Once Only Imagined
KK: Joe, your conviction regarding the necessity of drawing artists into the conversation on Blake dates back to 1982 at least, when you wrote the following as part of your contribution to the special issue of SiR: "Art and literary critics should stay in contact with artists . . . if we do not include the workshop and artist in our conversations about Blake's art, I think we will continue to remark: 'This is not what pictorial art is all about' " (405). In bringing your own studio arts background to bear on your—and our—understanding of Blake over the years, you've helped us see what Blake's pictorial art is all about. Blake and the Idea of the Book has helped refresh the perception of many literary scholars who, though they may have wrestled with the meaning or iconography of the illuminated books, hadn't previously thought to peer behind the images to their execution on copper or transference to paper. By recreating Blake's workshop with its pigments, inks, rags, brushes, and glues, you've brilliantly illuminated the material dimension of Blake's artistry.
Continuing in this vein, I'd like it if you could talk some about how the artist-practitioner in you has influenced the facsimilist, both before and after 1992, when you set out with Morris and Bob to build a comprehensive electronic archive of Blake's work. Having spent a good deal of time over the last six years watching you hash out issues of digital reproduction with John Unsworth, director of the Institute for Advanced Technology in the Humanities; Matt Kirschenbaum and Andrea Laue, technical editor and project manager of the WBA, respectively; IATH programmers; and your fellow editors, I've become attuned to some of the ways in which the hand and eye of Viscomi the artist overlap with the hand and eye of Viscomi the facsimilist. Partly it's a page taken from the book of history: illusionary achievements in visual reproduction have time and again profited from an artist's knowledge of materials and processes. Weighing in the balance which intaglio process—mezzotint or stipple—will more accurately reproduce an oil portrait painting is the natural prerogative of the artist and technician alike. I would venture to say your acute sense of the limitations of digital media to imitate all characteristics of Blake's hand-colored relief prints owes much to your direct contact and experience with copper, acid, brushes, and ink. In what other ways do the artist and facsimilist in you co-mingle?
JV: My interest in facsimiles began in 1975 while assisting on an exhibition of nineteenth-century paper toys and theatres. Technically, these were etchings and large lithographs, with each sheet containing many parts, figures, or stages. Children would cut out the parts and color and assemble them into the toy or theatrical scene. We had a few original models assembled, but mostly we had uncut sheets, which were visually interesting in themselves—at least to me—but required the constructions to really make their point. So I selected various sheets from different periods and countries, had them photomechanically reproduced, and proceeded to play the industrious child, cutting and coloring and gluing to make the models. Oh yes—I also aged the paper and colors so the models looked authentic.
Working at the Museum of the City of New York, with eighteenth- and nineteenth-century English and American books, toys, prints, drawings, and paper theatres was great training in material culture. I didn't expect it would also provide training in forgery, but it did. I learned a few valuable lessons working on that show: a facsimile intends to deceive legally, an unacknowledged facsimile can easily become a forgery, and an undetected forgery is an original. For me, these dubious distinctions between original and copy commented interestingly on Benjamin's idea of the "aura"—or historical authenticity—of the original. The "aura" is created not by the object but by the belief in the object as authentic and unique.
Two years later I began my recreations of Blake's illuminated prints. I wanted to reproduce his prints by reproducing his production process, by using the tools, materials, and processes he did. You don't make relief etchings by hand, though, if you intend to reproduce the model exactly. For that you need photography, and the resulting plate does not even need to be a relief etching. I found that out a year later when I discovered two forgeries in a monochrome copy of America. The two plates were lithographs with faked embossments; the images easily fooled the eye (they had been reproduced in books before) but not the hand. Because of my recreations, I could tell that the paper was wrong and that led me to examine the images more closely. Their ink was too flat and they were slightly elongated along the diagonals, signifying the projection of a negative onto a sensitized zinc plate. I don't know if the two pages were produced with the intention to deceive or as facsimiles to complete a copy missing these pages, but when the copy changed hands, they were not documented and hence were taken as Blake's prints. I do know that I was able to detect these fakes and a number of others over the years—as well as authenticating prints as Blake's—because I was also able to make them or envision making them. Redrawing Blake's designs on copper forces you to see the subtlest events in the originals, because, as any drawing master from Blake's day would tell you, drawing is the art of seeing. It trains your eye from looking to seeing, or, as Blake would put it, from general forms to minute particulars, and "Unless. You Consult. Particulars. You Cannot. even Know or See Mich: Ang. or Rafael or any Thing Else" (anno. to Reynolds, E 645). Such direct consultation, I would hope, can only help an editor and scholar.
Being able to make and print plates that looked and felt like Blake's may have sensitized my eye but it also spoiled me. I expected facsimiles of Blake to be works of art in their own right while remaining true to the original, like those produced in 1983 by the Manchester Etching Workshop. I expected the best of them to recreate the initial reading experience. What a shock it was to see digital reproductions on IATH monitors in 1993, when Morris, Bob, and I visited IATH for the first time. With the Web in its infancy (the Mosaic Web browser was just then morphing into Netscape 1.0.), computer monitors were designed for word processing and not graphics. Their size, resolution, number of colors, and color management system were not impressive, to say the least. The reproduction was way off in color and size and looked different on each monitor, and I thought, "Hell no, I'm out of here. I'm not signing on to a project whose reproductions are no better than color photocopies." It took John Unsworth, the director of IATH, and a high-end monitor designed for graphics to reassure me that such display problems would soon be resolved for the general, non-professional user. Planning for the future while working within the present material and physical limitations of our medium has been a hallmark of the project—and at times a difficult juggling act, technically, editorially, and mentally.
Monitors have come a very long way since then, as have scanners, computers, storage systems, and bandwidth. But the best digital facsimile is not going to get confused with the original or require—or reward—the "suspension of disbelief" of printed facsimiles. I made this simple point a few years back in a lecture at the NASSR conference in Halifax, on Romanticism and the New Technology. For a lecture on Blake and reproductions, I brought Bob's impression from plate 4 of Urizen copy G—or that is what I claimed it was. It was actually a doctored-up color photocopy. At the end of the lecture, I took it out of its gold frame and mat to demonstrate how an original would be examined bibliographically—and proceeded to "accidentally" spill a glass of water on it. The audience gasped in horror. Then I told them the truth: for one hour that copy had the "aura" of the original, something a digital facsimile could never have. And I miss that; I think humans are hard-wired to touch things, to know with our hands. But that is a different kind of knowledge and experience, and the image you can't touch can be more accurate to the original image and provide more visual information than any printed reproductions. And I think this raises one of the crucial issues addressed by the Archive: when do you need to consult the original and when will a reproduction suffice? I think the answer has changed radically because of the Archive.
The Archive's 100dpi images are displayed in an applet that enables the viewer to receive them true size, which is why we refer to them as digital facsimiles, and they are connected to 300dpi enlargements. Neither can recreate the original reading experience, nor can they reproduce the material artifact. But then, the best of the printed facsimiles do not fully reproduce the artifact either. As Bob and I show in our color-printing article, Blake's plates were rarely well-aligned on the page; they fall, slant, rise, or are too close to the edge, but this registration information is unrecorded because reproductions are cropped to the image and facsimiles are centered on the sheet. Misregistrations, along with the lack of uniformity in plate size and margins, are, in effect, edited out. Nor do printed facsimiles reproduce the original condition of an illuminated book, as it left the "Printing house in Hell," but rather, at best, a nicely bound version of the book as it was once "socially situated" in a collector's library. The Archive does not reproduce the full sheet either; it is cropped to the image, albeit for technical rather than aesthetic reasons. Cropping allows us to get the image as large as possible on the reproductive source (usually a 4 x 5 inch color transparency), allows most illuminated book images to be displayed to size, minimizes scrolling, and keeps file sizes smaller. Cropping to the image is yet another compromise to the materiality of our medium, to monitor sizes, storage capacities, bandwidths, and reproductive sources. Recognizing that we are not alone in making this editorial decision and that our medium is not transparent makes it easier to accept the limitations of digital reproductions and to focus on their strengths.
KK: I'm curious to know if the path of influence runs both ways: has your academic work on Blake in turn shaped your development as an artist over the years?
JV: Around 1977, when I began studying Blake seriously, I had already been painting and drawing for over ten years but making prints for only a few years, mostly book-size etchings, relief etchings, and woodcuts. I had my own small press, which enabled me to experiment with monoprints and color printing. So, I was already working small and in media that Blake used, though I did not know that. Nonetheless, there would be an influence, but not of the obvious kind. I didn't start to combine words and images, illustrate poems, draw Blakean subjects, or develop a more linear style. I did, however, rethink my idea of what a print was and abandoned the idea of it needing to be exactly repeatable. I stopped numbering prints in editions and started to focus more on image-making rather than printmaking, using etchings or relief etchings as basic matrixes but not as models I had to duplicate in paper. In effect, my prints became monoprints, which I'd continue to work up in watercolors and/or pastels. I wasn't bothered by the variations. In fact, creating drawings in series that are basically variations on a theme dominates my art to this day. I used to think this was from having played in bands that improvised a great deal (or needing to do something till I got it right), but your question makes me realize that it started not long after I began studying Blake.
If you were to place on a table all the impressions printed from one of Blake's illuminated plates, you would have a series of images that are basically variations of the plate's design; you would also see how an image or motif can evolve through its production, that is, how execution can generate invention. The works at the end of the series differ significantly from those at the beginning but could not have been reached without all the intermediate works. What for Blake was a historical process, with overt changes the inevitable result of change in production styles, is for me collapsed into a three to six month period, with changes very deliberately evoked. For example, I am currently working on three related series of drawings of fruit with violin, and recently finished a series of fifty drawings of a still life that I think of as "six circles in a basket." In these series, no two drawings are alike; they differ in colors, textures, and medium, but rarely in size (14 x 17") and only minimally in composition. Last year I did six series of drawings of various articles of clothing, from twenty to fifty in a series, produced one after the other, each one suggesting the subsequent one. These drawings, many of which are true size or larger, range from being very sketchy to more representational and yet most look more like color prints or serigraphs (silk screens) than drawings because the colors are flat and subordinate to strong black lines. In regards to Blake, the violin series is probably the more interesting, because these drawings are a kind of print taken from relief outlines (figure 1). I make a key drawing and then with glues and resin build it up into relief, like a collotype, but instead of printing it, I take a rubbing of it to transfer the outline (the technique is called "frottage"), which I then work up in various media and usually combine four to a frame. The drawings in all of these series, though, however they were executed, differ from what I was doing in the 80s, which were smaller, more abstract, almost minimalist, but even then, when I first began making more drawings than prints, I liked working up the same theme over and over again in different media, from graphite to charcoal to hard and soft pastels to oil pastels to watercolors and oil paint and combinations of these media. It is repetition without duplication, and with variations and visual effects created and encouraged by the manner and materials of execution.
Blake made prints that look like drawings; I like making drawings in the spirit of print production that often look like prints. I guess Blake's largest impact on me personally is his unorthodox ideas of print and his experiments at combining drawing and printmaking.
KK: Joe, perhaps the connections I want to make here are too forced—you can let me know—but the printmaking metaphor lying behind your drawing experiments takes on added dimensions and theoretical interest when considered in conjunction with Bob's comments a little further down on the virtual concept of "copy" that Dynaweb, the Archive's search engine and display tool, imposes on categories like "drawings" and "paintings." It seems to me that each of the objects in your series based on a still life, for example, could with a little license be designated as a "copy" (copy 1 of 50, copy 2 of 50 . . . copy 50 of 50). But of course those are your drawings and metaphors—not Blake's. I realize I'm setting myself up for accusations of methodological fallacies, the most obvious of which is anachronism, in projecting your practices onto his, but bear with me. As you point out, Blake modeled relief etching on drawing, taking up brush and pen rather than the tools of engraving to execute his designs. It's a topic to which you give dedicated space in Blake and the Idea of the Book. Is there any evidence, Joe, that Blake ever put the metaphor in reverse, in effect transposing its two parts, tenor (printing) and vehicle (drawing)? Relief etching is like drawing, but drawing and painting are also like printing? Blake's monotype prints of 1795—the so-called "color-printed drawings"—come to mind, for example, for their fusion of drawing, painting, and printmaking paradigms.
One of the things that has always provided grist for Morris's intellectual mill is the intensity of Blake's metaphoric imagination. In his introduction to the Cambridge Companion, he writes that Blake was "by nature it seems . . . a synthesizer whose electrified senses tended to experience, because they desired to experience, everything in terms of everything else, to see all channels of life as the tributaries of one vast waterway." How, then, did that synthetic insight influence Blake's thinking about the various media in which he worked? Did he see a convergence of all media? If so, maybe Dynaweb isn't such a procrustean bed after all—not when it comes to Blake.
ME: May I interrupt? As far as synthesis and Blake's illuminated-printing process are concerned, I wouldn't say that he took up drawing tools rather than engraving tools. It wasn't either/or but both. Of course, he needed the basic materials of etching—copperplates, stopout mixtures, acids—to make his texts and images reproducible. But even in writing the texts and drawing the images on the plate, while he used stopout varnishes as if they were ink for brushes and pens and he used copper as if it were paper in some respects, he also used burins, etching needles on the same copperplates. He saw the possibilities for relief in plates designed for intaglio and the possibilities for intaglio in plates designed for relief work. His mind wasn't the mind of a printmaker or the mind of a painter (or, for that matter, the mind of a poet) but a fusion of all. I suspect that this characteristic was both a matter of natural talent and cultivated skill reinforced by attitude (he thought he should be that way, and he was determined to be). Though he seems to have been comfortable with simple combinations of rudimentary tools, materials, and procedures—technical solutions that required intricate mechanical manipulation and elaborate coordinations don't seem to have appealed to him—he was resistant to narrow constructions of his task that would force him into a single groove (I guess that's a hiphop metaphor converted into an engraving metaphor). But he was also a technician proud of his inventions and discoveries. The pride comes through in his announcement of illuminated printing and its advantages in 1793 (E 692-693), his recipes for engraving and "woodcutting" on pewter (E 694), and his self-identification as the inventor of "W Blake's Original Stereotype . . .1788" at the end of The Ghost of Abel (1822), one of his last illuminated works (pl. 2, E 272).
But none of that changes the fact that Dynaweb is a procrustean bed. But so is XML, so are all tools and materials, including pen, ink, and paper: they allow you to do some things, they force you to do other things, and they keep you from doing yet other things. That is the price of admission—we must pay to play.
JV: I'm sure Morris has heard the phrase, "No Representation without Taxation." Every medium comes with a price, and that of course includes Blake's as well as ours. "Pay to play" indeed. We are outlining some of the costs—and some of the benefits they pay for—in this interview.
Blake's experiencing everything in terms of everything else might explain why he can so easily move from graphic art to painting and drawing and back again in his Public Address and most of his other commentaries about art. What he says about one appears equally true of the others. I am not sure, though, if Blake's sense of equality among media, which I assumed was an insight born of practice, constitutes a vision of media convergence. If there is a convergence, then I suppose it would be similar to how "all religions are one," that is to say, maybe all arts are one in that they all come from the same source, the "poetic genius," with their differences reflecting material limitations or the medium's natural language—the "tax"—rather than inherent values or rankings.
Theoretically, Blake reduced all art to drawing, as he explicitly states numerous times: he who draws best is the best artist and engraving is drawing on copper and painting is drawing on canvas and nothing else. This is not literally true, of course, but by reducing all art making to drawing, as verb and noun, that is, to the inventing process and to the product defined by strong lines, Blake eliminated the grounds for valuing one medium over another. Just as in genre, there was a hierarchy in media: oil painting was above watercolors and both were above engravings—or works on canvas over works on paper over works on copper. Reducing all art making to drawing was Blake's way of leveling the playing field, of removing the taint of craft from his work as a printmaker. This is a very smart strategy for a printmaker who hopes to raise his status from craftsman to artist and have his original prints taken seriously. But maybe what Morris calls his "synthetic imagination" was playing its part as well, and the vision of equality among the arts was more than theory born of defensiveness or practice.
In any event, in practice, the pull was "upward," with his relief-etched prints and color prints moving to the status of the unique, autographic work on paper and away from the mechanical and multiple. Works in the Small and Large Book of Designs and, as you mention, the large color-print drawings are the beautiful results of combining printing and painting with finishing in strong pen and ink outlines. The direction or influence was mostly one way but not exclusively. His willingness to experiment in graphic art to create visual effects that exploited the tools and language unique to that medium, as in "woodcut-on-pewter" and white-line etchings, or simulated alla-prima painting, as in color prints, spilled over into painting, as is evinced by the experimental paintings in the Descriptive Catalogue, such as the Spiritual Form of Nelson, and his so-called "tempera" and "fresco" paintings, which capture the visual effects of color printing. His attention to "minute particulars," as in the fine pen and ink line work in the Milton illustrations, appears influenced by his working small as well as working as a line engraver.
But I am not sure if any of this reveals a mind thinking of convergence, unless we assume the still point of convergence to be line and drawing as Blake defines them. What I see is an intense sensitivity to each medium he worked in and to what it shared with other media and what made it unique. I agree with Morris that Blake resisted narrow constructions of his tasks—and avoided unnecessary complexities or expensive ways of doing things (what working man doesn't?)—and that it was never an either/or type of thing for Blake. Indeed, one of his great strengths as an artist lay in his ability to mix it up within one media and to work simultaneously in various media that seem mutually exclusive. Dynaweb, on the other hand, is either/or, and it has prints as its paradigm only because we used it first with the illuminated books. What worked perfectly well in marking up prints, however, forces us, in the metadata (not in the display), to treat unique works as though they were copies or parts of a series. This is another case of how our medium's physical or structural limitations affect our choices—and I look forward to the day we move beyond Dynaweb to a search engine that recognizes a more flexible mark-up, "and/both" rather than "either/or." I think it would be pretty cool if Dynaweb inadvertently revealed or reflected how Blake thought about media—and I can see theoretically how a new medium of reproduction could reveal new insights about the things reproduced or the manner or spirit in which they were produced—but I don't think it does in this case. But then maybe I just can't fully transcend traditional categories of production.
KK: Looking back on the original SiR symposium on the future of Blake studies, I find myself somewhat surprised, as I've already mentioned, by the backseat representation took in the prophetic musings of the various contributors, who chose to frontload interpretation and its futures instead: Foucault, deconstruction, formalism, evaluative criticism—all of them more or less eclipsed things bibliographic and editorial. Nelson Hilton and David Erdman's prescient but brief remarks aside, the only contributor to engage reproduction with any intensity was John Grant, who, to his credit, adumbrated something not unlike the William Blake Archive in scale, comprehensiveness, and image-intensiveness. As to why the original coven of prophets overlooked "future editors," as Grant addressed them in the vocative, we can perhaps point an accusatory finger at Butlin's magnificent catalogue raisonné of Blake's paintings and drawings, hot off the press in 1982, whose stature cast a shadow so long that it was difficult at the time to see what might supercede it.
In 1996, with the publication of The Book of Thel, copy F, the William Blake Archive put issues of digital reproduction on center stage in a big way. I'd like to quote an editorial statement excerpted from some of the WBA's online materials:
We are hoping, of course, that the Archive, once extended to encompass the full range of Blake's work, will ultimately set a new standard of accessibility to a vast collection of visual and textual materials that are central to an adequate grasp of the art and literature of the eighteenth and nineteenth centuries. But we have also come to see the Blake project as a pacesetting instance of a fundamental shift in the ideas of "archive," "catalogue," and "edition" as both processes and products. Though "edition" and "archive" are the terms we have fallen back on, in fact we have envisioned a unique resource unlike any other currently available—a hybrid all-in-one edition, catalogue, database, and set of scholarly tools capable of taking full advantage of the opportunities offered by new information technology.>
Can I get your reflections on the new era of representation that the WBA has ushered in? What kind of precedent does it set for the next generation of Blake editors?
RNE: A difficult question, or really set of questions. The Archive was originally designed for scholars, but its full impact on scholarship has yet to emerge. It may take a few more years to move beyond the "my, that's a pretty picture" stage and into the full exploitation of some of the Archive's resources, particularly text and image searches. I suspect that younger scholars, now just entering graduate school (or kindergarten?) will figure out ways to use the Archive unimagined in the dreams of older folks like myself.
The Archive's influence in the long term may have more to do with the issues it raises, in terms of editorial theory and concepts of representation, than its utility in support of traditional types of textual and iconographic research. Our concentration on the object, and on the unique qualities of each exemplar of a particular "work" or "image," leads to a multiplication of textual instantiations of what was previously thought to be a single text—not the text of The Marriage of Heaven and Hell but a text of copy A, a text of copy B, etc. The approach accords with certain materialist strains in recent critical theory and works counter to any notions of an image or text that transcends its physical embodiments. In that sense, the Archive can appear to be anti-intellectual, too much concerned with objects and not sufficiently oriented toward ideas. But criticisms in that vein (which I personally enjoy) fail to grasp the way the Archive implicitly raises some rather complex ideas in the form of questions about what constitutes a text, a work, a copy, an image, a picture, a representation.
Let me give just one example about how the Archive breaks down some of our traditional ways of defining these terms. The common-sense notion of a "copy" of a book is deeply embedded in the architecture of the Archive as originally developed for the presentation of Blake's illuminated books. But what happens when we move from books to drawings and paintings? The hierarchical structure of our Dynaweb program makes it very difficult to eliminate the "copy" level without serious distortions in functionality and display. Thus, we have been forced to continue with an implicit (or virtual) concept of "copy" even for drawings. When there is more than one drawing of a particular design, even with considerable variation among them, the concept of "copies" takes on a meaning we generally express through the term "versions." When there is only one drawing of a design, then we are dealing with "copy 1 of 1." But what is being "copied"? An image in Blake's mind, one could imagine—although following that line of thinking has transcendentalizing implications I'm not too comfortable with. But "copy 1 of 1" also might suggest that Blake could have drawn or painted more exemplars of the "same" (and just what does that mean?) design. And maybe he did; we just need to turn up "copy 2 of 2" one day. In the meantime, we can contemplate the ways the Internet influences our thinking about very basic concepts.
JV: Bob's right, of course, that the Archive's full impact on scholarship has yet to emerge and is probably still a ways off, but there are signs that is has started. A special session on new technology and Blake, with a focus on the Archive, is scheduled for 2002 MLA, and there have already been similar sessions at MLA (most notably two in 1998 that resulted in a special issue of The Wordsworth Circle, Summer 1999) and at NASSR, and there have been master theses and articles written on the Archive from the perspective of humanities computing. The Archive will not have its fullest impact on Blake scholarship, though, until we add even more Blakes, particularly the non-illuminated works. At the moment, we have over 2000 unpublished images. These are still being scanned, color corrected, transcribed, or tagged to make them searchable, and they include all of Blake's original and commercial prints; all 537 Night Thoughts designs; nearly all of Blake's designs to Milton's works (including two copies of Comus, On the Morning of Christ's Nativity, and Paradise Lost); all the drawings, sketches, watercolors, and prints relevant to the study of The Book of Job and Dante; and all the drawings and sketches in the British Museum. We will continue to add Blake's art and manuscripts from various collections over the next few years, including never-before-reproduced illuminated books. The lag time between what the editors and staff do and what the public sees will decrease significantly for the non-illuminated works, because we have created a new "Preview" wing in the Archive in which they can be viewed with all their bells and whistles without being searchable. Eventually they will be, of course, but because making them so is so labor intensive, requiring our marking them up in Standard Generalized Markup Language (SGML), that is, identifying or "tagging" every "characteristic" of every component of an image—the searchable terms on our search page give you an idea of how one figure alone could have many dozens of characteristics—we have decided to publish them when they are visually and bibliographically ready.
More works—most never reproduced in color before—and many works never before reproduced at all should by themselves have an effect on Blake scholarship. But I think the Archive's full impact will be felt when the "virtual lightbox" which our technical editor, Matthew Kirschenbaum, and his team are developing for the Archive, is fully operational. A lightbox is used to sort transparencies (e.g., 35mm slides, 4 x 5" transparencies and larger), and it is literally that, a box with a frosted glass top over a light source (usually color-corrected florescent tube illuminated at 5000K so the colors are displayed accurately). Imagine turning your monitor screen or a portion of it into a lightbox and dragging images or their thumbnails from anywhere in the Archive onto it, and imagine being able to resize those images or any part of them by clicking till you have whatever sizes you want, with the details remaining clear and unpixelated. (The new JPEG2000 compression formats for images will make this possible.) The comparison feature we currently have in the Archive for the illuminated books is a powerful tool, but it can only compare different versions of the same object (e.g., "The Tyger" from all the Archive's copies of Songs). The lightbox will enable you to compare images from across media, juxtapose them as you wish, and resize them to provide the details you need. Some of these things can be done now using a robust image editor like Adobe Photoshop, but the more user-friendly "virtual lightbox" will be an integral part of the Archive's environment and will consolidate those editing features that will facilitate critical, editorial, and art-historical analyses.
The lightbox will be a boon to teaching as well as research. The images that you bring to and arrange on the lightbox—say a sequence consisting of preliminary drawings, finished watercolor, and later variants of the composition or key motif, or a sequence consisting of pages from the Four Zoas and their counterparts in Milton and Jerusalem—can be projected from one's laptop via a projector onto a slide screen in the classroom. You can use images in the Archive, in other words, as slides, literally, placing them into folders from which you can bring them to the lightbox/carousel. For personal research, though, the ideal will be large, high-resolution flat monitors, the size of the literal "desktop." In such an environment, you can use part of the screen for the lightbox and work with multiple documents scaled to size; browse or search the Archive in another part of the screen; and use Inote, the Archive's Image Annotation tool, to outline areas on images that you wish to annotate for your private use or attach to the images you wish to project. Teachers and researchers, of course, will need to keep in mind the "Fair Use" clause of copyright law.
Now, for me, the best of all possible worlds will arrive when there are other scholarly sites like the Archive and we all have borrowing privileges from one another. Imagine working with Blake's texts or images and bringing onto the lightbox images downloaded from related sites, devoted to such artists as Romney, Hogarth, Barry, Rowlandson, Fuseli, Flaxman, Reynolds, or Turner, as well as from more specialized exhibitions of art works in the virtual Romantic Circles Art Gallery. Working in this manner will, I think, make possible new ways of thinking and writing about Blake. I would hope to see more studies in which images generate the text rather than interpretations looking for images as illustrations.
I seem, with this emphasis on "the future availability of Everything pictorial, by precursors, cursors, and the Interpreter himself" (392) to have recapitulated Edman's expectations. I seem to have also modulated back to the first question, but while I am here, let me say that a few things wished for did come about. Adams and Gleckner hoped for studies that examined what Blake read to see if his readings of others were as "eccentric in his time as we have tacitly supposed" (400). I think in aesthetics Morris has done much to answer this question, showing in his Counter-Arts Conspiracy that Blake joined arguments and did not start them, and used the language of those contemporary discourses, with all their analogies, plots, narratives, and metaphors intact. But along the same lines, Gleckner noted that there are no shortcuts to good interpretations, to knowing Blake's connections with the literature of his day and before him: "We make ourselves art critics and iconographers but, as Erdman suggests, few of us work at learning to read seriously Night Thoughts or The Grave or The Faerie Queene or The Canterbury Tales or Pilgrim's Progress. Blake did—but oddly enough we have not really got very far beyond the Bible and Milton, considerable as they are" (432). I wonder sometimes if the Archive's image search engine is providing shortcuts to knowing Blake. Why study 537 illustrations when you can ask the machine to find things for you? Are search engines convenient scholarly tools like concordances or "only the Contents or Index of already publish'd books" (MHH 21, E 42)? To finish my recapitulation of Erdman: are they an inevitable result of being "overwhelmed" by the "availability of Everything pictorial" or the "new language" of humanities computing?
ME: Trying to talk about representation makes me feel like I'm getting a stomach virus. There are tons of things worth saying about it if only I were smart enough to think of them. But, as Kari says, one of the notable features of those 1982 responses was indeed the attention to interpretation. Deconstruction, the new historicism, and signs of the class-race-gender triad that was soon codified as cultural studies were everywhere, while editing was pretty much nowhere—a pale artifact of the 50s and 60s (Fredson Bowers, the overweight American-edition projects generated by the MLA and NEH, and so on). That soon began to change, though, with the challenges to fundamental editorial assumptions and methods issued by Jerry McGann and others in the 80s and turned into pointed, protracted, and productive debates by the responses from Thomas Tanselle and others. Textual criticism suddenly took off and became what it seldom has been allowed to be, an interesting subject—and so it fortunately remains, in my opinion. Those helped prepare the editorial ground for taking advantage of the rise of networked desktop computing and the Web in the 90s. Let's put down a couple of then-and-now markers: by 1982 I think I may have seen an Osborne 1 "portable" desktop computer and soon after bought a KayPro 2 with a 6-inch monitor for "word processing." In 1992 I was using email regularly, but I'm not sure I had yet heard of the World Wide Web. In 2002, well, everyone knows the story—wireless everything, gigabytes as common as kilobytes used to be, the human genome, the resurrection of Shakespeare from the dead. I may be gullible, but it's all enough to make me want to say that in those twenty years representation has become everything—or many things anyway. (An interview like this gives me the right to say reckless things that will turn into ridiculous things in far less than twenty years.) So when Jack Grant—the only 1982 soothsayer to mention computers, I think—brings computers into his prediction, as he frequently does, they're always one hundred years away at the end of the 21st century, when scholars will be able to push a button on their computers and get the Variorum Blake. It's all very 2001. Someone else mentions videodiscs, a medium now replaced by the DVD—temporarily, until its replacement comes along, which it will soon enough, or too soon. Most of the respondents seem to envision the medium of the future as microform, of all things. (This must have been about the time the MLA got into the business of promoting the purchase of a handy desktop microform reader by all its members!—before it turned to promoting Nota Bene as its favorite word processing application.) That shows you how absurd it is for English professors to predict the future.
On the other hand, representation certainly isn't absent from those twenty-year-old predictions. On the very first page David Erdman asks rhetorically if readers have ever felt the scholarly agony of trying to compare several of Blake's 537 Night Thoughts drawings in the print room of the British Museum. He asks, "Will the future availability of Everything pictorial by [Blake's predecessors, contemporaries, and Blake himself] in full color in projectible and comparable microform—will this overwhelm our team—or teach us a new language?" (392). Those are problems and promises of representation that are central to the purpose and design of the Blake Archive. Books and microform aren't very well suited to solving such problems (which is not a casual dismissal of microform, much less books, just a straightforward acknowledgement of their limitations as specialized forms of information storage).
I agree with Bob that working on an editorial project like the Archive forces (not too strong a word) you to think about representation in alarmingly particular and general ways. It can be quite a surprise to find yourself acting out, as it were, the McLuhanesque idea (speaking of reckless—and McLuhan, typically, recycled the idea from someone else, in this case Lyman Bryson) that technology is explicitness: you can't make a machine that spins yarn or weaves yarn into fabric without being explicit about the processes and the assumptions that underlie them, principle by principle and step by painful step. The machine represents, is a dynamic metaphor for, that explicitness. The same is true when it comes to computing machines, which is why we, in trying to figure out how to make a Blake Archive, necessarily end up trying to figure out what a Blake Archive is, what texts and images are, and how they can all live together under one roof of hardware and software. That's why it's interesting work and not just a lot of work (which it is too).
Consequently you end up arguing over the most absurd things. A classic example from the Archive would be the months we spent trying to decide how to number the lines of words (poems, prose, "texts") that constitute transcriptions—the transformations of words that have been extracted from one source or platform or support, such as the printed impression of a plate from one of Blake's illuminated books, and rematerialized onto another, such as your monitor in Darwin, Australia. We struggled mightily with that issue—it soon stopped being an issue and became a problem—in all its ramifications: editorial assumptions of the conventional kind, of course (what are our editorial principles? Is the Blake Archive making "critical" editions of Blake, or "documentary" editions, or something else entirely? Are we attempting to honor Blake's artistic intentions in some way, even in line numbering?). But the struggle quickly expanded to include the editorial history of aids to reading (such as divisions between words, page numbers, paragraphs, chapters, indexes, tables of contents). Where would we locate ourselves in that history? Should we try to make our numbering more convenient by following previously established conventions (which are, we discovered, irregular, inconsistent, and incomplete) of numbering Blake's lines, or should we break with convention in favor of a new system designed specifically for our specific purposes? And at some point we must always factor in the issues raised by our electronic medium: what will the hardware and software permit us to do? In electronic editing you meet the medium very forcefully because it's always an unfamiliar medium no matter how familiar you are with it, and you meet it as an editorial issue, to be weighed along with the rest of the entire editorial legacy (that is, the way other editors have represented Blake), pure principle, and so on. All over a silly thing like line numbering, which no one ever notices.
I'm sure this all sounds incredibly tedious to people who aren't directly involved in the argument (are you serious? line numbering?). But even in protracted arguments about line numbering—and I could name a quick hundred other examples, from standards of image reproduction (no matter how precise and elaborate, are they ultimately groundless?) to the structure of a table of contents to the design of the BAD, the Blake Archive DTD, to the ins and outs of image-and text-searching—I've experienced over and over the outright thrill of getting down to that editorial bedrock where basic assumptions are exposed—a sort of private viewing of the most intimate recesses of representation and its secrets. But I've said much too much already.
KK: Bob, you predict above that "it may take a few more years to move beyond the 'my, that's a pretty picture' stage and into the full exploitation of some of the Archive's resources, particularly text and image searches." Joe points to a forthcoming MLA session, organized by Sheila Spector, that aims to lay the groundwork for just that kind of sustained exploitation. Spector's central question—how do electronic resources like the WBA and Nelson Hilton's Blake Digital Text Project extend traditional scholarship?—seems to me a timely and important one. Joe and Morris, I'd be curious to know how each of you might answer it. Can you imagine a hypothetical book project, for example, that couldn't have been written pre-WBA? I'll tip my hand by saying that I'm particularly curious as to how image searches might be imaginatively exploited, not least because of the cautionary tales I've heard the three of you tell about how not to use them: e.g., in the service of heavy-handed semiotic analyses that revel in every infinitesimally small image variant. But let's assume for the sake of argument that our hypothetical researcher intends to heed all the important caveats; in that case, what new knowledge might emerge from systematic use of WBA resources?
ME: Electronic resources like the Archive and Nelson Hilton's project "extend traditional scholarship" by making it potentially deeper, wider, and less parochial because less bound to a limited pool of information. Joe has pointed to an interesting feature of the history of reproductions of Blake's illuminated books: a few copies get reproduced over and over, while others have never been reproduced, and most research has understandably focused on those copies that are most easily available—maybe in reproductions, but maybe also in museums that welcome curious students rather than in museums that don't, or museums that are in attractive locations that are cheap to get to rather than in hostile far-flung outposts. We lament this the way Thel laments mutability, but it's inevitable, and any one limitation stands for a host of others that we must always live with; it's better when we're aware of those restrictions than when we're not, but being aware does not necessarily remove the restrictions, it just makes you a bit more cautious about the conclusions you draw.
I've described so many images in such tedious detail for the Archive that I can hardly bring myself to think about how those might be used in constructing image searches that would actually benefit someone else. It's my impression that so far no one has made much use at all of the image-search function in the Archive; if they had, I'm sure we'd have received many more questions and complaints, because the system is far from perfect. But the only complaints we've received have been from people who use the image-search function to look for the titles of works they've heard about, and image searches aren't designed for that. That said, I'm pretty sure that Bob, Joe, and I are united in the belief that the image search is full of wonderful possibilities for future research, simply because it allows you to find images (shapes, characters in poses, actions, natural objects, manmade structures) that are like other images across the Blake canon. Our instincts tell us that this has to be a useful thing to be able to do, if only because it puts images into a searchable category parallel to words, and everyone understands the value of being able to search through a stock of words. Now if you want to find out how Blake uses "rock[s]" in his work, you don't have at your disposal a nice, efficient concordance for words on the one hand and an utterly frustrating, undifferentiated, unsystematic pile of catalogues and other books with reproductions in them on the other.
Of course you do have to take the time to figure out how to conduct an image search on our terms. You have to know, for instance, that we use a controlled vocabulary, and that our term is "stone" not "rock." Image searching is a hot topic in computer science, because there are so many compelling, and not a few lucrative, uses for the ability to search efficiently across huge stocks of pictures, whether you're looking for shirts to buy or targets to hit. And our way of making image searches possible isn't of much interest to computer scientists because it's way too labor intensive. But it has many virtues, such as that it's possible here and now, unlike more automated systems, and that, rather than start from raw data, it builds on the legacy of what we already know, or think we know, about the content of Blake's designs (we sometimes know Urizen from Los, or a sheep from a serpent), and what we know about the history of Western visual representation in general (that "contrapposto" is a conventional pose, and we can label it when we see it). I won't go on like this, because it gets so boring, but one point worth making, when it comes to predictions of future use, is that our system is probably better for extending what we already know than for discovering the never-before-known because we do build on the scholarly legacy. We are bound to it—and we pass our bound condition on to users of our image-search function the way families pass on their attitudes from generation to generation, the disease with the health, the blindness with the insight. I can't deny that starting over with raw data has its advantages, but quick results isn't likely to be one of them, and we wanted to get quick results for our pains—and to give other scholars quick results for their collaboration, or call it complicity, with us.
Meanwhile, for now, the critical challenges remain basically the same—though I'm very hesitant to say that, and of course "basically" is a big word, really just a hedge against being dead wrong before they even close the casket. But that's where you suddenly feel the puniness of your powers of prediction. You know, "as far as I can see": you try to peer beyond the horizon of what you know, and all you see is a more or less clever take on what you know, heavily shaped by what you don't know. But it's pretty safe if vague to note that every time the pool of accessible information gets bigger, the more likely it becomes that new things will turn up. When you learn new things, inevitably you or someone else figures out how to spring surprising new thoughts on you. How does Blake put it: what is now known, was once only imagined? It is unimaginable that the availability of so much matter for thought won't alter the scholarship of the next generation.
JV: I have already expressed my fears of how search engines, as short cuts to Blake's images and texts, can create an illusion of knowing Blake that actually undermines knowledge of his works. But I also think that they have the potential to assist and deepen our understanding enormously. They have this potential because they enable you to search for multiple terms and motifs (up to nineteen) at the same time, and this can reveal patterns of thought and visual formulaes. We know that there are lots of snakes in Blake, but how often does an image of a snake occur with, say, a book or writing instrument, or, in the illuminated books, without the word "snake" or "serpent" anywhere near? Are these patterns across categories or restricted to specific genres and periods? In other words: What occurs with what else more often than not? Where and when? I am sure we are in for some surprises. I got one when I did a word search for "God," "king," and "priest." If I were looking for any one of these very Blakean words/concepts, I would get hundreds of hits. But in combination as a triad I got only one, very much to my surprise. The words were used as a phrase in "The Chimney Sweeper" from Songs of Experience. When all the illuminated books, engravings, and illustrations to Job, Milton, Gray, Young, the Bible, and Dante are reproduced and searchable, I think we are going to see more studies focused on Blake's iconography and more of Blake's art figure into literary analyses.
KK: I'd like to return to the topic of editing by posing a curricular question. Morris has a terrific essay on the history of Blake editing in a recently published collection entitled Reimagining Textuality. I'll return to that essay later, but for now want only to single out a blurb on the back of the book, which enthusiastically touts the collection as one that "deserves to push Textual Studies into the mainstream of humanistic scholarship." This mainstreaming sounds good in theory, but could be disastrous in practice within the context of Blake studies, given the awesome demands Blake's oeuvre places on the aspiring editor; it seems to me that the dilettante —unless properly forewarned—is well advised to steer clear of the pictorial Blake altogether. He exacts too high an admission price of an aspiring bibliographer, asking at the very least that she be both art historian and literary critic alike; in the end the probability that she'll end up a jack of all trades and a master of none seems pretty high. The dangers of dilettantism are addressed head on, Bob and Joe, in your recent essay on Blake's method of color printing, which offers an indictment of narrowly literary approaches to Blake as material artist.
Given such challenges, I'm curious as to what advice you would offer students of Blake who wish to study him from an editorial or bibliographic perspective. As artist (Joe), collector (Bob), and historian of technology (Morris), each of you, for example, brings some form of extra-literary knowledge to bear on your editorial work. Does this triad constitute a sufficient curriculum for editing Blake? What else does the modern student of Blake need to know now that technologies of representation embrace silicon and pixels—is it mostly a question of skill sets?
RNE: Some answers are implicit in the question. "All of the above," to begin with. Just as editors of texts extant only in the form of letterpress printings must know a good deal about how books were printed, and medievalists working with manuscripts must know about calligraphy and codicology, Blake's editors must know as much as possible about the way Blake produced his texts. Further—and this is crucial—Blake's editors have to understand the nature of the media in which their own, edited texts are going to be represented, be it letterpress, digital texts, digital images, or some combination of these and other technologies. The same general and two-fold rule—understanding Blake's means of production, understanding one's own means of production—also pertains to the reproduction of his pictorial images. In that regard, we need to learn a good deal more about Blake's habits as a draftsman and painter. And those who reproduce his work need to understand how images are reproduced today on paper or on monitors. I have a long way to go to live up to these criteria.
ME: No one is born an expert on anything, and no one can ever learn enough to be the world's leading expert in anything, even the tiniest area of human knowledge, for very long. That's simultaneously humbling and reassuring. I've come to see that for me personally Blake has been a shortcut, a focusing device, a knowledge-crutch—editorially and otherwise. I mean that he's been an opportunity to ask large questions in small places. All three of us are English professors, no matter what else we like to think we are or might once have been or might be in another life. Each of us has special areas of knowledge that make us (I believe) a good team of collaborators, because we have a base of shared knowledge (and interest—all of us have always been interested in the material/editorial representations of Blake) that is complemented by special knowledge and interests and perspectives that each of us has that the other ones don't have. That sounds like a cliché about teamwork and team spirit, but it's an honest observation. It's also largely accidental, so I'm properly grateful for the opportunity to work alongside such wonderful colleagues, which has been priceless.
As for what Kari calls the "price of entry," one of the special attractions of Blake's work is the very challenge it presents to the categories in which we store specialized knowledge, which for lack of better things to say I've said too many times. That makes everyone a dilettante. Literary expertise is by no means sufficient, and neither is expertise in art history or studio practice or any other kind of history or practice. We could adjust this statement to fit anyone's work, Byron's or Verdi's or Giorgio de Chirico's—or any subject whatever. When it comes to putting demands on knowledge, Blake is not a unique case. But his extreme demands do make you realize sooner than other artists might that the cultural conventions of information storage—I mean information about texts and how to read them, images and how to view them, music and how to listen to it—are far too restrictive to give the poor reader/viewer a handle on his work. To quote Harold Bloom, "I read one of the most eloquent descriptive passages in the language; I stare, disbelievingly, at an inadequate engraved illumination, and then I try, too strenuously, to isolate an image that Blake, as a poet, knew better than to isolate." Bloom wrote this as a negative comment on Blake's imagemaking talents, but it's also a comment on the limits of Bloom's perspective, which can stand for the limits of all of our perspectives, even those developed by intense specialization over a lifetime. With specialization comes inflexibility and blinkered knowledge.
A second difficulty with Blake's work is its unconventional draw on the resources in each of the conventional cultural categories where we store information about him (I mean, again, words and pictures, literature and art). It's not just that the combination of words and images is unusual (and therefore difficult to understand, record, and remember—speaking editorially rather than critically, of editorial understanding, recording, and remembering), but that the poems and the images are themselves unconventional. They don't fit all that well into the channels that guide our understanding of the information that flows through them. What this means, in effect, is that no conventional expertise can claim the best chance of understanding Blake even within a conventional category: literary critics aren't necessarily at the greatest advantage in understanding his writing, art historians aren't necessarily going to understand the visual work best, and artists aren't necessarily going to deliver the best understanding of his artistic techniques. Specialists, after all, are specialists in convention—literary conventions of the eighteenth century, for instance. Now all these kinds of expertise underwrite bibliographical and editorial expertise, and if there's structural failure at one point, failure across the board is likely. A useful cautionary tale, because it's all about facts and techniques and not about highflying criticism and theory, is the commendable effort, in the 1940s, to understand Blake's illuminated-printing techniques. Who was better qualified for the job than Stanley William Hayter (1901-1988), "probably the best-known printmaker of his time" (Time, 19 June 1961) and certainly among the most accomplished technically, and Ruthven Todd (1914-1978), who had attended the Edinburgh School of Art and made himself a formidable Blake expert? Hayter, Todd, and artist Joan Miró teamed up on an elaborate scholarly explanation of how Blake did illuminated printing that was, as Bob and Joe have demonstrated, dead wrong. Their artistic training, experience, and outlook counted for something, no doubt, but it clearly wasn't sufficient. And one rightly wonders why no art historian quarreled with their judgments. You can multiply stories like that endlessly, of experts who have fallen afoul of Blake's unconventional imagination.
I can't afford to exaggerate his unconventionality, since I've invested years in trying to persuade my colleagues that, in important respects, Blake's ideas about art, at least, can be understood only in the context of other contemporary theories about art. Joe and Bob have shown in various ways how Blake was a conventional technician. And I don't know enough to deny Northrop Frye's attractive contention that key elements of Blake's literary works make them utterly fundamental products of the literary imagination—and utterly conventional in that positive sense. But you can't escape his strangeness, either, or tame him with history and context. He won't hold still for it. As Tom Mitchell cleverly put it in his 1982 remarks, Blake is a truant who won't behave, messy and undignified, out of the box, beyond the pale. You can sometimes see where's he coming from, but you much less often can tell where's he's headed.
But you asked if I have any advice for the editorial and bibliographical crowd who may want to manage information about Blake in the future. Only of a uselessly general kind. I think it's less a matter of "skill sets" than attitude and maybe temperament. Curiosity, a sense of adventure, and skepticism are essential. Intense curiosity will pull new information into your field of vision; a sense of intellectual adventure will motivate fearless category-jumping and keep the field of vision growing. Editing in new media I think of as Xediting (I've copyrighted that)—experimental editing not as an occasional thing but as a permanent condition, where you make the rules as you proceed, and where nothing offers even the illusion of permanence: any solution that works today probably won't work tomorrow. You have to be able to thrive in that environment. Skepticism is helpful in filtering new information and provides a degree of resistance to facile and opportunistic formulations—a proven danger with Blake, whose complexity gives a special allure to simple formulations that promise new insight at low cost (the price of entry again). And tying all these together is the sine qua non for all editors and bibliographers: if you don't have a large measure of anal retentiveness, you'll never get off editorial first base. And perhaps one final thought. You don't have to understand Blake to be his editor or bibliographer. If that were a requirement, he wouldn't have any editors. But then editors and bibliographers, more than critics, are used to that. They often don't feel the need to understand, in the usual sense, the objects of their attention, because they can operate at a certain conceptual distance from them, as they work with the objects so that other people can have the illusion of working inside them.
JV: I think that Morris is correct about one's approach, pedagogy, perspective, discipline, training, expertise, specialty—call it what you will—being both enabling and limiting simultaneously, which is why working collaboratively makes sense. Apparently scientists and health professionals, who are accustomed to working so, have an easier time acknowledging this than humanists, who are by training, if not nature, solitary creatures. Maybe learning how to work with others is one of the skills the next generation of Blake editors needs to acquire. And by others, I do not mean only other scholars. In creating the Archive, Bob, Morris, and I had to learn to work with computers and computer folks who spoke different languages and thought in different terms. We all had to learn to become a team, which in turn enabled us to envision something new and, as Morris notes, question every assumption we had about texts, images, editing, documents, reading, etc. And we had to unlearn "print-mind"—the idea that you publish things only in their final form—and adapt to a constant "work-in-progress" environment in which last year's sensible solutions are tomorrow's problems.
But as Morris also points out, even the collaboration of experts is no guarantee of success. His example, though, is worth examining closely. Hayter and team, with their extensive experience in etching and lithography, assumed that Blake's core technical problem was reproducing text without having to write it backward on the plate. They knew that the transfer methods then available to Blake did not work in relief etching, so they had him invent a method perfectly analogous to the one Senefelder invented for lithography—ten years later. Historically, their theory is anachronistic and fails to recognize Blake's ability (well known among his friends) to write backwards. They thought too much about how they would solve a problem Blake never encountered and too little about the historical Blake.
Of course, knowing Blake in his times is difficult even for dedicated Blakeans, as Erdman and others recognized twenty years ago. And as Bob and I pointed out back then, knowing how and when Blake created and produced his books is important too because it affects how editors think about Blake's texts. No doubt I would say that, given my own visual and graphic biases in these matters. As Blake said, "as the Eye is formed, such are its Powers" (letter of 23 August 1799, E 702) and "As the Eye—Such the Object" (anno. to Reynolds, E 645). So, yes, the way my eyes have been formed has affected my eye as an editor and scholar and my vision for and of Blake. Guilty as charged. But I think editors need to understand illuminated printing because misunderstanding it has had dire consequences—and still does, as is demonstrated in the essay you noted a moment ago on Blake's color printing. Twenty years ago, the books were thought to have been produced one at a time, usually by commission, leading reasonably to the assumption that each and every difference among copies signified intentional revision. The books were incorrectly dated, making it seem like Blake was printing a few copies of them every year for most of his life. Copies were loosely identified as "early" or "late," which was as historically placed as they could be. With no obvious copy text and the privileging of any specific copy on very shaky ground, editing Blake became, ironically, ahistorical. I am not referring to Keynes's reader's texts, which do not pretend to represent the historical texts, but to those established by Erdman and Bentley. Their texts purport to represent the texts as originally executed on the copperplates, though they arrived at their reconstructions differently. Erdman established composite texts—what are in effect texts by consensus—by comparing various copies of the same work and averaging out the markings. Bentley relied on the clearest monochrome and posthumous copies in the belief that they could serve as copy texts.
Reconstructing the nonextant plate text, like recovering manuscript or fair copy, aims at capturing the historical moment of first creation, but it resulted in abstractions of the kind Bob referred to as the "work" level, because the excavations cannot be verified and the methods deducing them are suspect. Whatever its source—notebook or back of a napkin—the text rewritten on the plate was in acid-resistant "ink" that was etched into relief, which could have removed fine details, like tails on commas. And the prints that reflect the etched texts do so imperfectly, because their appearance was affected by the way they were printed (e.g., underinked, overinked, inked with a stiff or oily ink onto paper of varying thickness and dampness). The text as originally written on the plate, in other words, is two removes from the printed text we read and is not recoverable.
That said, I think establishing the etched text for each book is an editorial project probably still worth pursuing. Bentley and Erdman had the right idea; the etched text is the constant matrix for all copies and provides the grounds for identifying variants. They just had less than optimal methods for establishing it. Today, with digital imaging, we can create three-dimensional models of the plates and examine each letter and mark in far greater detail than was possible using magnifying glasses, black and white photographs, and one's notes. The Archive's reproductions could possibly contribute to the reconstruction of etched texts, though all would require the reader to know Blake technically and have image-editing skills. Computer enhancement of a few copies of a book from different printings (and hence production styles) could yield and verify reconstructions of the plates (but not necessarily of the artifact, that is, the order of its plates) more exacting than those from consensus or any one particular copy. However, even with a methodology more systematic, consistent, and rigorous, the results would still be speculative, closer but still with built-in ambiguities, because a core problem in editing Blake—the defining of punctuation marks—is a subjective call: what appears oblong and comma-like to one editor can appear like a period to another and—more troubling—the same shape in a line or page of well-defined commas or an excess of round dots will appear/be read differently. Maybe the next generation of editors will need to study the phenomenology of reading?
Today we have a good idea of which copies of a book were printed together and when, which enables editors to establish a book's history and to recognize or at least question whether differences are deliberate revisions or the results of different printing styles years apart. Editing Blake historically, though, is still a challenge, for all the reasons cited by Morris and Bob and for all those still worth repeating. Illuminated poetry is pictorial; Blake's words are images and are versioned and embodied in particular copies and much of the punctuation requires interpretation and/or is not translatable into type. There really is no copy text for any book existing in more than one copy, and thus most of the theories governing editorial practice don't apply. Choosing the first or last copy of a book on the grounds that its text brings you closest to Blake's first or last intentions might seem like a good idea, but identifying such copies is impossible for all but one or two books.
The Archive doesn't attempt anything so grand (or impossible?) as establishing etched, base, or ideal texts, or deducing Blake's first or last intentions. We reproduce exemplary copies of each book from each of its printings, first, last, and everything in between, but, in effect, treat each copy as its own copy text and, within the limits of the medium (that price of entry again), establish diplomatic transcriptions of each copy as printed and retouched in pen and ink by Blake. This is the object orientation that Bob mentioned earlier, which keeps us historically grounded. From the start we saw our goal as providing trustworthy reproductions of Blake's pages for study—and as raw materials for editorial projects envisioned and conducted by others.
30. For more on the Archive's ImageSizer applet, see "The Persistence of Vision: Images and Imaging at the William Blake Archive," RLG DigiNews 4.1 (2000) <http://www.rlg.org/preserv/diginews/diginews4-1.html> and "A Tour of the William Blake Archive," WBA <http://www.blakearchive.org/public/about/tour/index.html>.
32. "No representation without taxation": for a clever metaphoric deployment of this phrase in a literary context, see W. J. T. Mitchell, "Representation," in Critical Terms for Literary Study, ed. Frank Lentricchia and Thomas McLaughlin (Chicago: U of Chicago P, 1990) 21.
33. From the Public Address: "He who Draws best must be the best Artist" (E 582); "I request the Society to inspect my Print of which Drawing is the Foundation & indeed the Superstructure it is Drawing on Copper as Painting ought to be Drawing on Canvas or any other [table] <surface> & nothing Else" (E 572).
35. "The Archive at a Glance," WBA <http://www.blakearchive.org/public/about/glance/>.
37. Matthew G. Kirschenbaum, Jerzy W. Jaromczyk, Amit Kumar, The Virtual Lightbox 1.1.1, Beta Version July 2001 <http://www.glue.umd.edu/~mgk/lightbox/>.
38. For JPEG2000, an up-and-coming image compression standard using wavelet technology, see the JPEG2000 pages at <http://www.jpeg.org/JPEG2000.htm>. Each image in the archive is currently preserved at multiple sizes, resolutions, and formats; JPEG2000 promises to change all that by delivering the image at multiple resolutions in the same codestream. Such efficiency would obviate the need on the production end to store multiple versions of an image and on the user end for multiple downloads.
42. A "DTD," i.e., Document Type Definition, is a specially designed tag set used to describe the structure of a given document or group of related documents (a document genre). For a technical discussion of the WBA's BAD DTD, see Daniel Pitti and John Unsworth, "After the Fall: Structured Data at IATH," ACH/ALLC Conference, Debrecen, Hungary, 7 July 1998. Available online at <http://www.iath.virginia.edu/~jmu2m/ach98.html>.
43. Hilton, Blake Digital Text Project <http://www.english.uga.edu/~wblake/home1.html>.
44. For more on this reproductive rut, see Viscomi, "Digital Facsimiles: Reading the William Blake Archive," Computers and the Humanities 36 (2002): 30-31 and 30n6. Viscomi provides some rough statistics to drive home the point: "Typographic editions and reproductions of only about 20% of Blake's illuminated canon (40 or so of the 175 copies of the 19 illuminated books Blake produced between 1788 and 1827, when he died), reproduced sometimes well, sometimes execrably, but in no coherent historical order and insufficient detail to sustain scholarly and editorial research--this was the state of Blake studies when in 1993 the editors of the Blake Archive began to conceive of reproducing Blake digitally" (30-31).
49. From Mitchell's contribution to the round robin: "Everything suggests to me that we are about to rediscover the dangerous Blake, the angry, flawed Blake, the crank who knew and repeated just about every bit of nonsense ever thought in the eighteenth century; Blake the ingrate, the sexist, the madman, the religious fanatic, the tyrannical husband, the second-rate draughtsman" 410-411.