 | |  |
Printing History and Other History
by
G. Thomas Tanselle
[*]
History is a subject about which everyone seems to have a heated opinion, including the determination to ignore it. Fashions in the way history is approached are a product of history itself, with one set of attitudes toward the past, and the place of the past in the present, succeeding another. The study of history, or the use of history in intellectual endeavor, has periodically been out of vogue among prominent critics of the arts; but the perennial attempts to reject history, in one form or another, show how insistently present, how inescapable, history always is. To the human mind, it appears to be a ubiquitous concern, for we bring past associations to our surroundings. G. M. Trevelyan summed up this phenomenon when he mused that "once, on this earth, once, on this familiar spot of ground, walked other men and women, as actual as we are to-day, thinking their own thoughts, swayed by their own passions, but now all gone."[1] A related point was made, no less eloquently, by Noël Coward in his song "London Pride" (1941): "Cockney feet mark the beat of history. / Ev'ry street pins a memory down."[2]
Indeed, Coward's reference to "memory" may be taken as a significant enlargement of the concept of historical associationism, adding personal experience to acquired knowledge. One can bring to one's current surroundings a sense of the distant past, developed through learning the so-called facts about the occurrences that took place there over the centuries. But even if one has never been exposed to this information, or has no interest in acquiring it, one's memory is constantly at work,
But our engagements with our surroundings are complex transactions, in which what we bring is mixed, in varying proportions, with what we encounter. We are surrounded by what we take to be physical masses, material objects; some are products of nature, some of human artifice, but all are from the past, from some moment prior to the present one. Our individual associations and temperaments may make each of us see these entities differently; but the masses themselves—however obscurely their objective status may be identified—do seem to contribute their share in setting the associations in motion. If, as R. G. Collingwood has said, the aim of historical study is "thought" or "mental activity,"[3] our means of getting at it must nevertheless be principally the examination of physical objects; for thought is not tangible, and the thought of the past resides mainly in the artifacts it produced, since artifacts, being physical, have a chance of surviving, though perhaps in altered form, from one moment or era to another. Oral traditions, which form the other link with past thought, should be examined as well; but they may often be suspected of reflecting a greater change through the passage of time than physical objects usually undergo.
Thus one may say that the study of human history is largely the study of physical objects,[4] for the double reason that they both reflect and
Of the rich variety of artifacts, I wish to concentrate on one category, printed items with verbal texts, as a way of looking at material culture. These printed items—books, pamphlets, newspapers, and so-called ephemera—are certainly one of the largest classes of artifacts from the past five hundred years, and possibly the most influential class.[6] The
These fields have long been recognized specialties within the overarching field of History with a capital H. So have many others, of course: economic history, diplomatic history, intellectual history, cultural history, biography, the histories of literature, of music, of architecture, and so on. It is true that not long ago the history of the book would not have been seen as on a par with these inquiries; people who stated their interest as book history—and there have always been such people—would have been regarded as narrow specialists obsessed with a tangential topic. What they were doing was history, yes, but almost as far removed from real history as the work of amateur genealogists pursuing their family
The development of the new kind of book history is in general, therefore, a cause for celebration. But its rise in stature has not brought about a greater appreciation of all the studies that it necessarily comprises. The history of printing and typography, which one might logically assume to be a fundamental part of book history, does not seem to have acquired greater glamour within the new approach, and one even hears disparaging hints about its narrowness. Some superb scholarship has been produced in a number of countries during the past century in the areas of the history of letterforms, of typefounding, and of printing; but there has been a tendency on the part of some of the newer book historians to associate Britain, and the English-speaking countries generally, with the study of the physical production of books, and then to contrast the French school of histoire du livre and its broad social concerns with a supposed English school of more strictly technological history. Although this view of the so-called English approach is unfair, the association of the English with the study of books as physical objects probably arose from the fact that they were indeed the pioneers in what is now called analytical bibliography—the analysis of the physical evidence in books for clues that reveal information about the printing process—and in the application of this information to literary study.[8]
There is an irony in the fact that this field, analytical bibliography, has never been fully understood or accepted by the more traditional printing historians. Perhaps its development in connection with the textual criticism and scholarly editing of literary works has caused it to seem a branch of literary studies. In any case, analytical bibliography has remained on the fringes of printing history, just as printing history
The first step in the argument is to note that texts are affected by the printing process. This simple, but profoundly important, point has remained outside the ken of most readers (even highly perceptive ones) over the centuries. The reasons for this neglect, along with an account of the emerging interest from the mid-nineteenth century onward (at least on the part of a few scholars) in the relationship between text and object, constitute one of the most fascinating stories of intellectual history. It is largely unsung, however, because the point at its core is still not widely understood; and thus the accomplishment of that band of British scholars who were the forerunners and founders of the Bibliographical Society is not properly appreciated.[9] Yet what they set in motion was a revolution in the way we approach printed texts, a revolution still in progress and unlikely to be completed for a long time, judging from the slow pace of its inroads into entrenched patterns of thinking. (I should say, parenthetically, that any prognostication about attitudes toward printed texts in the future is likely to seem quixotic in an age full of predictions about the demise of the printed book; but of
The revolution instigated by the pioneers of the "New Bibliography" consisted of approaching artifacts that have words in or on them as one would approach other artifacts—treating them as objects produced by artisans skilled in particular crafts, as objects that can be studied for physical details revelatory of their own production history. These bibliographers' view was in contrast to the notion, still prevalent, that books are merely the containers of texts, which can readily be extracted without the necessity of paying much attention to the containers themselves. The approach also necessitated a recognition of the individuality of every copy, in opposition to the common assumption that copies from the same edition are identical. Examining books as objects suggests to most people an interest in the graphic arts; what is less readily understood is that an interest in the texts of books also requires the reading of physical evidence (which, because it is physical, may vary from copy to copy).
Even sophisticated historians of material culture sometimes have difficulty regarding books as just another category of artifact, for the presence of words in them tends to suggest that they can speak to us in a more direct way—in our own language, as it were.[11] When words are
When Henry Bradshaw in the nineteenth century observed a correlation between textual divisions and paper stocks (or gatherings of sheets) in certain incunables, he was noting a connection between printing and text;[12] so is a present-day editor who discovers that a blank-verse passage in an Elizabethan play was set as prose at a point where the space would not otherwise have held the pre-assigned text.[13] Some recent students of
Another, more visible, residue of the printing process is also always present: the design of the pages. Every book, whether by default or by careful planning, has a design, created by the letterforms, their sizes,
That problem is not a simple one, however. For even if everyone granted that the details of what happened in the printing shop are a fundamental ingredient in the study of the dissemination of ideas in printed form, there would remain the question of how those details can be known. And on that issue there is no unanimity of opinion among printing historians themselves. The primary reason for this situation is uncertainty over the nature and respectability of analytical bibliography. Some printing historians have neglected it on the assumption that it is a branch of literary study, of little relevance to printing history; others have actively rejected it because they believe it to be suspect as history. The first response results from a simple misconception; the second raises the basic question of what constitutes validity in historical study.
As to the first, one need only observe that printing history clearly encompasses information on such topics as how many compositors were at work on a given book, whether they set the type pages in numerical order (or according to the sequence in which the pages would be placed on the press), and how proofreading and correcting were accomplished.[19] Generalizations about printing-shop procedures in particular periods and places can be made by combining such data from large numbers of individual books. And these data are precisely what literary scholars
If printing historians and other historians have frequently shared a misunderstanding of what analytical bibliography attempts to accomplish, they also have shared a skeptical view of the procedures it follows. In the first place, historians are inclined to prefer archival records—manuscript and typescript documents—over printed books when they are available. Obviously printing historians do know that to discuss typography and layout they must look at the actual printed items, just as the newer historians of reading understand why the visual appearance of printed pages is a key class of data for their work. But when historians wish to know something about how books were produced, rather than how they finally looked, the books themselves are not routinely thought of as an archival source, presumably because of the number of inferences often required in the attempt to coax their own stories out of them. If a printer's ledger records the format of a book, or the amount of paper needed for the edition, or the number of compositors who set the type, or the number of presses used for the printing, those pieces of information are conveyed as direct (or relatively direct) statements, whereas extracting the same details from the finished books would entail inferences, often a network of inferences. Archival records are frequently regarded as "primary" documents, and printed books as "secondary"; but primariness is relative to the subject of the inquiry, and when the subject is printing-shop procedures, the evidence from the objects that resulted from those procedures (objects that have survived and are available for first-hand examination) must take precedence over statements about the procedures in other contemporary documents.
Printers' and publishers' records are valuable, of course, but no one
Elementary this point may be; but a tendency to assume that archival records are correct is curiously persistent. One telling example is J. D. Fleeman's investigation of William Somervile's The Chace (1735) in the light of the surviving ledgers of its printer, William Bowyer. He notes several discrepancies between the record of presswork in the ledgers and the press figures printed in the book—one of which is that both formes of sheet A are listed in the column for press number 3 in the ledger, whereas the figure 8 appears in printed copies on a page of the outer forme of this sheet (the verso of the second leaf), and the inner forme bears no press figure. Fleeman does not suggest the possibility that the ledger is incorrect. Instead he concludes that figure 8 here must refer to the pressmen named in the ledger as operating press number 3 (Fowle and Davis); the other two occurrences of 8 in the book, however, he attributes to a different pair of pressmen (Mazemore and Jethro, named in the ledger as operating press number 8), because the two formes involved are credited to press number 8 in the ledger. In other words, the assumption that the ledger is correct forces Fleeman to conclude that the printed figures designate pressmen and that their signification can shift—a conclusion that leaves one wondering what purpose the printed figures serve, if the ledger is the only true (and thus necessary) guide to the allocation of presswork. A simpler hypothesis that maintains the accuracy of the ledger is to regard the printed figure 8 in sheet A as an
A similar instance is Robert Darnton's discussion of what he calls "hidden editions" of Diderot's Encyclopédie.[23] He examines the wage book of the printer of the quarto Encyclopédie to learn "what actually happened in the printing shops" (p. 78), and he makes no attempt to verify the data from that record by reference to surviving copies. In other words, "what actually happened" in the printing process can be known, he implies, independently of analyzing the actually surviving product of that process. Indeed, he explicitly rejects such analysis: "It would be vain to draw inferences about the printing process from the examination of actual copies of the book, because the copies must vary endlessly" (p. 81) —as if complexity were grounds for refusing to examine evidence.[24] It is hard to imagine any other class of available evidence that a historian would so willingly ignore.
A more realistic attitude toward bibliographical evidence and its
The essence of any inductive process, such as the search for scientific "laws" or the pursuit of the past, is uncertainty; and what are called facts—scientific or historical—are simply conjectures that seem relatively certain (that is, relatively unlikely to be overturned by contrary evidence) to a substantial number of informed observers. (The historian's conclusions are testable in the same way that the scientist's are: by repeated reenactments of the events that led to the conclusions, whether those events took place in a laboratory or in a library.)[26] If the validity of analytical bibliographers' hypotheses is frequently questioned, those bibliographers are in no different position from other historians, or other sifters of inductive, and thus effectively infinite, evidence. Yet sometimes bibliographical analysis is criticized for involving the same inferential process that is taken for granted in other branches of history. A notable example, which goes to the heart of the issue, is D. F. McKenzie's statement that compositorial analysis has exhibited "virtuosity in discovering patterns in evidence which is entirely internal, if not wholly fictional."[27]
What we agree to call historical knowledge is built up by the accretion of individual acts of pattern-finding, some of which invalidate previous acts and some of which confirm and extend them. It is true that each act must be interpreted in the light of whatever relevant context has already been formulated; any "internal" analysis that does not at some point recognize appropriate available sources "external" to it can of course be criticized, for we should obviously use all relevant data that we can locate. But McKenzie's belittling of analyses based on evidence internal to books, if not corroborated by evidence external to them, ultimately fails to acknowledge the limited nature of all bodies of evidence, however broadly conceived. In the first place, documents external to books hold just as many pitfalls for interpretation as do the books themselves, and any conclusions based solely on such evidence are potentially flawed for the same reasons that conclusions based on any other single body of evidence are. Beyond that, one does not in many instances have the option of checking books against printers' ledgers, for the survival of printer's archives is not common. One has to use whatever is available and interpret it in the light of what is already considered established. McKenzie overstates the authority of the latter, however, when he says that bibliographical analysis "depends absolutely upon antecedent historical knowledge" (p. 2): such knowledge (as in every other field) is itself always an hypothesis, and the new analysis may modify it. This kind of interplay, after all, is what produced the antecedent knowledge, and each new analysis plays a role in determining what is regarded as established by the next analyst. In the end, all evidence is "internal," for sooner or later one reaches a point where there is nothing outside to relate to. We are trapped in Wallace Stevens's "island solitude, unsponsored, free, / Of that wide water, inescapable." Virtuosity in making sense of what we have is thus the only procedure available to us. The degree of ingenuity
Analytical bibliography seems to me a perfect paradigm of historical scholarship because it illustrates, with particular clarity, how dependent our view of the past is upon the creativity of those who engage in historical investigations. A current instance is Paul Needham's hypothesis that three works were printed by the Catholicon Press of Mainz in ca. 1459-60 (and reprinted later) from two-line slugs, cast from a setting of movable type, rather than being printed from the type itself. This startling thesis is an example of creative thinking on the part of a learned bibliographer who argues with careful deliberation and logical rigor. Whether it is accepted as "true" depends on whether it can stand up under criticism. Since its first announcement in 1982, it has in fact been criticized, and an alternative explanation has been proposed for the puzzles raised by these editions. Needham has shown, to the satisfaction of many but not all, why the direct criticisms of his position have missed their mark and why the alternative thesis does not adequately account for the observed evidence. His hypothesis is well on its way toward acceptance as a significant fact in the history of early printing. Needham fully understands the nature of this process and has on more than one occasion invited his readers to evaluate the controversy as an exercise in clear thinking (the requirements of which are themselves a human invention, of course). In 1993, for example, he said, "The issues in question do not depend in any way on a specific knowledge of incunables, but should be accessible to anyone familiar with the fundamental principles of analytical bibliography."[29] If a consensus forms in support of
The idea that history is revised by each generation, that what we see in the past is contingent on who we are in the present, is of course not new; most historians by this time surely subscribe to it in some form, and it is a natural element in any skeptical or antifoundational philosophy. As Charles Gullans wrote in his poem "Research," "It is ourselves we summon from the past."[30] Yet surprisingly it is still easy to arouse controversy by embracing the subjective element in historical writing, to say nothing of openly accepting the role of creativity in it. A prominent recent illustration is the celebrity achieved by Simon Schama, in both the journalistic and the academic press, from proclaiming this view in lectures and in Dead Certainties (1991), a book that mixes archival data and invented details in retelling the divergent accounts of the deaths of General James Wolfe (1759) and Dr. George Parkman (1849). The teasing subtitle of the book, placed in parentheses, is Unwarranted Speculations, a phrase that accurately conveys what some readers feel the book contains; but its parenthetical apposition to the
Some reviewers were predictably bothered by this position and felt that it presented an overly negative outlook. But even persons sympathetic with Schama's general view may find him strangely reserved here, for in his "Afterword" he repeatedly undercuts his own relativistic remarks. He says, for example, that he does not "scorn the boundary between fact and fiction"; perhaps not, but what he does in the body of his work is to show brilliantly that such a boundary is always in flux. Similarly, he calls the narratives in his book "works of the imagination, not scholarship" (p. 320); yet presumably one of his purposes is to raise questions about how scholarship is defined, as in his assertion that scholarly reports involve "the inventive faculty." He states as an article of faith "the rather banal axiom that claims for historical knowledge must always be fatally circumscribed by the character and prejudices of its narrator" (p. 322). But why "fatally"? The banality of the observation, if it is banal, arises from its restating of a universal condition, one that can be taken for granted. In that case, the hint of a complaint underlying the word "fatally" is pointless. Historical study is not prevented by our inherent subjectivity; the limit that human perception places on verification is simply a given in everything we do, and we proceed from there.[31]
An equally fundamental point is that, even if we could be objective in reporting observed data, what has survived for us to inspect is inevitably fragmentary and requires supplementing by informed intuition. A creative account written by a learned historian may bring us closer to the past than documents alone. Whether we believe that it does so in any given instance depends on our evaluation of the process by which the account is constructed. Setting standards for responsible and irresponsible argument is not in any way incompatible with recognizing the
In the approaching years, as we are constantly being told, electronic texts may become the dominant form of visible language.[33] Whether the study of the processes of producing these texts will be carried out under the rubric of "printing history" is less important than seeing the continuity from handwriting to letterpress to offset to electronic digitization. In his Centenary Lecture to the Bibliographical Society in London, D. F. McKenzie appropriately called for the Society to include in its domain the electronic production and dissemination of texts. Indeed, if bibliography is thought of as the field that deals with the processes resulting in the placement of texts in physical objects,[34] along with
The history of the production, use, and influence of letterforms is fundamental to humanistic study not only because letterforms are one of the dominant vehicles for the transmission of language but also because their shapes and deployment—as with all artifactual details—convey
E. M. Forster, in his memoir of Edward Carpenter, said that Carpenter would not "figure in history," because his life did not produce the kind of "words and deeds" that attract chroniclers; but Carpenter's genius, according to Forster, was to make those who came "under his spell" realize how inadequately the human spirit has been recorded by historians.[37] Forster's observations, both on Carpenter and on historical tradition, are themselves a part of history through their embodiment in visible language printed on the pages of a book (and subject to the vicissitudes of the printing process). Our best chance of capturing the human spirit, as it has existed in different times and places, is through studying the artifacts it has produced, reading their significance in the light of how they came into being. Printing history is essential for examining a major class of those artifacts by helping us to decipher, in the fullest way possible, the physical marks that constitute verbal messages from the past.
Notes
This paper was delivered as the J. Ben Lieberman Memorial Lecture of the American Printing History Association during Rare Book School at the University of Virginia, 18 July 1994. It was also delivered at the University of Toronto on 28 September 1994 as the inaugural lecture of the Toronto Centre for the Book.
"Autobiography of an Historian," in An Autobiography & Other Essays (1949), p. 13. He made a similar statement in a 1948 piece, "Stray Thoughts on History," included in the same volume: "Here, long before us, dwelt folk as real as we are to-day, now utterly vanished, as we in our turn shall vanish" (p. 82).
John Gielgud, in a 1993 interview reported by Mel Gussow, said "Every street in London is full of memory" (New York Times, 28 October 1993, p. C1).
The Idea of History (1946), pp. 305, 306. Earlier he states that the "proper task" of historians is "penetrating to the thought of the agents whose acts they are studying"; history is "the re-enactment of past thought in the historian's mind" (p. 228).
An historian who has made this point the basis for his approach to history is Arthur Bestor, who said that the historian "recovers the past . . . by carefully studying a group of objects or artifacts that exist in the present, and drawing logical conclusions therefrom about the particular past event which must have produced, or shaped, or at least put its mark upon these observable things"; this statement occurs in the opening paragraph of "History as Verifiable Knowledge: The Logic of Historical Inquiry and Explanation," in Research Methods in Librarianship: Historical and Bibliographical Methods in Library Research, ed. Rolland E. Stevens (1971), pp. 106-127.
A useful introduction is Jules David Prown's "Mind in Matter: An Introduction to Material Culture Theory and Method," Winterthur Portfolio, 17 (1982), 1-19 (with a "Selective Bibliography" on pp. 16-19). For some indication of the difficulty historians have had with this approach, see note 11 below.
I do not underestimate the importance of handwriting as a means of transmitting visible language; even after the advent of printing from movable types, handwriting continued, and has continued to the present, to be indispensable in conveying verbal messages from one person to another. But for a significant part of the past five centuries, printed books were the primary means for disseminating verbal works, both new and old. What I say later about the connections between texts and the physical means of their production applies equally to manuscripts.
See Karen J. Winkler, "In Electronic Age, Scholars Are Drawn to Study of Print," Chronicle of Higher Education, 14 July 1993, pp. A6-A8. The Society for the History of Authorship, Reading, and Publishing was founded in 1991 and has published the SHARP Newsletter since then; its first conference was held in New York on 9-11 June 1993 (and is reported in Printing History, 34 [1993], 71-101).
On national trends in book history, see John Feather, "Cross-Channel Currents: Historical Bibliography and L'Histoire du Livre," Library, 6th ser., 2 (1980), 1-15. A collection of essays in this field is Books and Society in History, ed. Kenneth E. Carpenter (1983), which includes "A Statement on the History of the Book" (signed by the participants in the 1980 conference of the Rare Books and Manuscripts Section of the Association of College and Research Libraries) and Robert Darnton's "What Is the History of Books?" (In my introduction to this volume, I attempted to show the relation of analytical bibliography to histoire du livre, as I also did in my Hanes Lecture, The History of Books as a Field of Study [1981].) A recent theoretical overview is Thomas R. Adams and Nicolas Barker's "A New Model for the Study of the Book," in A Potencie of Life: Books in Society, ed. Nicolas Barker (1993), pp. 5-43.
The best account of the development of analytical bibliography is F. P. Wilson's "Shakespeare and the 'New Bibliography,'" in The Bibliographical Society 1892-1942: Studies in Retrospect (1945), pp. 76-135 (reprinted as a separate volume, edited by Helen Gardner, in 1970); the early history of the Bibliographical Society is recounted by F. C. Francis in the same voume, pp. 1-22, and by Julian Roberts in "The Bibliographical Society as a Band of Pioneers," in Pioneers in Bibliography, ed. Robin Myers and Michael Harris (1988), pp. 86-100. See also my "Physical Bibliography in the Twentieth Century," in Books, Manuscripts, and the History of Medicine: Essays on the Fiftieth Anniversary of the Osler Library, ed. Philip M. Teigen (1982), pp. 55-79, or "The Evolving Role of Bibliography, 1884-1984," in Books and Prints, Past and Future: Papers Presented at The Grolier Club Centennial Convocation (1984), pp. 15-31. The fullest listing of examples of bibliographical analysis is in my Introduction to Bibliography: Seminar Syllabus (periodically revised; see, for example, pp. 135-162 of the 1994 revision).
On this general subject, see my "Reproductions and Scholarship," Studies in Bibliography, 42 (1989), 25-54, and "The Latest Forms of Book-Burning," Common Knowledge, 2.3 (Winter 1993), 172-177.
Jules David Prown, for example, in his important essay "Mind in Matter" (see note 5 above), unwittingly illustrates the difficulty people have had in thinking about objects containing verbal texts; when, near the end, he says that artifacts "tell us something, but facts are transmitted better by verbal documents," he is failing to recognize the physicality of visible texts or to acknowledge that such texts are artifacts themselves, requiring the same kind of examination and interpretation as other artifacts. The neglect of this point on the part of many historians has caused them to believe that verbal texts speak to them more directly than do physical objects containing no words and thus to ignore the physical aspects of "verbal documents." (The common use of the word "artifact" to refer only to objects without verbal texts reflects the mistaken assumption that "verbal documents" are in a class apart from other objects.) This line of thinking has often led historians not merely to undervalue physical evidence but actively to disparage its use. Thirty years ago John Chavis criticized this attitude in a piece for Curator (the magazine of the American Museum of Natural History) entitled "The Artifact and the Study of History" (7 [1964], 156-162), deploring "the willingness of the academic historian to do without the utilization of the artifact." To represent the historians' position, he summarized William B. Hesseltine's "The Challenge of the Artifact," in The Present World of History, ed. James H. Rodabaugh (1959): although Hesseltine recognizes that verbal documents preserve physical evidence, he thinks of it as "external" to what the text "says" and naïvely believes that one cannot determine what nonverbal objects say. Chavis contrasted this view with E. McClung Fleming's in "Early American Decorative Arts as Social Documents," Mississippi Valley Historical Review, 45 (1958-59), 276-284: Fleming believes that the artifact is "a social document" and regrets that "the historian has tended to ignore this primary source in his preoccupation with printed and manuscript materials" (p. 276). Some years later Fleming developed his point in "Artifact Study: A Proposed Model," Winterthur Portfolio, 9 (1974), 153-173 (in which he again laments the fact that the "use of material culture by historians and social scientists is minimal" [p. 154]). Similarly, Henry Glassie has said, "Because of his commitment to the primacy of print, the historian has been unable to produce an authentic history"; see "Archaeology and Folklore: Common Anxieties, Common Hopes," in Historical Archaeology and the Importance of Material Things, ed. Leland G. Ferguson (1979), pp. 23-35 (quotation from p. 29). It is noteworthy, however, that even these cogent calls for greater attention to physical evidence do not explicitly place books and manuscripts in the world of artifacts. Anyone who wishes to encourage use of the physical evidence that resides in objects with verbal texts thus faces a double obstacle: not only the neglect of physical evidence in general but also the failure to regard verbal documents as artifacts. A notable exception to this common failure is Arthur Bestor's "History as Verifiable Knowledge" (see note 4 above): "a document is simply a special kind of artifact. The marks that were impressed upon it by past events are intelligible marks, produced as the result of the operations of a mind like our own." Furthermore, Bestor effectively makes the point that logically follows: however persuasive the verbal text may be, "the historian must not be misled into imagining that it speaks to him with such authority as to relieve him of the obligation of carrying out a critical and logical inquiry of his own" (pp. 108-109).
For an excellent account of Bradshaw's concern with the "relation of text to structure," see Paul Needham's Hanes Lecture, The Bradshaw Method (1988), esp. pp. 5-6, 10-12.
The classic examination of the Elizabethan compositors' practice of "setting by formes" (rather than following the numerical order of the pages) is Charlton Hinman's "Cast-Off Copy for the First Folio of Shakespeare," Shakespeare Quarterly, 6 (1955), 259-273. An admirable brief treatment is provided by Peter W. M. Blayney in The First Folio of Shakespeare (1991), pp. 12-14. For some history of the bibliographical analysis of setting by formes, see my "Analytical Bibliography and Renaissance Printing History," Printing History, 3.1 (1981), 24-33.
On "frisket bite," see Walter J. Partridge, "The Type-Setting and Printing of the Mainz Catholicon," Book Collector, 35 (1986), 21-52 (esp. pp. 42-44); and Paul Needham, "Slipped Lines in the Mainz Catholicon: A Second Opinion," Gutenberg Jahrbuch, 1993, pp. 25-29 (esp. p. 29).
See Peter L. Shillingsburg, "Detecting the Use of Stereotype Plates," Editorial Quarterly, 1.1 (1975), 2-3. Two illustrations of such damage are figures 3 and 6 appended to the condensed reprinting of my article on "The Use of Type Damage as Evidence in Bibliographical Description" in Journal of Typographic Research, 3 (1969), 259-276.
For a reproduction of the corrected state of the text, see Charlton Hinman's The Norton Fascimile: The First Folio of Shakespeare (1968), p. 737. (In Hinman's "through line numbering," the variant—"Looke where he haue [not] crown'd dead Cassius"—occurs in line 2587.) A list of substantive variants in copies of the First Folio appears on pp. xxi-xxii. The large number of press-variants of all kinds (those in punctuation and spelling as well as in substantives) are listed and discussed in Hinman's The Printing and Proof-Reading of the First Folio of Shakespeare (1963), pp. 226-334.
This variant is discussed in the Northwestern-Newberry Edition of Typee, ed. Harrison Hayford, Hershel Parker, and G. T. Tanselle (1968), pp. 310-311. (The point of variation—"religious attendants of the grove[s]"—occurs on p. 101, line 31, of the 1846 edition, and on p. 91, line 37, of the Northwestern-Newberry Edition.) See also my "Textual Study and Literary Judgment," Papers of the Bibliographical Society of America, 65 (1971), 109-122 (esp. pp. 115-116), reprinted in Textual Criticism and Scholarly Editing (1990), pp. 325-337 (see pp. 330-331).
For an excellent introduction to this subject, see David McKitterick, "Old Faces and New Acquaintances: Typography and the Association of Ideas," PBSA, 87 (1993), 163-186. He points out that "bibliographical analysis has much more to offer the social and enumerative historian [of reading] than, so far, seems to have been demonstrated" (pp. 164-165).
That analytical bibliography furnishes information for printing history was the point of my article entitled "Analytical Bibliography and Renaissance Printing History" (see note 13 above). And in 1987 I emphasized analytical bibliography in outlining an agenda for printing history—see "Thoughts on Research in Printing History," Printing History, 9.2 (1987), 24-25.
"Paper Chase," Metropolis, 4.4 (November 1984), 24-26, 32 (quotation from p. 32). (Prudon supervised the restoration of the Woolworth Building.)
A Life of Learning (the Haskins Lecture of the American Council of Learned Societies, 1994), p. 2 Cf. Arthur Bestor's comment quoted at the end of note 11 above and John Lancaster's quoted in note 22 below.
See "William Somervile's 'The Chace,' 1735," PBSA, 58 (1964), 1-7 (which was the first bibliographical article to make use of the Bowyer ledgers). Fleeman's discussion of other discrepancies follows the same pattern that he uses for the one in sheet A: thus his assumption that the attribution of pressmen in the ledgers is correct forces him to state that the Long Primer figure 3, which appears in the inner forme of C and in outer E and G, signifies one pair of pressmen in the first instance and a different pair in the other two instances; similarly, he claims that the larger figure 3 designates one pair of pressmen in outer H and a different pair in M, N, and O (and, further, that this latter pair of pressmen was the same pair that used the Long Primer figure 3 in inner C). Because Fleeman does not ask what reason there could be for such a shifting use of figures, it is ironic that he concludes, "without the benefit of the printing records kept by Bowyer, a good many serious misinterpretations would have been the result of the usual kinds of analysis." For information on the ledgers themselves, see the admirable work by Keith Maslen and John Lancaster, The Bowyer Ledgers (1991); their entry for The Chace is no. 2156, p. 170. Lancaster called attention to the Somervile example in his address to the Grolier Club on 15 June 1993, during his exhibition marking the publication of the ledgers. That address, in its published form in Gazette of the Grolier Club, n.s., 45 (1993), 63-81, notes Fleeman's "premise that the ledgers are accurate" and provides an appropriate warning: "the ledgers —like most human documents—are not infallible, and must be used with caution like any other historical evidence" (p. 68). I am grateful to Lancaster for providing me with the materials for examining this instance of discrepancies between ledger entries and the printed item to which they refer.
"A Bibliographical Imbroglio: Hidden Editions of the 'Encyclopédie,'" in Cinq Siècles d'Imprimerie Genevoise, ed. Jean-Daniel Candaux and Bernard Lescaze (1981), pp. 71-101.
The variation among copies would not prevent, for example, the attempt to determine whether the pages supposedly set by different compositors do in fact display different characteristics. Darnton's comments throughout on the confounding of bibliographers by the randomness of the process by which sheets were mixed together in individual copies displays a lack of understanding of bibliography as a form of history; the goal of descriptive bibliography is not to present neat collational formulas or to seek uniformity among copies but to report as accurately as possible the true situation.
See "The Dunciad in Four Books and the Bibliography of Pope," PBSA, 83 (1989), 293-310. The points in question are Bowyer's references to cancelled and reprinted sheets and leaves (pp. 302-303).
The events that lead to historians' conclusions are examinations of documents and other artifacts. Some elaboration of this point, showing how scientific investigation involves historical method, can be found in Arthur Bestor's "History as Verifiable Knowledge" (see note 4 above), esp. pp. 111-113.
See McKenzie's Panizzi Lectures. Bibliography and the Sociology of Texts (1986), p. 7. I have discussed these lectures, and this passage, further in "Textual Criticism and Literary Sociology," SB, 44 (1991), 83-143 (esp. pp. 87-99). (McKenzie has been expressing doubts about analytical bibliography since at least 1969, when he published "Printers of the Mind" in SB, 22: 1-75; for responses to that essay, see Peter Davison's "Science, Method, and the Textual Critic," SB, 25 [1972], 1-28, and my "Bibliography and Science," SB, 27 [1974], 55-89 [esp. pp. 73-78].) In a later lecture, "What's Past is Prologue": The Bibliographical Society and History of the Book (1993), McKenzie continues to show a bias against analytical bibliography, referring to A. W. Pollard's activities as "sensitive to the primacy of trade documents as historical evidence" (p. 12) and asserting, "Enumerative bibliography in Britain opened up riches . . . well beyond the reach of descriptive and analytical bibliography" (p. 17)—as if he thinks of these approaches as rivals, competing with each other. He does, however, also express a recognition of the importance of physical evidence and offers a useful statement of the "premise" underlying analytical bibliography: "that the forms [of books] themselves encode the history of their production" (p. 24).
Cf. Richard Rorty's concept of "conversation" in Philosophy and the Mirror of Nature (1979): e.g., the idea that "objective truth" is "the normal result of normal discourse" (p. 377). See also such statements in his Consequences of Pragmatism (1982) as the proposition that there is "no rigorous argumentation that is not obedience to our own conventions" (p. xlii).
"Slipped Lines in the Mainz Catholicon: A Second Opinion" (see note 14 above), p. 25. Cf. the following comment in his "Corrective Notes on the Date of the Catholicon Press" (Gutenberg Jahrbuch, 1990, pp. 46-64): "The arguments involved in these preliminary issues do not depend on bibliographical knowledge. I hope they will be accessible to every reader who is even peripherally aware of the Catholicon 'question'" (p. 61). Lotte Hellinga, in the same volume of the Gutenberg Jahrbuch ("Comments on Paul Needham's Notes," pp. 65-69), makes this unconsidered remark: "The more elaborate our arguments and the more they are confined to what is presented by the printed books themselves, the greater the need to recognize these inevitable limitations of our perception, and to be alert to the possibilities that lie beyond them" (p. 69). The last part of her sentence is undeniably correct: we must always be mindful of the "limitations of our perception" and the "possibilities that lie beyond them." But the "need to recognize" these points is equally present in all situations and is not "greater" when arguments are "more elaborate" or "confined" to the evidence found within printed books. The elaborateness with which one argues and the sources of evidence one employs do not in themselves make arguments questionable; what is crucial to the acceptance of an argument is the care with which one goes about collecting and evaluating the evidence (whatever it is) and then constructing inferences from it (however elaborate they may be). The controversy between Needham and Hellinga should be examined by all who are interested in the process of scholarly argumentation by which historical facts come to be established. See (in addition to the articles already mentioned) Needham, "Johan Gutenberg and the Catholicon Press," PBSA, 76 (1982), 395-456; Hellinga, "Analytical Bibliography and the Study of Early Printed Books with a Case-Study of the Mainz Catholicon," Gutenberg Jahrbuch, 1989, pp. 47-96; Needham, "Further Corrective Notes on the Date of the Catholicon Press," Gutenberg Jahrbuch, 1991, pp. 101-126; Hellinga, "Proof for the Date of Printing of the Mainz Catholicon," Bulletin du bibliophile, 1991, 1:143-147; Hellinga, "Slipped Lines and Fallen Type in the Mainz Catholicon," Gutenberg Jahrbuch, 1992, pp. 35-40; Hellinga, "Eltville and Mainz: A Tale of Two Compositors," Book Collector, 41 (1992), 28-54; Needham, "Mainz and Eltville: The True Tale of Three Compositors," Bulletin du bibliophile, 1992, 2:257-304. (See also the Partridge article cited in note 14 above and Needham's reply, "The Type-Setting of the Mainz Catholicon: A Reply to W. J. Partridge," Book Collector, 35 [1986], 293-304.)
Letter from Los Angeles (1990), pp. 6-7. More than half a century ago G. M. Trevelyan called it "a common opinion" (which, he said, "I myself share") that "the historian's work is partly scientific, partly artistic"; see "Bias in History," in An Autobiography & Other Essays (1949), p. 81. There have of course been many discussions of the subjectivity of historical scholarship. Two recent treatments (which include the history of such discussions) are Peter Novick, That Noble Dream: The "Objectivity Question" and the American Historical Profession (1988) and Joyce Appleby, Lynn Hunt, and Margaret Jacob, Telling the Truth about History (1994). (See also note 32 below.)
Although Schama has somewhat clouded the reasons for such a view of the bases for proceeding, his last line does seem to me to be in this positive spirit: if, he says, "our flickering glimpses of dead worlds fall far short of ghostly immersion, that perhaps is still enough to be going on with." But he could have made clearer, I think, that "enough" has no connection with quantity; whatever we have must always be "enough," if we are to do any thinking at all. He is right to speak of the "gap separating a lived event and its subsequent narration"—that is, historians' "unavoidable remoteness from their subjects" (p. 320). (And he makes eloquent use of James's The Sense of the Past in this regard.) But his wavering over the relation between imagination and scholarship in the effort to build a bridge to the past does not provide a satisfying preparation for a positive conclusion.
Quoted from a 1991 lecture at the New-York Historical Society in Peter Stevenson, "Lights, Camera . . . Schama!", New York Observer, 27 September 1993, p. 15. Another historian, an economic historian, who has recently received considerable attention for stressing the importance of literary methods in historical writing is Donald N. McCloskey. In The Rhetoric of Economics (1985) and If You're So Smart: The Narrative of Economic Expertise (1990), he argues that narratives and metaphors are an indispensable part of all "factual" discourse, whether in history or in science. James Atlas, in an article occasioned by a 1991 conference in Albany on the writing of nonfiction, says that the truth of a narration is not simply a matter of "facts" (which are themselves "slippery"): "The truth is in the prose, the style, the quality of representation that compels us to believe" ("Stranger Than Fiction," New York Times Magazine, 23 June 1991, pp. 22-23, 41, 43 [quotation from p. 43]). The way was paved for this kind of discussion by Hayden White's pioneering and influential Metahistory (1973); two recent examples are his The Content of the Form: Narrative Discourse and Historical Representation (1987) and Lionel Gossman's Between History and Literature (1990).
The phrase "visible language" has been given some prominence by its use as the title of a journal (on matters relating to typography, handwriting, and reading) now published by the Rhode Island School of Design.
I use this phrase in order to encompass such texts as those of motion-picture films and sound recordings as well as of objects (like books) that display written language. My specific concern in this paragraph, as in the essay as a whole, is with the latter, but I recognize that the study of textual transmission involves other kinds of objects as well.
At the end of "What's Past is Prologue" (see note 27 above), McKenzie seems to think that, if the field of bibliography embraces (as he hopes it will) "the dynamics of the increasingly volatile texts of our new age," the move will "represent a radical departure" from the earlier concept of bibliography based on "the primacy of the physical artefact (and the evidence it bears of its own making)" (p. 29). But the electronic processing of texts, like the earlier methods of processing them, uses physical means to produce physical texts. (How could it be otherwise?) McKenzie refers to the capacity of computers for "modelling" as if this characteristic were new to the history of text-producing processes; but all of the technologies that have made the reproduction of texts easier—such as movable type or xerography—have also made easier the process of using old texts to generate new ones. One thinks of stop-press corrections and the alteration of standing type for a new impression, for example, or cutting and pasting followed by xerographic copying. We have lived for a long time with technology that encourages the proliferation of variant texts; and the latest instance, computerization, marks a change of degree, not of kind. It has tremendously eased the process of textual manipulation, but it has not altered the fact that visible texts are the products of physical routines, which leave their traces in the texts. McKenzie's suggestion that bibliography in the electronic age "could come to deal, less in specific manifestations of a work, than in the formulae for their realisation" is like saying of earlier bibliography that it could focus on printing technology rather than on printed texts. Of course the technology is always important in its own right, but it is also always important for understanding the resulting texts, whether they are on paper or computer tapes and disks. The future study of texts will have to take new technology into account, but it will be dealing with an old situation.
It is useful to recall, in this connection, the title of a pair of lectures delivered by Alvin Eisenman at Dartmouth on 28-29 October 1992: "Printing as Memory." This wording implies not only the idea of printed texts as the memory of human thought but also the role of printing, as a technical and artistic activity, in the memorial process.
| | |