Last Update: May 2, 2000         Navigation:   Main Menu     Vol II (1969)     Next Article

[page 1, column 1:]

The State of Poe Bibliography

G. Thomas Tanselle

It has become a commonplace, when Poe studies are surveyed, to lament the small amount of useful bibliographical work which has been devoted to Poe. In this respect Poe is hardly different from the other major nineteenth-century American writers, few of whom have been provided with anything approaching satisfactory full-scale bibliographies. What does distinguish Poe bibliography is the fact that two elaborate attempts have been made to produce detailed primary bibliographies — and both of them are failures. The scarcity of the early printings of Poe's work has caused dealers and collectors over the years to give considerable attention to the complicated problems of Poe bibliography; but it is unfortunate that so much interest has resulted in so little solid accomplishment. To J. Albert Robbins' description of Poe, in the first number of this journal, as a "scholarly and critical orphan," one may add that he is also a "bibliographical orphan."

    At present, the state of the secondary bibliography of Poe is more encouraging than that of the primary. Until quite recently, however, Poe scholars had few specialized lists of secondary material to turn to and had to rely on the PMLA, MHRA, and AL bibliographies, on Leary and Woodress, on CHAL and LHUS. Jay B. Hubbell's brief section on bibliography in his essay on Poe for Eight American Authors (1956) revealed the paucity of work in this area. Since that time, several important advances have been made. For current items, Robbins' annual essay on nineteenth-century poetry in American Literary Scholarship has furnished a critical evaluation of Poe scholarship since 1963, and Richard P. Benton's checklists have provided comprehensive coverage since 1962 (in the Emerson Society Quarterly in 1965 and 1967, and now in the Poe Newsletter, beginning with the current number). For earlier material, John Carl Miller's John Henry Ingram's Poe Collection at the University of Virginia (1960) includes the first extensive published listing of nineteenth-century criticism since Killis Campbell's CHAL bibliography (1918). A more comprehensive — but unpublished — list, including the nineteenth-century items and bringing the coverage through 1941, is I. B. Cauthen's 1942 Master's thesis at the University of Virginia. Twenty years later J. Lasley Dameron, for his doctoral dissertation at the University of Tennessee, extended Cauthen's record through [column 2:] 1960 (and described Poe's critical reception since 1928, the point at which Dudley R. Hutcherson's earlier dissertation on Poe's reputation had stopped); this annotated checklist, covering 1942-1960, was published in 1966 by the Bibliographical Society of the University of Virginia. Dameron and Cauthen have now joined forces and are together adding items to their previous lists in the preparation of a complete bibliography of Poe criticism (including foreign material) from 1827 to the present. When this work appears, it will become the central research tool in Poe scholarship, and the problems of Poe secondary bibliography will have been essentially solved; inevitably obscure references will continue to turn up after its publication (and the Poe Newsletter offers the appropriate place in which they can be recorded), but the basic job will be done once and for all. In addition, G. R. Thompson's forthcoming list of dissertations and theses on Poe promises definitive coverage of that area of scholarship, for it will contain both abstracts and critical comments. The present number of PN — with its annotated lists of published scholarship, of fugitive references, and of dissertations — provides (with Benton's two previous lists) more thorough coverage of the last eight years than of any previous period. It is only through annotated listings of this kind — and the much-needed attention given to European research by Claude Richard — that scholars can find their way through the rapidly increasing mass of material with any efficiency and discrimination.

    If there is good reason to be optimistic about the immediate future of Poe secondary bibliography, there is no corresponding evidence that the problems of the primary bibliography are going to be solved soon. It is true that much research has been directed toward matters of the Poe canon, and articles continually appear announcing new facts about Poe's appearances in periodicals, gift-books, and anthologies; but no bibliography exists which brings all this information together, along with physical descriptions — based on detailed bibliographical analysis — of Poe's publications in book form. A number of useful contributions to Poe bibliography have been issued in connection with editions or with exhibitions and catalogues of individual collections. Thus such editions as James A. Harrison's (Virginia Edition, 1902), Killis Campbell's (poems, 1917; tales, 1927), Edward H. O'Neill's (Borzoi Edition, 1946), and Floyd Stovall's (poems, 1965) are important achievements in their synthesis of bibliographical data, but they are not intended to be full-dress bibliographies. Similarly, such catalogues as the Stephen H. Wakeman sale catalogue (1924), John D. Gordan's characteristically annotated Berg exhibition catalogue (1949), Carroll A. Wilson's Thirteen Author Collections of the [page 2:] Nineteenth Century (1950), Richard Gimbel's full catalogue of the 1959 Yale exhibit (YULG, XXXIII, 139-189), and David A. Randall's account of the J. K. Lilly collection in the Indiana University Bookman in 1960 (IV, 46-58) — excellent as they are — cannot substitute for a complete bibliography. Other lists — such as the two 1941 compilations, the Index to Early American Periodical Literature, 1728-1870: Part 2, Edgar Allan Poe and John Cook Wyllie's "A List of the Texts of Poe's Tales" (in Humanistic Studies in Honor of John Calvin Metcalf ) — are necessarily limited in scope. T. O. Mabbott announced, at least as early as 1922 (see LJ, XLVII, 1086-88), his intention to produce a complete bibliography of Poe's writings, but it had not appeared at the time of his death in 1968; even if notes for such a bibliography remain among his papers, any attempt to edit them carefully and perform the necessary checking of additional copies would be a lengthy process, and thus no new Poe bibliography is to be expected from this source in the near future. Jacob Blanck, in his Bibliography of American Literature, is gradually working his way toward Poe, having now arrived alphabetically at Longfellow. When his section on Poe appears, scholars will finally have a reliable descriptive bibliography of Poe's books and appearances in books — but they will still not have the full-length bibliography that is needed. The BAL, which omits periodical contributions and employs an abbreviated form of description, is not designed to take the place of more detailed book-length bibliographies of the major authors. Nevertheless, the appearance of the BAL volume containing Poe will mark an immense step forward in Poe primary bibliography; but its publication cannot realistically be expected to occur for a considerable time.

    Meanwhile, we are left with the two bibliographies currently available, both of them products of a flurry of bibliographical activity on Poe in the 1930's. In 1932 Kenneth Rede and Charles F. Heartman put together a "Census of First Editions and Source Materials by Edgar Allan Poe in American Collections," published serially in Heartman's American Book Collector (I, 45-49, 80-84, 143-147, 207-211, 339-343; II, 28-36, 141-153, 232-234, 338-342) and then separately in two volumes. Two years later John W. Robertson published his Bibliography of the Writings of Edgar A. Poe, beautifully printed by the Grabhorn Press, with a second volume of "Commentary" on the bibliography in the first. In 1940 the Heartman-Rede census was extensively revised by Heartman and James R. Canny, and the resulting Bibliography of First Printings of the Writings of Edgar Allan Poe was issued as No. 53 of Heartman's Historical Series (a revised edition of which appeared in 1943). These two bibliographies — the Robertson and the Heartman in its last version — are obviously the products of great industry and enthusiasm, but they are also undependable and poorly conceived. David A. Randall reviewed each of them in Publishers' Weekly at the time of their publication; the Robertson bibliography he found "practically useless" (CXXV [21 April 1934], 1540-43), and the Heartman-Canny he described as "amateurish, inaccurate and speculative" (CXXXVIII [30 Nov. 1940], 2033-38). In each case Randall went on to support these judgments with detailed discussion of specific weaknesses and errors, ranging [column 2:] from a demonstration that Robertson was not abreast of the scholarship to a thorough examination of the Heartman-Canny treatment of the Raven and Tales volumes of 1845. The two reviews are outstanding examples of the kind of informed criticism to which bibliographies should be subjected; no one who attempts to use these bibliographies should neglect to consult Randall's reviews at the same time.

    In general terms, one may point out four ways in which the bibliographies are deficient. First is the underlying conception of the function of a bibliography, as reflected in the record of copies examined. The Robertson bibliography is essentially based on the Robertson collection, and, while a catalogue of an important collection has its uses, it must not be confused with a bibliography — for a bibliography is a documented history of the forms in which an author's works have appeared, and such a history cannot be limited to the particular copies in one collection any more than other kinds of history can be limited to the materials available in one library. The Heartman-Canny bibliography, having grown out of the Heartman-Rede census, does give a partial list of the locations of copies, but it proceeds gratuitously to point out, "It is really not within the scope of a bibliography to give a census of existing copies" (p. 16). One can agree that a bibliography need not provide a census merely for the sake of the census, but it is absolutely essential that a bibliographer document his findings by reporting the locations of the particular copies he has examined. The Heartman-Canny bibliography gives no indication of which of the listed copies formed the basis for the descriptions; in some cases no locations at all are recorded, and one finds instead the useless statement, "Copies are owned by many institutions and collectors" (e.g., p. 129). If a bibliography is to fulfill its function as a history of the physical forms of an author's work, it must represent a thorough examination of as many as possible of the surviving copies; one can never know what variants may be lurking in the next unexamined copy. Any successful piece of research must be based on a comprehensive knowledge of the material; since the bibliographer is dealing with physical forms, the body of material relevant to his purposes includes every surviving copy of the books with which he is concerned. In practice, of course, it is not always feasible to examine every copy, as in the case of modern books that exist in numerous copies; but several copies should be examined in every instance, and, for scarce books, it can be said that the nearer a bibliographer's list of examined copies approaches a census of surviving copies, the closer he has come to fulfilling his responsibility. If there had been more articles like William B. Todd's analysis (in the Fall 1961 number of Library Chronicle of the University of Texas [VII, i, 13-17]) of "The Early Issues of Poe's Tales (1845)" — in which eighteen copies are examined — the state of Poe bibliography would be much beyond where it now stands.

    A second area of difficulty is the content of the descriptions themselves. Quasi-facsimile transcriptions of title pages are indispensable, and photographic reproductions do not eliminate the necessity for them. Neither bibliography presents title-page transcriptions that conform to current standards, and Robertson makes a futile attempt [page 3:] in his transcriptions to suggest the spacing of the originals. As for the collation paragraph, neither bibliography provides a collation by gatherings; in each, "Collation" consists simply of the leaf size, the total number of pages, and a listing of the contents. However, a notation indicating the collation of gatherings (preferably in the Greg-Bowers style) should be the central fact of a bibliographical description, for a knowledge of the physical structure of the book is basic to any bibliographical analysis. Furthermore, the two bibliographies give only scanty attention to bindings and inserted advertisements and no attention whatever to typography and paper — all of which are obviously important elements in the physical makeup of a book. Lacking all these details, the bibliographies fail to provide necessary information not only to the Poe scholar but also to the historian of publishing.

    A third kind of limitation is that both books are concerned primarily with first printings of Poe's works (as the Heartman-Canny makes clear in its title). Later printings, however, may contain important textual variants, and any edition or impression published during an author's lifetime is of potential textual significance. Such editions and impressions (English as well as American) therefore deserve detailed examination also; still later editions and impressions (both English and American) need not be described as fully but should nevertheless be recorded, since they form part of the history of the publication of the texts involved and furnish essential data for a study of the author's reputation. In 1953 Francis B. Dedmond compiled "A Check-List of Edgar Allan Poe's Works in Book Form Published in the British Isles" (BB, XXI, 16-20), but this sort of information needs to be incorporated in a full-scale bibliography, arranged (with expanded descriptions) so that the history of various sets of plates, as they passed from one publisher to another, becomes clear. For a writer like Poe, the publication history of his works in their original language is complicated enough that it is probably wiser to leave the translations for a separate volume; but a complete bibliography of the translations is a desideratum and, in the case of Poe, is a particularly fascinating project. Some attention has been given in the past to the translations of Poe — John E. Englekirk's 1937 list of Mexican translations (PMLA, LII, 524-525) is an example — but a single comprehensive bibliography needs to be provided.
 
    Finally, the two bibliographies are deficient in their treatment of textual matters. Heartman-Canny scarcely touches the subject, though some variant readings are mentioned within chatty notes; Robertson, on the other hand, includes long extracts from Poe's reviews and, in his volume of "Commentary," prints different versions of many poems in parallel columns. The one gives too little attention to the text; the other is on the verge of becoming an edition. Just how detailed a record of textual variants is necessary or appropriate in a descriptive bibliography has always been a difficult problem. A descriptive bibliography is not expected to establish a critical text; at the same time, a descriptive bibliographer must investigate textual variants if he is adequately to distinguish the impressions, issues, or states of a given edition. The research for a critical edition and the research for a descriptive bibliography are interdependent, a fact currently recognized [column 2:] by the editors of several of the editions being prepared under the auspices of the MLA Center for Editions of American Authors. The Robertson and Heartman-Canny bibliographies, inadequate even at the time of their publication, are unquestionably outmoded today.

    These two bibliographies stem from the old tradition which conceived of author-bibliography as a record of "first editions" for the use of collectors. Fortunately collectors and scholars are increasingly recognizing the fact that their interests and needs, far from being divergent, actually coincide: scholars are beginning to understand the importance of bibliographical analysis in the establishment of texts, and collectors are beginning to recognize the significance of printings later than the first. Whether modern textual research on Poe will stimulate the production of a descriptive bibliography remains to be seen. At the moment, one can only repeat David Randall's comment of thirty-five years ago: "There is a very pressing need for a definitive bibliography of Poe." If the state of Poe secondary bibliography is hopeful indeed, the prospect for the primary bibliography is not entirely bleak — the BAL is in progress, and both the current interest in Poe (manifested in this Newsletter) and the growing understanding of the relationship between editing and descriptive bibliography are encouraging signs.

[S:1 - PSDR, 1969]