures "were designed for an informed general audience such as is found in a university." The subject was "method alone," and the examples were "discussed only as illustrations of method." The two preceding series of lectures could be described the same way, and both of them could be said to deal with "the nature of bibliographical thinking." Thus one might uncharitably conclude that this book was mere repetition and fail to understand its different focus.

Bowers reported that his original plan had been a "fuller survey" of

the relationship between bibliography and textual criticism, but what came to interest him more was "the nature of bibliographical evidence as applied to textual problems," his feeling of a need to examine "the way in which the bibliographical mind works." As a result this book is more abstract than the preceding two. One might say—oversimplifying to suggest the general tendencies of the three books—that the earlier two are driven by the illustrative examples, whereas in this one abstract concepts are the driving force; and although all three are concerned with the bibliographical basis of editing, the focus of the first two is clearly on editing, whereas the third one is really about analytical bibliography and its role in working out textual genealogy. What it sets out to do is to explore "the nature of the evidence on which textual bibliography operates, the logical forms of its reasoning, the techniques it uses, and the results it can achieve" (p. 7). The first two lectures treat the first two of these topics (and contain the famous, or perhaps notorious, sentence, "When bibliographical and critical judgement clash, the critic must accept the bibliographical findings and somehow come to terms with them" [p. 29]); the next three lectures take up the third topic by examining "The Interpretation of Evidence" under the heads of "three orders of certainty" ("the demonstrable," "the probable," and "the possible"); and the sixth lecture, offering a case study of "The Copy for the Folio Othello," illustrates the results of bibliographical analysis. The book does not attempt to provide for analytical bibliography what the Principles did for descriptive bibliography, and it is thus not a systematic guide to the techniques of analytical bibliography (as Hinman's work in many ways is, though not intended as a textbook). But what it tries to do is even more basic: to furnish a model for thinking about physical evidence and a framework for evaluating conclusions based on such evidence.

The most searching criticisms this book received came from J. K. Walton and Trevor H. Howard-Hill. Walton (in Shakespeare Survey, 1966), besides responding in detail to Bowers's disagreements with his work on the copy for folio Richard III, complained that Bowers "overrates the role of bibliography in establishing genetic relationship." Although Bowers's enthusiasm did sometimes result in overstatements, I think he would not disagree with Walton's point that "we cannot study the history of a text as we would, say, the history of a fossil," because the transmission of a text from one edition to another requires the intervention of a compositor's mind. Bowers was certainly not claiming to eliminate literary judgments from the analysis of textual relationships,

but the insistent tone of his emphasis on those instances in which physical evidence precluded or limited literary arguments does make understandable the kind of response Walton engaged in, which tended, in its turn, to concentrate on the limitations of physical evidence. Howard-Hill (in Shakespeare Studies, 1968) showed himself to be one of those not entirely happy with Bowers's tendency to spend more time as a spokesman for the field than as a research scholar. His two principal criticisms were that Bowers's interest in discussing bibliographical evidence led him to neglect other kinds of evidence and that his treatment of the logical processes of bibliographical method was "simplistic." In regard to the first, Howard-Hill recognized that Bowers's concern was "only with those problems for which bibliographical analysis has, or pretends to have, some relevance." It is the second criticism that is more serious, because many readers may have shared Howard-Hill's "difficulty in deciding whether a simple text-book of logic would not better have served the novice in these studies, for although Professor Bowers writes of 'bibliographical logic,' reflection assures us that there is nothing peculiar to bibliographical evidence which demands a logical treatment different from that of other disciplines in order to obtain valid conclusions." Bowers's rhetoric was no doubt—as in so many other instances—the culprit, for he obviously knew that he was not proposing a distinct kind of logic. His discussion may indeed seem simplistic to those who are experienced in bibliographical analysis; but he was addressing his remarks to the vast majority of literary scholars (including many editors) who did not (and still today do not) understand what "bibliographical evidence" is or know how to think in terms of such evidence. Bowers's lectures attempted to explain how to apply logic—ordinary logic—to an unfamiliar category of evidence, and therein lies their importance.

When, five years later, D. F. McKenzie published his influential article "Printers of the Mind," he was raising questions that Bowers in these lectures had provided the framework for answering. After citing instances of incautious generalizations based on bibliographical analysis, McKenzie reported "a feeling of mild despondency about the prospects for analytical bibliography." Bowers's open-mindedness is shown by his printing this article in Studies, when he had in effect already replied to it. His Lyell Lectures are founded on an understanding of the risks of inductive reasoning and, at the same time, a recognition of the necessity for it, if the great body of physical evidence hidden in books is to be tapped. That such evidence is sometimes incorrectly interpreted, or inappropriately applied to textual criticism, offers no reason for being

any less excited over the prospects for its use; Bowers's book conveys that excitement while outlining a cautious and responsible approach to bibliographical evidence and its place in scholarly editing.

Bowers's campaign to promote bibliographical and textual studies was carried on in a number of other lectures during the 1950s, in addition to the three major series. Although his lectures were adjusted to fit audiences with varying degrees of bibliographical knowledge, nearly all of them were general treatments of the importance and methodology of bibliographical work, bolstered with numerous illustrations. For example, in one of his earliest bibliographical lectures outside Charlottesville, he spoke on "Bibliography and the University" at the University of Pennsylvania on 13 May 1949; in the fall of that year (9 September 1949) he chose the topic "Some Relations of Bibliography to Editorial Problems" for the English Institute at Columbia University; on 9 May 1952 he addressed a joint meeting in Charlottesville of the Bibliographical Society of America and the Bibliographical Society of the University of Virginia on "Bibliography, Pure Bibliography, and Literary Studies"; six months later (18 November 1952), in London, he took up "Purposes of Descriptive Bibliography" before the Bibliographical Society; in the spring of that 1952-53 Fulbright year he delivered a lecture at Bedford College, University of London, on "Shakespeare's Text and the Bibliographical Method" (4 March 1953); in the spring of 1954 he gave the Phi Beta Kappa address at Randolph-Macon College on "Scholarship, Research, and the Undergraduate Teacher"; on 17 September 1954 he addressed the English Institute on "McKerrow's Editorial Principles for Shakespeare Reconsidered"; at the 1957 Modern Language Association meeting in Madison, Wisconsin, he read a paper on "Old-Spelling Editions of Dramatic Texts" (10 September 1957); when the Grolier Club met in Charlottesville (with the Bibliographical Society of the University of Virginia) on 25 October 1958, he spoke on "Textual Criticism and the Literary Critics"; and on 14 November 1958 he delivered, at Charlton Hinman's university, the sixth University of Kansas Annual Public Lecture on Books and Bibliography, entitling it "The Bibliographical Way" (published as a pamphlet the next year). Several times during the 1950s he also came to the aid of his own Bibliographical Society, at Virginia, delivering not only the inaugural paper in 1947 but such later ones as "Textual and Literary Study" (9 April 1957) and "The Growth of Whitman's Leaves of Grass as Shown by Its Manuscripts" (10 April 1958).

Most of these lectures were later published, and altogether in the

fifteen years after the war he published some five dozen bibliographical articles, along with nearly two dozen reviews. Among the most significant were two of the earliest, his "Notes on Standing Type in Elizabethan Printing" in Papers of the Bibliographical Society of America in 1946 and "An Examination of the Method of Proof Correction in Lear," in The Library the next year (a paper extended into "Elizabethan Proofing" for the 1948 festschrift for Joseph Quincy Adams and regarded as standard for some years and as "ingenious"—to use Peter Blayney's 1982 adjective, in The Texts of King Lear and Their Origins—even when questioned). Another important piece was "The Problem of the Variant Forme in a Facsimile Edition" (The Library, 1952), which underlay his highly critical review (Modern Philology, 1955) of the Yale facsimile of the Shakespeare First Folio and which set forth questions first addressed in a facsimile in Hinman's Norton Facsimile of the First Folio (1968). One other article that attracted considerable attention was "Hamlet's 'Sullied' or 'Solid' Flesh: A Bibliographical Case-History" in Shakespeare Survey for 1956. Bowers entered the tangled debate over this famous crux in order to point out that the Second Quarto's "sallied" (1.2.129) and "sallies" (2.1.39) were set by different compositors and printed on different presses and thus to "prove on physical evidence" that no connection between the two existed—allowing for the conclusion that both words are correct (i.e., that the meaning in each case is "sully" and that "sally" is an acceptable variant spelling). The article did not end the controversy but from then on played a prominent role in it. His short piece on The Wits in Strickland Gibson's A Bibliography of Francis Kirkman (1949) deserves brief mention as an indication of some of the work that Bowers was doing in connection with his bibliography of Restoration drama: Gibson's preface noted that Bowers was engaged in an examination of Kirkman's piracies and that his chapter on The Wits was a part of an uncompleted larger study.

Although Bowers was primarily recognized for his bibliographical and textual work, he continued to write critical essays and published during this period a dozen of them. One that became particularly well known was "Hamlet as Minister and Scourge" (read in December 1954 at the Shakespeare Section of the Modern Language Association convention and published in PMLA in 1955), which took Hamlet's self-assessment as "scourge and minister" to be his "clearest analysis . . . of his own predicament"; Bowers explained Hamlet's procrastination through an examination of the Elizabethan understanding of "scourge" (the damned agent of private revenge) and "minister" (the divinely appointed agent of public benefaction). Three years later, in the issue of PMLA celebrating

the seventy-fifth anniversary of the Modern Language Association, this article was named as one of the eight most "outstanding and influential" PMLA articles on Shakespeare: one of three anonymous Shakespeare specialists called on for reports named it, describing it as "both very learned and very sensible." Even Greg, who had severely criticized Bowers's style in the Rosenbach Lectures, said that this article "can be read with pleasure." And David Bevington represented many others when, years later, he testified to how "deeply" the essay had influenced his thinking (he is quoted on the jacket of Bowers's 1989 volume of Shakespeare essays).

On 4 March 1959, the day before Bowers arrived in England on his first Guggenheim Fellowship, Greg died at the age of eighty-three, and from that moment Bowers was unquestionably the most prominent bibliographical and textual scholar in the English-speaking world. Bowers's obituary of Greg in The Library was full of warm praise for his "Logical thinking, great thoroughness, and . . . original turn of mind operating on the frontiers of knowledge":

This moving tribute offers an insight into the influence Greg had on Bowers; it can also be taken as a fair statement of the way many scholars since then have felt about Bowers's own presence. As the 1950s ended, Bowers was the leader of his field, having earned that position by a remarkable fifteen years' work that included the writing of a classic book, the founding and editing of a seminal journal, the delivery of three major lecture series, and the production of two landmark editions.