Rubrique Smart

Dada Photomontage and Sitemaps

George L. Dillon
University of Washington
[published in PostModern Culture 10:2 (2000) ]

© 2000 George L. Dillon
All rights reserved.

Maps of websites represent them as networks of linked files and provide perspective on the sites and their parts. Some of these maps and perspectives are utterly standardized (those for information retrieval, for example), but some are as much for interpretation as navigation. This piece sketches a semiotics of website imagemaps. Even more than Cubist collage, Dada photomontage (Höch, Hausmann, Schwitters, Grosz/Heartfield) presents very suggestive images composed of signifying fragments ranging from small schematic arrangements to wildernesses of profusion. One sees several patterns (matrix, swirl, cascade) with their associated semantics. We then apply these categories and principles to website imagemaps to see what the imagemap is saying visually about the site and our experience of it.

  1. We find ourselves in the midst of a new medium, we with our various discourses, which does not mean we are all experiencing the same thing. HTML hypertext seems to have about as much intrinsic character as tofu. It lends itself to many deployments; people of very diverse interests and sensibilities find excitement working in it--and seek to claim it as their own, as "the way we think" (or ought to want to think?). People differ over using it as work or play, as transparent or mazey, as an art object or as a disposible.

    At least four different sensibilities are emerging on the web: the researcher's, the explorer's, the browser's, and (this last one lacks a good name) the connector's. These sensibilities link directly to different purposes for logging on. The researcher wants to find the best information currently available on the topic of her interest. This is the presumed purpose, for example, of the "user" in Patrick Lynch and Susan Horton's Style Guide,[ 1 ] and it is relative to that purpose that they warn the web writer that the user should never be confused about the hierarchical order of the website and her place in it at any given time. The user, they say, will form a mental image of your site, and an image like that of Figure One is a "Bad User Image."

    Lynch and Horton's Poor User Image
    Figure 1--Lynch and Horton's "Poor Site Image"

    This is "poor" for them because it is headless and lacking in hierarchy and because some of the nagivational links are uncertain –reflecting probably a user's sense of finding herself in the midst of information, but not sure how she got there or where she can get to from there–not good if you are trying to compare contraindications for various hypertension medications. As Edward Tufte puts it in Visualizing Information : "Ideally, structures that organize information should be transparent, straightforward, obvious, natural, ordinary, conventional--with no need for hesitation or questioning on the part of the reader." [2]

    Browsing is not so task or goal oriented; its sensibility is nicely evoked by Mark Bernstein, himself a recovering location-and-navigation zealot, in Hypertext Gardens: Delightful Vistas , a tasteful stroll in the Horatian vein, emphasizing the pleasure of "unexpected delights" to be found in well-crafted "gardens" that supply neither the unsorted profusion of the wilderness nor the practical predictable grid of the downtown highrise. "Hypertext Gardens" itself exemplifies a hypertext garden, though Bernstein also includes a diagram of the site (in his own Storyspace format) in a technical footnote.

    Exploring is looking to get a sense of 'what's there' in a certain area; the explorer is seeking public knowledge, or at least a sample of the current discourse in a domain, not a "cabinet of curiosities," but is not pursuing a systematic or thorough search. Exploring is what you do in a wilderness. Such sites are often teacherly collections of resources on a subject. One might pursue the topic of contemporary Dadaists at Virgin Megaweb: Rubrique Smart to get the flavor and some leads to this admittedly unsystematizable domain.

    In fact, one is likely to be propelled from this site into connecting , which I mean to link with the practices of making and resolving disparate and incongruous juxtapositions. On the face of it, the Web seems to offer a huge and unsorted collection of texts and images and a simple way of pulling some of them together with the "A" linking mechanism. It is a bricoleur's dream world where the maker is like a metal sculptor in a junk yard with the A anchor for a welding torch. This sensibility or positioning toward materials is I think what Michael Joyce called a "constructive" relation to hypertext some years before the Web was up and running, and it is what George Landow sees as the underlying tendency of hypertext. The connector likes to make sense of jumps in unexpected, incongruous directions, to resolve catachreses. In Hypertext 2.0 Landow says,

    if part of the pleasure of linking arises in the act of joining two different things, then this aesthetic of juxtaposition inevitably tends towards catachresis and difference for their own ends and for the effect of surprise, sometimes surprised pleasure, that they produce. (p. 167)

    He likens such hypertext sites to Cubist collages in the ways they appropriate bits of other things (wicker, rope) and juxtapose them in a whole (p. 167). The comparison is fruitful, but it bogs down a bit in the physicality of things: hypertext anchors and targets are all text, image, sound, or movie, not stranded rope.

  2. Similar to cubist collage, but far more involved with text, is the practice of Dada photomontage as developed by Hannah Höch, Raoul Hausmann, Kurt Schwitters and others (Georg Grösz, John Heartfield, Max Ernst). The hypertext page has words and images linking to other words and images; Dada photomontage is made up of bits of photos and other images along with words and phrases from the media, not "things" but signifiers. These signifiers are recomposed into a new whole but point always to another "page" from which they were snipped. So the Dada photomontage is like a sitemap--an image of one way all the fragments go together. Of course, not all sitemaps resemble Dada assemblages: the great majority are simple itemized lists or file-and-folder menus and serve the purposes of the Researcher only. They do not suggest specific relations of individual elements or overall patterns, and they reinscribe rather than subvert conventional indications of order and relation. In the area of webspace called, however, the spirit and designs of Dada assail the orderly stable- marks-for-standard-meanings code of the Researchers with all of the ingenuity and glee of Hausmann, Höch, or Schwitters. This is the parallel that I will pursue in some detail, not just because it is remarkable and illuminating in its own right, but also because the analysis takes us into quite general issues of visual meaning. We begin by identifying the units of analysis; (2) then, formal principles for composing the units; (3) only then, the semantics of composing the units. So we will finally get to brief discussions of what the individual works mean, but (very artificially) only after we have climbed from the ground up. We will (4) then go beyond the photomontage-sitemap parallel to examine imagemap sitemaps currently on line, concluding with avowedly Dadaist ones and the Dadaifying of Post Modern Culture itself.

  3. The fragments

    Dada photomontages (and collages, for that matter) are made up of fragments of images and text from the popular culture. Not just words, but clipped bits of newspapers, posters, catalogs, tickets, letters, and fakes of the same. The development of halftone photogravure and offset printing had set loose what seemed an avalanche of photographs in newspapers and magazines, and already advertisers, first of all in America, had begun to combine photographs in one poster or advert. In 1919, Hannah Höch was working for one of the new illustrated magazines when she and Raoul Hausmann (among others) realized that this technique of mass culture could be turned against it with great force to disrupt its depictions of a normal social world and political order--to demonstrate, as Johanna Drucker puts it, "the social reality made in and through image production" [3], or again, to pry signifying practices loose "from their conventional relations or easy recuperation as readily consumable modes." [4], Benjamin H. D. Buchloh speaks in very similar fashion of Dada's "extreme procedures of juxtaposition and fragmentation by which the origins in advertising were inverted and where the constructed artificiality of the artifact destroyed the mythical nature of the commodity" [5]. More on the semantics of this juxtaposition below.

    Words too could be clipped from the headlines and pages of magazines, or scribbled on the picture indecorously, thus destabilizing the conventional relations of visual and verbal modes ("label," "illustration," "product," "pitch") in journalism and advertising.

    Whenever images from the popular media are "recycled", Dada and its oppositional politics is invoked. See for example Victor Burgin's well-known "appropriated image" work from the early 1970s [6], and Richard Hamilton's photomontage, "What makes the contemporary home such a wonderful place to be?" (1956) [7] It is of course not always clear what those oppositional positions are, and sometimes it is not certain they are there at all (as in the case of Pop art, or more recently Richard Prince, or Sandy Skoglund) or almost certain that they are not (as in the case of Jeff Koons). But the expectation of critique is very strong, amounting to a convention of appropriation or citation of mass consumer culture.

  4. These fragments have clear edges--no attempt is made to smoothe or blend one into another. (Smooth gradation from one image to the other attracts the semantics of identity and transformation rather than juxtaposition.) Because they are fragments from "elsewhere," they function in certain ways like hypertext links: they do not of course "take" you to another page or more complete image, but they do stimulate you to locate them in a recognised or imagined whole and context. That is, they function as Lücke–"gaps"in Wolfgang Iser's terms–which the reader tries to supply from memory or surmise. In some cases, the Dadaists fragmented Gestalt "good forms"–especially the human body and face; gaps so produced are often shocking and clamour for resolution. In the case of Dada works from the Weimar era in Germany, the modern reader welcomes the work of scholars who have set about identifying images and sometimes translating text to recreate as it were the common knowledge of contemporary readers/viewers. (The recognition in many cases may be tentative or approximate, of the "looks like some kind of" variety.) The fragment, in other words, is an anchor that links not to another page but to a "page" in the reader's mind. What is important for our purpose is not just this trigger effect, but the fact that they are composed-–placed in relation to other fragments.

  5. The Arrangements

    Formally, Dada photomontage is two dimensional, after the pattern of posters. There is no perspective or common scale or palette, so there is no unified point of view for the viewer to take up. Edward Tufte, develops the term "confection" to describe such divided, patterned surface without perspective or a single space. Confection, he notes, is the characteristic use of images on the Web.

  6. The fragments can be arranged in a rows-and-columns style table (which is generally rather stable/static) or in more dynamic patterns. Here I will describe four kinds of arrangements (Grid, Schematic, Swirl, Cascade), holding the discussion of reading the arrangements until they have been described.
    1. Grid/matrix (table with cells)

    Kurt Schwitters: Die Handlung

    Figure 2 –Kurt Schwitters,
    Die Handlung Spielt in Theben und Memphis zur Zeit der Herrschaft der Pharoanen
    (The Action Takes Place in Thebes and Memphis Under the Pharaohs' Rule.)
    (192x) 16.2 x 20 cm.
    [Click on image to see enlarged view]
    Hannah Hoch: Meine Hausspruche

    Figure 3 –Hannah Höch,
    Meine Hausspruche
    (My Household Proverbs)
    (1923) 32 x 41.1cm
    [Click on image to see enlarged view]
  7. When the fragments are not very numerous, they are often arranged in an open space with some sorts of relations of one to the other indicated:

    Hannah Hoch: Und Wenn Du Denkst Hannah Hoch: Marlene

    Figure 4 –Hannah Höch,
    Und wenn du denkst, der Mond geht unter
    (And When You Think the Moon is Setting).
    (1921) 21 x 13.4 cm.
    [Click on image to see enlarged view]
    Figure 5
    –Hannah Höch
    (1930) 36.7 x 41.2 cm.
    [Click on image to see enlarged view]
  8. More fragments with diagonal placement suggest movement and a center-periphery scheme; equivalence is suggested by parallels, opposition by intersecting or oblique angles.
    3. Swirl

    Kurt Schwitters: Das Kotsbild Raoul Hausmann, Dada Cino

    Figure 6 –Kurt Schwitters
    Das Kotsbild
    (Vomit Picture)
    (1922) 27 x 19.5 cm.
    [Click on image to see enlarged view]
    Figure 7 –Raoul Hausmann
    Dada Cino
    (1920) 31.7 x 22.5 cm.
    [Click on image to see enlarged view]
  9. Perhaps the epitomes of Dada photomontage are the large works with a huge number of fragments spilling forth in abundance to the point of disorder. Victor Burgin calls attention to Jacques Durand's "Rhetoric et image publicitaire" where he cites the classical figure of epitrochasm or "abundance" in which "the relations of identity and opposition are not only absent, they are denied." [8]
    3. Cascade

    Grosz/Heartfield: Life and Work in Universal City 12:05 Noon, 1919

    Figure 8 –Georg Grosz and John Heartfield
    Leben und trieben im Universal-City, 12 Uhr 5 Mittags
    (Life and Work in Universal City, 12:05 Noon)
    [Click on image to see enlarged view]

    Hannah Hoch: Schnitt mit dem Kuchenmesser Dada

    Figure 9 –Hannah Höch
    Schnitt mit dem Köchenmesser Dada durch die letzte Weimarer Bierbauchkulturepoche Deutschlands
    (Cut with the Kitchen Knife Dada through the Last Weimar Beer-Belly Cultural Epoch of Germany)
    (1919-1920) 114 x 90 cm.
    [Click on image to see enlarged view]
  10. Semantics

    According to one tradition, images make no claims, state no propositions, and therefore to render what an image means as statements of what it says clumsily forces it to perform unnatural acts. Under the canons of high Modernism, the semantics of images would be wholly other than those of texts. The claim of modernism of the Clement Greenberg kind, Johanna Drucker says, was "that visuality was equivalent to muteness, that modern art had fulfilled its teleological aim through achieving a condition of plenitudinous presence, and that all ties with literature, literary modes, or linguistic signification had been severed as part of the so-called autonomy of modern art." [9] Along with that comes the view that images do not refer to the world of common experience, they merely present a possible one. Photographers have been fairly resistant to this view of their art, and photomontage fragments are lifted from common experience, namely the popular media, and mean by standing in for larger texts and images from which they were taken. So one interprets photomontage by tracing the fragment back to its missing page-of-origin and by grouping them by provenance.

  11. Another view holds that images may well have paraphrasable meanings, but images are inherently less determinate, more polysemous than texts, since they are mute and not of the same material as words. They are not language and hence not linked to the whole apparatus of text reading and interpretation. Hence their meanings should be suggested delicately and circumspectly. Certainly some images are quite inscrutable or private. (My working definition distinguishing Surrealistic photomontage (like some of Max Ernst's) from Dada is my relative cluelessness as to what is being "said" as well as the smoothness and seamlessness of the surrealist surface.) Dada, however, is always constructing a perspective, comment on, or stance toward contemporary life. Martha Rosler, herself a contemporary, political Dadaist, writes of the original Dada photomontage as a breaking of the surfaces to reveal the social and moral relationships in Weimar Germany. [10] And Raoul Hausmann, by most accounts one of the inventors of it, wrote in the third person:

    They were the first to use photography to create, from often totally disparate spatial and material elements, a new unity in which was revealed a visually and conceptually new image of the chaos of an age of war and revolution. And they were aware that their method possessed a power for propaganda purposes which their contemporaries had not the courage to exploit ... [11]

  12. John Heartfield was certainly one who did employ photomontage in very explicit anti-Nazi propaganda, but when doing political posters, his main mode of visual statement was the highly codified repertoire of bloody knife, pig, jackel, hanging rope, stacks of gold coins--blunt and powerful, but not "visually and conceptually new." Similarly, the Grosz/Heartfield amputated mannikin with a revolver for a right arm, a lightbulb for a head, dentures for genitals, and an Iron Cross decoration on its chest ("dedicated to the Social Democratic delegates who voted for the war") was exhibited in the Dada exhibition but uses these objects as highly conventionalized tokens. Dada photomontage, however, definitely has playful and enigmatic streaks that do not lend itself to propaganda. It is an inspired amalgam of nonsense and critique, edgy and in your face, sure of the penetrating discernment of its withering gaze and confident of the viewers' abilities to make sense of the allusions and incongruities and to take their meanings. These abilities do not work upon just the formal cues of image, arrangement, texture, color and so on, but tap into a shared knowledge of the contemporary popular culture. At almost 80 years, one language, and an ocean's remove, the diligent work of various scholars identifying bits is indispensible, and, although we may begin with the formal arrangements, we quickly look beyond the edges of the work into the newpapers and magazines which the fragments are fragments of.

  13. The first type is the grid/matrix/table array where the fragments, being of roughly equal size, appear to be equivalent "panels". The layout suggests either a table (grouped by likeness according to row or column) or a comic from the comics page. In this case, we tend to read top left to bottom right and to look for some sort of sequence or causation. In the case of Schwitter's Handlung ( Figure 1 ), the reference to Egypt in the time of the Pharoahs combined with medieval and modern images of women suggest some sort of panoramic sweep through time, but the panels begin with a scene of movie-making and then pass to a portal of a cathedral (Rheims?), then more or less alternate medieval images (angel and madonnas) with contemporary fashion illustrations, timeless slogans and a contemporary street scene. The effect is not of development or progess but "immer so". The equivalence of the "good women" images is underlined by their gazes, which are downcast and to the left to various degrees. What sort of time-stamp the barouche makes is not clear to me. Given the date (ca. 1922) the composition might be a charter for a certain kind of normality, stabililty, continuity in the admiring depiction of women (in womanly costume and role, to be sure).

  14. The second matrix (Figure 3) is from virtually the same year by Hannah Höch and deals with the contents of her own personal "house" as it incorporates slogans and profundities uttered by many of the leading Dadaists (Arp, Hülsenbech, Schwitters, Johannes Baader, Hausmann, Goethe, Nietzsche, and Friedländer). These deal with time, death, and various Dada paradoxes. Some of the images seem to be bits for a scrapbook or album–photos, beetles, cross-stitch, stick figures, and chalk drawing of a tree leaf–and to suggest a theme of personal development. The clock with no hands, however, puts quick stop to thoughts of chronological sequencing. Scholars have noted puns (e.g. cross and cross-stitch) and other groupings and correspondences. [12] It is a rich composition with which and upon which to reflect.

  15. With the schematic group, the principles of the layout seem clear enough, but their semantics may be somewhat enigmatic. The first of these, Figure 4 , uses Gerhard Hauptmann's head, minus his signature shock of white hair, as the setting moon. The other identifiable person, anyway, is that of John D. Rockefeller, and the body with the gorilla head has the head of a girl with hair streaming behind her as she runs. The splice makes her/it one of the three left-facing figures. The legs and feet of the man with serape were clipped from a woman depicted in Die Dame August, 1920. So we have major gender- and species-splicing here (with Höch, almost always male head onto female body). The title refers to a popular song which proclaimed that Germany, though in apparent decline, would rise again. Hauptmann had endorsed the Weimar government and was regarded as having sold out by the German left, among whom the Dadaists were certainly to be numbered. I suppose the scale is there to measure decline. Is it the signs of returning glory that these discerning financiers are looking at?

  16. Marlene , which was done in 1930 when The Blue Angel was all the rage, literalizes "putting woman on a pedestal" (with a concommitant rearrangement of her body). The little men look on with wonder and admiration, while she smiles like the Cheshire cat on their devotion. This does depart somewhat from earlier, classic Dada flatness (or non-space) in its rather complicated gazed and perspective lines where we are placed high in the superior space of the air (along with Marlene's face). "Marlene" recalls a line from Hausmann's "Definition der Foto-Montage": "The ability to manage the most striking contrasts, to the achievement of perfect states of equilibrium, in other words the formal dialectic qualities which are inherent in photomontage, ensures the medium a long and richly productive life span." (cited from Richter, p. 116.) In context, Hausmann is talking about the photomontage fragment as representing not only its object but the camera's eye view of the object, resulting in multiple angles and distances of the viewer from the objects. These do not cohere in a single or even plausible sequence of viewer positions; rather, the scene, as here, balances; physical space and magnitude merge into ideological space and value.

  17. With the Swirl pattern, diagonals set up lines of travel for the eyes, often with a center-periphery layout which is sometimes called centrifugal. Das Kotsbild is Schwitters again on women, this time on their professions ( Frauenberufe ), with the phrase Frauenberufe next to and parallel to the word KOTS ("vomit") on the one hand and Hundehalsbänder ("dog collars") on the other. (At right angles to this array is Samishgares Rindleder ("chamois skins"). Taken together, these can make a case for anti-feminist revulsion on Schwitter's part (assuming these words are "uttered" by Schwitters and not cited as "in the air.") Dorothea Dietrich in fact makes that case. [13]

  18. There is some reason, though, to hold off this conclusion for a while, partly because it is based on taking one part of the image as dominant. The center of the composition is at least shared by the Polish one mark note (with its two female portraits) and the "Anna Blume" tag. Anna Blume was an imaginary beloved to whom Schwitters had written some Dada love poems, which were an immediate international sensation. And there are again bits of fashion catalog and high fashion strolling with your gentleman. One might read Das Kotsbild as displaying the options and attitudes toward women in early Weimar.

  19. Although very similar in visual dynamics, Dada Cino is surely more exuberant and triumphal, carrying on the "Dada Siegt" theme that is the title of another of Hausmann's photomontages, this time with some sort of tank or personnel carrier. Inscribed on it is a note to "Lieber Kurt Schwitters" joining Merz and Dada and the montage has Schwitters' fashion heads along with Höch's dancing ladies (and a reference to Picabia's journal CANNIBALE). Some critics see the whole Dada effusion rising from the source of all human life at the bottom of the piece, but Hausmann's anatomical cut-aways more often signify the penetrating gaze of Dada (and in any case, the child is still in utero). The "Cino" theme is said to indicate Hausmann's desire to capture the experience of cinema in photomontage. [14]

  20. Already in the Swirl arrangement, individual fragments are losing definition; even greater numbers pour forth in a Cascade, which profusion Victor Burgin reminds us, was called Epitrochasm , the classical rhetorical name for one figure of arrangement, namely, one of profusion to the point of disorder. [15] Burgin is following Jacques Durand here, whose point is that the profusion does not lead the eye to groups, oppositions, parallels, or lines of action. We are here at Mark Bernstein's wilderness. Epitrochasm says "Not to be read in order, or exhaustively."

  21. Life and Work in Universal City 12:05 Noon in general celebrates the prolific new American culture industry the exotic new world of cinema, which, though "Universal," is strongly American The piece begins with a Grosz sketch of a flow of unpleasant looking people, but this is overlaid with a snowfall of publicity snippets, actor's faces, logos, slogans and sayings (what could "Son of a Gun" (a movie title) have meant to them?!) along with markers of the technology that made it "Universal" (or, universal at 12:05pm--a time with complex Nietzschean resonances for Hanne Bergius). [16] It is an exuberant urban wilderness for exploring.

  22. With patience, disorder sometimes begins to sort itself out. Schnitt is widely recognised as Höch's masterpiece and the astonishing portrait of the life of an era–its ethos, its enthusiasms, its major figures, its vitality, and its tensions and contradictions. It is the object of a full-length study by Gertrud Jula Dech, who identifies every fragment she can by content and source, groups them into five major content groups (buildings and cities, animals, machinery, people, words and letters, etc.), with three major thematic clusters (eyes, seeing, and gaze; gender and genderblending; and movement and balance) and takes us up and down, back and forth in the picture. [17] "Schnitt" would make a wonderful imagemap with Dech's identifications linked to each identified item. There are clearly plenty of paths and traversals through this "site" that would prove striking, thought provoking, and finally illuminating.

  23. To recapitulate to this point: Dada photomontages are like representations of some hypertext sites with all of the targets of the main links turned up, as it were; on the usual hypertext page, the source links do not reveal much about their targets--the targets are face down and clicking on one turns it up. On the Dada type site, some of these links will be unexpected--computable, given enough wit, knowledge, and the right frame of mind--but not expected. Second, these fragments are not just tokens for the larger image or text from which they were taken and to which they point, but they are arranged into patterns and configurations that are themselves integrative and meaningful. As Benjamin Buchloh says, "the network of cuts and lines of jutting edges and unmediated transitions from fragment to fragment was as important, if not more so, as the actual iconic representations contained within the fragment itself" (p. 64). A website, after all, is not just a list of all the links made from at least one page. Having outlined Dada photomontage in some detail, we can trace its reappearance as reitique and recoding of signifying practices in the sitemaps of recent web pages.

  24. Web site maps

    Sitemaps (by which I mean graphic representations of the structure of a site where the subparts are "hot" links to subportions of the site) can serve at least two purposes: they enable people ("users") to find things and get them up on screen (aka "navigate to them"), and they can model the relation of the parts to each other and the whole graphically. The first purpose is best served by using a standard schematic format, usually employing the by-now-thoroughly-dead metaphor of files and folders. The second purpose profits from original use of graphic signifiers, but risks unintelligibility. For this reason, writers often supply a standard hypertext table of contents along with their imagemap. One can have a graphic that entirely serves the second purpose: it would not have links (i.e. would not be an imagemap) but would represent the parts of the site and their relations. You just couldn't navigate with it. It would be like Bernstein's diagram of "Hypertext Gardens." [18]

  25. It is possible to have an "abstract" site imagemap, where we have a basic indication of parts and their relation to the whole, but no metaphor. If you have a tree with links as leaves, you have a metaphor; but Figure 10 shows an almost Islamic refusal of figuration:
    Mola Project top Page

    top Page of the Mola Project

    Figure 10

    One might suppose that the five top squares would be the main subparts, and that would turn out to be correct. The squares more in the background, however, are split into many little hotspot links; in fact, the map has 73 links from it. Because the fragments are not taken from any place recognizable, they do not function as the fragments do in Dada photomontage. The Mola Project is very cerebral experimental hypertext; it does have a corresponding text-based page, but that is no more perspicuous that this image, in that it is continuous text in which every single word is a link to some place--solid blue underline. In fact, except for a scattering of small images, the other pages on the site are all text, all interlinked, all blue (to start with). Each page links to 6 or 7 other pages in the set (on average) so the site is a dense network--but one without head nodes or hierarchy (by design, to be sure). The site offers no perspective or map of itself--only the five squares of the quilt-mola. It is often said that the essence of hypertext is choice, and normally one has some reason for choosing--a hunch, a stab, a clear indication. But here, where there are many links from one apparently continuous piece of "fabric", you have not a clue what you may be choosing. You are not only in the wilderness, you are in the dark and without a compass. The effect of the text version is very similar, since a hypertext anchor is usually highlighted (in blue or whatever) and can focus the consideration of whether to "go there." When the text is all link, and the links merge into each other, one quickly adopts the strategy of stabbing at one thing or another without much reason. At least with the text version, the links to the pages visited turn color, but with the imagemap, not even that is registered. Moral: we expect links to be visually distinct--to have edges--as well as visually salient.

  26. Thought of (as is usual) as a set of pages linked by hypertext anchors, a hypertext site has the structure of a web or "lattice." This structure can be represented as a set of nodes with the connections between them. (In basic HTML, the connections are one-way). Give the nodes names (or icons) and you have your standard sitemap. It represents the topology of a site, not the topography--nodes are not distinguished by size or proximity/distance. But when making a map with a drawing program, it is very tempting to use size, placement, color, thickness and nature of line to create a space with ranks and messages beyond the web of connectivity. [19] One could make up one's own little coding system ("text files are green"), ("large type is major node"). That is about the level at which we start with the Dutch group V2's opening page: [20]
    V2 Home Page ImageMap

    Homepage of V2

    Figure 11
    (white boxes added to mark hotspots)
  27. This has very much the look of a schematic diagram--center and periphery. All phrases are hot and many drawn shapes are as well. The basic grid is slightly warped or melted at the top, allowing "DEAF98 THE ART OF THE ACCIDENT" to show (push?) through with top billing. The warp has the effect of propelling this schematic toward metaphor--it is not merely the good site designer's earnest effort to draw clearly the layout of the site (it does come with a key to colors--file types) but in addition reminds us that it is a rendering of structure in a particular medium (or pseudo-medium) and has elements of nonschematic, idiosyncratic use of shape and placement. (Historical note: The DEAF conference no longer being current, it has been removed at the top and the pattern straightened.) Below this imagemap on the V2 splash page is a straight text menu dividing the site into three heads (V2 Organisatie, V2 Archief, Free Zone) with sub-links.

  28. "My Body" Site Imagemap
    Shelly Jackson: My Body

    Figure 12
    (white boxes in original)
    (original is 50% larger)
    Shelly Jackson's imagemap of her site "my body" [21] is a chalk-drawing outline of her body with labelled boxes around body parts that link to pages of text with the body part as topic. This imagemap is a coherent single image with its own meanings, not a schematic, but it is a nearly literal map of the site, which is made up of pages of thoughts, memories, stories about the various parts of her body. It refuses to be a perfectly accurate or faithful map of all the links from the top page, since some parts are linked without being labelled, and the boxes only roughly overlay the body part. But what it does tell us is that the parts of the site have no more narrative order, groupings, or thematic oppositions than one part of the body has to another.[22]

  29. When the image represents something other than the the topic(s) of the site, then we have a visual metaphor for site structure (which of course is not necessarily profound). Consider the imagemap Scott Kerlin makes of his VLO--Virtual Learning Organization–a term that desperately wants some sort of fleshing in. [23]
    Hometour Site Imagemap

    Homepage of Scott Kerlin's Virtual Learning Organization
    Figure 13
    (white boxes added)

    Kerlin creates a floor plan of a one story structure rather like a converted elementary school. All of the image anchors are bits of text (and interestingly, just the text, not the entire rooms, are "hot"--Kerlin assumes you will position your mouse on the words, as in continuous text). "Well," you may say, "but this is little more than a seven-by-seven table." But it is considerably more, with central points, an entrance, proximities, and much of the content of the academic version of life (no place for the Nordic track, the motorboat, or the Nin64). In fact, the "house" metaphor allows him to include his hobbies and interests (even a little corner room for the wife) in the way a table would not (or else it would look like Borges' Celestial Emporium of Benevolent Knowledge.). Further, Kerlin sustains the "subdirectory-as-room" image by repeating the room outline on several of the "go-to-room" entry pages. Here we see clearly the function of the image as metaphor: an image (including a photomontage) is a "take" on some part of the world; an imagemap is a "take" on a site.

  30. A single image imagemap, metaphoric or not, does not closely resemble the photomontages we have examined. Floor plans and body drawings are not Dada. For a Dada imagemap, we turn to a group that invokes the name of Dada ("post-Dada") and that had its primary impact in the 1960's and 70's, but still flies the flag of Fluxus on the Web. [24] Here is an site imagemap adapted from an old poster by George Maciunas, their leader until his death in 1978:
    Hometour Site Imagemap

    Homepage Imagemap of Fluxus

    Figure 14

    Here many of the cells and sectors are hot links. The arrangement suggests a roulette or game show spinner and the periphery a board game. The inner sectors straightforwardly take you to subdirectories with the indicated contents–the expected contents of a multimedia installation/presentation oriented collective. The symmetry, strong colors, and stylization all suggest a belonging together that is not exactly matched semantically. What, one may ask, does the box of matches have to do with someone self-administering an enema, a piece of video detritus, or some unintelligible letters? If the gameboard design is a metaphor for a site, it is one which readily accommodates the arbitrary and the incongruous, that has neat squares offering and answering unknown questions, and that confidently mingles bits of order we can grasp and use with bits that we cannot. It is eminently reasonable to call this a Dada site map (or, a Dada sitemap of a Dada site).

  31. To look a little more fully at how graphic site maps can function in web design, we will consider another site that goes by the name Dada Net Circus, (even though that has become the name of an entire performance art group). The main hypertext site is one by Jim Clarage called Click Me; it branches into four main "rides" as Clarage calls them, as indicated in Figure 12:
    Links from Click Me

    Tantalus datanetcircus Fitness Channel Rocco Rides Again Click Me splash page

    Figure 15

    The four rides illustrate different tactics for different content. Ride 1 is a reworking of Faust in which Mephisto offers Faust unlimited online service--it is strictly sequential and has no hub or imagemap, since the author does not want you to be able to enter the ride in the middle. Ride 3 ("Rocco Rides Again") is a string of episodes in the career of the porn star Rocco Siffredi. It imitates the non-consequential episodic structure of porn video: entry at any point is the same as another--so Clarage gives it a traditional (and decorous) Table of Contents. Ride 4 does have a little imagemap menu of 9 cells where cells are linked to pages featuring a larger version of the image in the cell. Such a matrix format for a top page occurs fairly often, usually with an understood sequence of left upper to right lower (if you want to go in sequence from the top). Quite often the 3x3 grid also has some coherence in layout. In this case, the "ride" represents the experience of channel surfing or switching of shows on the "Fitness Channel." Order of page does not signify sequence or progression here, for the pages give contradictory pitches on how to solve the "problem" of women's fat. The main principle of sequence is an intensifying hostility to woman's body which culminates in panel 8. The effect is somewhat confusing, given expectations of orderly navigation set in Rides 1 and 3. The theme of Ride 2, "Dada," is the experience of surfing itself, and it is Dada through and through.

  32. In "Dada," there are two pages of "reasons" posed to the reader as possible answers to the question, "Why do you do what you do when you do the net thing?" Each page has about a dozen links, one to each answer along with a picture illustrating the text. Here the guessing game aspect of hypertext is in the foreground, since the information derived from clicking on these links probably will not help clarify the question or assist in an answer and will provide further evidence of the reader's addiction to clicking. On the first page one suggested answer "keeps you up all night" combines with an old photo of a naked girl in her room with a drip coffee maker--and the target is a larger version of the image --as is the case with the first image of the late 1940s girls displaying their "new [vacuum-tube] computer," and with "peeping in celebrity windows" of Tonya Harding on her honeymoon. Here or on the second reasons page we find

  33. There is no narrative sequence here, or even staying on topic, no filtering or exclusion of the plethora of information just one or two clicks away. It is a schnitt cut through the sagging belly or brain of the addicted surfer in this last/first epoch of the virtual age. This is major epitrochasm, where to be sure the abundance in all directions is the work of surfing--the abundance of individual sites is multiplied by this clicking in all directions. The unsorted cascade of possible answers and illustrations portrays the in-depth probing of media news and talk shows as profoundly frivolous, with each answer getting its equal two instants before being replaced by the next. It is interesting and very unusual that most of these answers are hypertextual as well as conceptual dead ends; they require the use of the back-button to get back to the "reason" pages.

  34. In the "Dada" ride of "Click Me," Clarage seems to have epitomized certain qualities of "exploring" which turn the Web into a firehose of excessive access but which is the product of our own making as enabled by hypertext. (Some of the other seductive attractions are worked out more deliberately in the "Tantalus" ride as one version of the Faust's pact.)

  35. As noted at the outset, imagemaps or image arrays may be offered as navigating aids (or control panels), or they may map conceptual relations among the component pages. A step further in the direction of Dada is taken when the map itself becomes unstable--that is, when it changes from time to time, or when clicking on a spot or square does or does not take you anywhere. This sort of unreliable array is a design feature of the group äda 'web's work and is pursued with rigor by Vivian Selbo in her äda 'web project "Vertical Blanking Interval" (mounted online in December 1996 and now maintained in an archive by the Walker Art Museum). [25] It is thus very close in time as well as technique to Clarage's "Click Me:" both use screen captures of TV screens, especially advertisements (media bricolage) and both use timed automatic refresh of the screen. The transitions are abrupt, like the cut edges of Dada photomontage, not fade-dissolves.

    We have been pursuing the notion that certain Web hypertext juxtaposes in time what Dada photomontage does in space. "Vertical Blanking Interval" merges the two axes: when you click on one of the pages in the array, it changes (or else it changes a certain number of seconds later on its own), so that the spatial pattern of images changes over time and, to a limited degree, it is redesigned by the act of clicking. We cannot of course capture the changing array, but here is a snap shot of it soon after loading.

    Early State of Vertical Blanking Interval's Splash Page

    Splash page of Vertical Blanking Interval--early state

    Figure 16

    On first sight, we may try to work out a reading of the composition based on the semantics of the matrix noted above, but as soon as the individual cells begin to refresh themselves with new images, or you trigger a refresh by clicking on a cell, we quickly abandon attempts to read the array pattern. About every half minute or so there is a general refresh of the array which produces a number of new images and the impression of one or two images jumping to another place; overall, the sense is one of non-repeating randomness. In fact, however, there is stack of about 50 images than run through in the same order in each cell, starting at different points in the stack but always ending with the "submit" button screen that actually does "take" you somewhere beyond the changing 3x4 grid into single screen hypertext ("mind the gap") and a more usual set of paths and choices. "More usual" however is not completely so, for these later pages appear to offer more choices than they actually do: there are numerous directional arrows and image links, but many go to the same place--again, meaningless choices. The matrix array of choices is not just or primarily a signifying practice of the Net, of course; it strongly evokes the arrays of the big TV game shows and the player's decision whether to risk her bundle on one category or another.

    One of the places the central dispatching "Mind the Gap" page may take you is a tight serial tunnel of four pages explaining what a vertical blanking interval is. These pages do not look like continuous text, and you may have ceased to expect any continuity at all by the time you reach them, but they can certainly be read as such, and provide an anchor and initial point of departure for the various associations around the theme of "gap." The facts are little known and interestingly include a proposed use of the vertical blanking interval to transmit internet data. So it is a "meta" reflection on the Internet medium and on its own gap which it creates between the image or image-text and its original employment as part of an advert. The effect is to set vertical blanking interval (which is a technical capacity that can exploited and sold), the reader's experience of the page, and citationality as analogous, equivalent, or identical.

  36. A variant of and slight advance on this changing matrix is one at ctrl-alt-del, which opens with an 18-panel array that changes as soon as you move your mouse into any cell of it.[26] All cells change before you can read the URLs in the status window, so the only way to "choose" is to click on whatever has turned up under your mouse, or move the mouse randomly, hoping for something more interesting (but how would you know, anyway?) "With hypertext, the user is in charge." Right?

    Enough of matrices with their apparent stable set of alternative choices! Changeable buttons and icons, buttons that move, that sometimes go one place, sometimes another, sometimes nowhere at all--these have become signature traits and flourishes of Let us turn to a more open freestyle placement of fragments in space--the splash page of

    Early State of's Splash Page

    good luck reading this

    Figure 17[27]

    The cross-section-marked background suggests an abstract cyberspace in which a few pulsating jewels are placed. These prove to be hot and to provide brief moments, from which one returns ... and finds the scattering of jewel-gifs to be altered--some have moved, others are new. (Mouse over this if you have not done so already.) This happens after a certain interval or upon reloading the page (as when returning to it). As with Ctrl-Alt-Del, the notion of choice from among a set of stable alternatives is subverted and, for that matter,so is the notion of significant pattern or placement. The site is hard to grasp because too much is changing in a world where the bounds of variation are not apparent. There is an alternative Contents page, which is a simple listing of the 40 small sites, each with its icon, and the main page offers little more in the way of indicated structure. In fact, in one sense, it offers less, since the total set cannot be seen at a glance, but the entire list can be experienced if you wait long enough (and take notes). The effect is of profusion (like the epitrochasm of the Dada cascades) but it is achieved more by suggesting random scattering than by great numbers of icons presented at once.

  37. The Great Dada Sitemapping Machine. The grand prize for the descendants of Dada--The DADAMAX--goes to Mark Napier who offers the world the services of Shredder to Dadaize Web pages.[28] Shredder will restyle a page differently each time you apply it. Here for example is one restyling of the September 1999 issue of PMC:

    The links are still functional on this page and can be read in the status bar: it is just the relations of the links that are presented in an unusual manner. This page, and the unlimited number of other pages like it that Shredder can make, may carry the Cascade/epitrochasm arrangement as far as it can go. It makes sense that Shredder is a robot, since pages, especially Top pages, change frequently; we will always be able to knock off a few versions of the new page to choose a best one.

    Shredder of course is wholly dedicated to advancing information entropy, which in its most developed form (say, in has no Table of Contents and provides neither the means nor the basis to choose where one "goes" on the site, and hence of "where one is." A more mixed form (with which we shall conclude) is a site that seems always locked in a mortal struggle with cybernoise and degeneration--daily experiences on the Web, but ones we largely choose to ignore. The cybernoise of includes scraps of email, technobabble, warnings, contract exclusions, and other textual detritus. Degeneration is a major theme of the site, and an assault on the viewer's sanity which keeps struggling to grasp what it assumes is the site's somewhat idiosyncratic structure. They frankly tell us

    +++++++++dÊÊ-gÉnÉ+++ration+++ zeite+D+zéigners scientifically randomize position and relationships of all elements on the page, totally releasing control of any form of layout, whatsoever ..... dee-generation zÉites metaphorize consistent thematical overuse of rejected visual tokens to disgust and repel, thus creating a near-death experience for zÜrfers by++++++

    their eyeballs

    1gräBBing them by

    aND rhytmically hammering nail of visual knowledge into undeveloped cranïïs, by exposing to imægery beyond comprehension, with imploding cohesiveness of anti-climactic multilayers, developing unique experience, simulating n-dimentional field in the context of the one1-dimentional cortex, randomly supressing or stimulation urge to operate "back" button either by applying pointy-ing device with certain level of clickability, or leveraging knowledge of implicit k-level vocabulary which+++++

    This is not quite the way this material is displayed on their pages (and it continues), but the main tendency is clear. The computer we find epitomized on this site is unruly and unreliable, subject to runaway processes (including artificial life simulations) and various fragmentations, always teetering on the verge of collapse. And never are our strivings for control and meaning more active than when they are threatened by degenerated signal or receiver. We can even make quite a bit of meaning out of these paragraphs of explicit meaning-unmaking, which tells us that our very efforts to restore normal functioning will be used against us.

  38. These last several sites are Dada's progeny in several ways, most notably in their appropriation of mass media images and disruption of their signifying practices. We might call them Net Neo-Dada. Although the Dadists disbanded almost immediately, so as not to become a Movement, their spirit and practices have been revived many times in the decades since the first and last Dada exhibit in Berlin in 1920, notably by Ed Kienholz, Ed Rauschenberg, by the Pop movements in various countries, by Fluxus, and by the Conceptualists in both Britain and the US. Ah, but you will say, the times are altered: the sense of political, social, and cultural collapse that fueled Berlin Dada, the fury at the symbolic orders that had produced it and were attempting to reestablish themselves with minimal alteration--these things have not been repeated. That is true, but also true is that the mass media have continued to burgeon and now inundate the world with text and image to an extent the Dadists could only begin to imagine, all of it functioning much as it did in 1920 to focus and channel desire and to disseminate a whole symbolic world. And this is as true of the new Web medium as it is of the print or broadcast media. But the Web offers special capacities for citation and juxtaposition--hypertext links--and access to it has not yet come under corporate regulation and control. We might here snip and insert a strip of text from "Hausspruche:" "Die WWWelt ist eine durchaus dadaistiche Angelegenheit" (The Web is a thoroughly dada situation)--not, to be sure, throughout, but here and there in exploratory hyperspace.


1 The CAIM Yale Style Guide has come out as a book from Yale University Press, and is available on line .

2 Edward Tufte, Visualizing Information, Graphics Press, 1997, p. 125.

3 Johanna Drucker, Theorizing Modernism, Columbia University Press, 1994, p. 55.

4 Johanna Drucker, The Visible World, University of Chicago Press, 1994, p. 66.

5 Benjamin H. D. Buchloh, "From Faktura to Factography," in The Contest of Meaning: Critical Histories of Photography , ed. Richard Bolton, MIT Press, 1989, p. 64.

6 See Victor Burgin, Between, Blackwell, 1986.

7 Reproduced in Graham Clarke, The Photograph , Oxford UP, 1997,p.116.

8 Durand, cited in Thinking Photography, Victor Burgin, ed., Macmillan, 1982: p. 79.

9 Johanna Drucker , The Visible World, , University of Chicago Press, 1994,p. 4.

10 Rosler's views are summarized by Martin Lister in "Photography in the Age of Electronic Imaging" in Liz Wells, ed. Photography, a Critical Introduction (Routledge, 1997).

11 Hausmann, cited in Hans Richter, Dada: Art and Anti-Art , trans. David Britt (Thames and Hudson,1985), p.116.

12 See Gertrud Jula Dech, "Hannah Höch ist eine durchaus dadaistishe Angelegenheit," in Hannah Höch , Museen der Stadt Gotha (1993): 119-26; and Armin Schulz, "Bild- and Vokabelmischungen sind Weltanschauungen zu Hannah Höch's Collage 'Meine Hausspruche'", in Hannah Höch , Berlinisher Galerie (1989): 133-145.

13 Dorothea Dietrich, The Collages of Kurt Schwitters , Cambridge University Press, 1993: 142ff.

14 Timothy O. Benson, Raoul Hausmann and Berlin Dada, UMI Research Press, 1987, p. 185.

15 Jacques Durand, 'Rhetorique et image publicitaire,' cited in Victor Burgin, "Photographic Practice and Art Theory," in Thinking Photography , Victor Burgin, ed., Macmillan, 1982: p. 79.

16 Hanne Bergius, "Zur Wahrnehmung und Wahrnehmungskritik in der Dadaistishen Phase von Grosz und Heartfield," in Montage: John Heartfield , ed. Eckhard Siepmann, Elefanten Press Galerie, 1977, pp. 43-47.

17 Gertrud Jula Dech, Schnitt mit dem Küchenmesser DADA dur die letzte weimarer Bierbauchkulturepoche Deutschlands: Untersuchungen zur Fotomontage bei Hannah Höch, Lit Verlag (1981).

18One such site-image occurs at the end of the Symbology section of "Distorted Barbie" where it functions as a recap-and-synthesis.

19 Jacques Bertin ( Semiology of Graphics , trans. William J. Berg, University of Wisconsin Press, 1983) speaks of "retinal variables" augmenting the displayed web structure.



22 for another body imagemap where only certain files are linked to body parts, see Seiko Mikami's biotech art.


24 For a history of this group, particularly in its relation to Dada, see The Fluxus Reader , ed. Ken Friedman, Academy Editions, 1998.

25 —even the menu frame on the left side of the main splash page moves: it scrolls itself, so that to use it, you have to click it on the run.

26 ctrl-alt-del.

27 (or