Issue 2 Mapping Conversations Spring 2001

Utterance Is Place Enough: Mapping Conversation

Frances Richard

­The map image is a synthesis of spatially and temporally registered gestalten, each a synthesis in its own right. No degree of thematic constriction can silence the conversation among map signs.
—Denis Wood, The Power of Maps

A map is a picture, a rendering in two-dimensional space of three-dimensional topography ordered through the filter of four-dimensional experience. A map establishes spatial relations between landmarks, commits these relationships to a particular scale, and aligns the resulting picture so that a view­er (reader, orienteer) can enter it. Its use-value lies in articulating connections: "You Are Here," "This Is the Place." Such connections may be situated anywhere along a continuum of quantitative accuracy, from the pristine measurements of a United States Geological Survey plat to the sketch you scribble for a friend so she can find your house from the subway. Like naming and counting, mapping is a method for articulating the existence of things—an operation causing chosen features to rise like newborn islands from the chaotic welter of experience, fixing them in time-space and bestowing (or foisting) upon them a significance that allows these features to be found again, to be approached from new angles while still holding them in the context of previous encounters. Maps index reality in layers.

And conversation? Setting aside the various media in which one might occur—the language of flowers, body language, Morse code—a conversation is a more-or-less unscripted verbal exchange, an ostensibly non-hierarchical talk shared by two or more participants. It is a group endeavor, a multivocal whole that will break down if one voice dominates unduly, or if too many fade away. Conversations do not always happen in real time, but time is inextricable from conversations because they are inherently improvisatory. If every member knows beforehand what each is going to say, speech becomes rote exhibition, not exchange. Sociolinguists are fond of noting that the etymological root of conversation is the Latin convertere, to turn around. To succeed as conversation, shared speech requires such back-and-forth, with its intrinsic unpredictability (being unpredictable, of course, is not the same as being interesting). In their turns, conversants spin a discourse. Their talk creates a mesh of specifically phrased ideas, a tissue of inter-pretation and response in which thought moves from latent to kinetic. Conversation generates reality as temporary and co-operative.

A language is therefore a horizon, and style a vertical dimension, which together map out for the writer a Nature, since he does not choose either.
—Roland Barthes, Writing Degree Zero

Between "mapping" and "conversation" a third term migrates, now acknowledging the inscribed, representative character of the map, now gesturing to the elementally linguistic condition of conversation. How does "mapping conversation" differ from "writing"? Map and written text are artifacts, experience once (at least) removed. The conversation is experience up front. In its objecthood, the map detaches from whatever landscape it purports to render; it slips into its own register as a freestanding unit, a self-enclosed area operating by reflexively validated rules. The tropes of mapping—including scale, legend, color, the use of contour isobars and other conventions for translating three-dimensional elevation into two dimensions—are no less stylized than the parameters of any other discourse. The cartographer has a Nature too, derived from scientific tradition and perceptual habit, more or less unchosen, not the least bit "natural." The map is a particular kind of story about reality.

The cooperative unpredictability of conversation forms around its speakers a terrain. Its words are audible and obvious. The dynamic shapes limned by the progressions of those words are subtler. The modulations of tone, timing, and implication form an uneven surface, a topology of group exchange. To write conversation is a word:word transformation—no extra light is shed on shape. In mapping, terrain becomes obvious, the sine qua non, while the semiotic maneuvers undergirding its representation drift into background. To map location is a space:space transformation—no extra light is shed on language. But map!=land, just as conversation!=words. Mapping conversation differs from writing because it literally indexes language as shape. It arranges words across an armature of space.

I call [these drawings] "narrative structures" because each consists of a network of lines and notations which are meant to convey a story, typically about a recent event of interest to me like the collapse of a large international bank, trading company, or investment house.
—Mark Lombardi, "The Recent Drawings: An Overview"

I am mapping the political and social terrain in which I live. Shall we cut it there?
—Mark Lombardi, video conversation with Andy Mann, February 28, 1997

The drawings are graceful, intriguing, eminently readable—but also pleasing to the eye as abstract webs, spangles, clusters. Done in black ink with red accents, on cream or white paper, they signify "diagram" from a distance, well before anything legible enters the viewer's mind. Anticipating textual focus, the glancing eye understands that some plethora of information is being schematized, presented for maximum accessibility. Unconsciously indoctrinated within the symbol system of (possibly pseudo-) scientific notation, we approach these images expecting to be led through a thicket of data toward a factually supported conclusion. And this is exactly what happens—sort of.

Investing the time required to move from looking at to reading a Lombardi drawing, we depart from the generic "information structure" established by a two-dimensional arrangement of arrows and circles, and enter a "narrative structure" with specific characters, settings, events, and chronologies. What story is being told? Several overlays of interpretation power up together. Most obviously, we are shown that Henry Kissinger, or the Vatican Bank, or Charles Keating, or Flushing (NY) Federal Savings, have been involved in double-dealing, shady finance, inter-national scams. We get dates, patterns of influence, types of transactions. Do we believe?

The answer is probably yes. But this Q&A is a gateway to other layers of the story. Lombardi spends years researching his narratives; every item is culled from the public record as available in books, wire service reports, magazines, etc.; every mark we see is supported by thousands of cross-referenced notes on alphabetized index cards. In other words, like most maps, these drawings posit themselves as truthful documentation, field-tested and verifiable, presented only for the edification of an interested public. In writing terms, Lombardi plays the lone reporter, relentlessly ferreting out secrets and compiling devastating dossiers on the big boys from the modest war-room of his file-strewn apartment. We accept that Kissinger, et.al. are guilty as charged, and the subtextual hum of real-life fact gives the images poignancy and force. But in a sense it would not matter if the whole panorama of Lombardi's narrative were fabricated from his imagination. Just as his sweeping curves and dotted lines adapt a pre-existing vocabulary of corporate graphing, the archetypes of crusading informer and corrupt V.I.P. are readymades from our cultural milieu, and we receive them as such—the artist makes a remarkable case, but we probably "knew" this stuff already. Growing up post-Watergate, inured to the idea that no one gets rich or maintains power without resorting to espionage, intimidation, and payola, we expect revelations of high-level venality. Lombardi's work provides them. The visual elegance of his phrasing is analogous to the oratory of a great muckraker, the comeliness of David as he gears up for Goliath. Their ostensibly neutral or indexical format protects the drawings from looking like agitprop—they are too strange and beautiful for that, and yet their social and political content is direct. Free to enjoy the form and be disturbed by the content, we are brought into Lombardi's orbit as confidants. We become his interlocutors, people on whom the nuance of his project is not lost.

As conversation, then, Lombardi's diagrams speak in multiple directions. Following the patterns he has traced, from Rome to Havana, Iraq to Tennessee, we can practically hear the buzz of covert phone calls and smoke-filled-room agreements, the blur of accents, guffawed satisfaction and growled threats. We are flies on the wall, bugs in the drapes, righteous surveillants. Moreover, Lombardi shows himself in notational conversation with his sources, eliciting and weaving together disparate voices. As audience, we stand before the work in dialogue with it and our internal chronicles, recalling scraps of news we've picked up elsewhere, orienting our attention to take in what Lombardi is telling us.

Mapmaking emerges to facilitate the control of social processes in rapidly expanding societies.
—Denis Wood, The Power of Maps

To create their series "Argument Drawings," the artists Janet Cohen, Keith Frank, and Jon Ippolito map their own discussions, graphing in color-coded lines the twists and reversals of talks on such topics as "the impossibility of distinguishing art you think is historically important from art you like" and "the danger of Bill Gates."1 To diagram casual, private interaction in this functionalist way raises issues about the creative potential of argument, and the shift from chat to chart is a clever joke, a toying with the tropes of number-crunching. But the drawings do not leap into statistical overload. They retain the scale of the friendship they delineate. There are three artists; there are the red, green, and blue lines. It is easy to imagine having this conversation your-self; easy, as a viewer, to pretend, "I'll be the blue," as if choosing a marker in a board game, easy to enter the map without feeling a time/space distortion press in upon your conception of yourself.

Normally, this is not the case in map-making. Cartographic historian Denis Wood describes a difference between mapping and mapmaking, which corresponds to the degree of centralization and division of labor established in a given society. Mapping, he proposes, is a fundamental act of human cognition. Mapmaking is a function of bureaucracy. In mapping, then, the act of tracing and remembering landscape is its own reward, a total process that does not need to culminate in an external, circulating text called "the map." Held within the shared consciousness of people who know and use their land intimately, the knowledge construct also remains at a human scale, articulated in direct proportion to the bodies who have gathered and continue to navigate by it.

During field work in 1989, one Inuk elder told me that he had drawn detailed maps of Hiquligjuaq from memory, but he smiled and said that long ago he had thrown them away. It was the act of making them that was important, the recapitulation of environmental features, not the material objects themselves.2

The Inuk elder describes what would normally happen in conversation, where it is the act of voicing that is important, the recapitulation of thought. To append conversation with a material document like the "Argument Drawings" is excessive, a baroque intrusion into an organic event—which is the artists' point.

The difference between internal memory-map and surveyed document also runs parallel to the anthropological (and literary-critical) distinction between storytelling and writing, wherein the artifactual, written text signals a loss of communal interrelation-ship, a shift away from the conversational nature of oral transmission around the fire. According to this particular master narrative of culture, the sign disconnects from the referent—the hill becomes an inkblot and the lore becomes typography—in order to consolidate power in a stratified society. The novel is said to have created the bourgeoisie; in the same way, the map creates property. But Wood also enumerates instances of what might be called subversive mapmaking, the most famous of which is probably the Peters projection map of the world, which allots to each continent a graphic area proportionate to the area it occupies on the globe. By correctly showing Africa larger than North America and Greenland smaller than South America, the Peters projection "distorts" the familiar Mercator projection, stretching it vertically, as though the landmass had sagged, in order to correct the politically-inflected connotations of a world-view in which North America and Europe are the biggest, at the center.

Both the "Argument Drawings" and Lombardi's work subvert mapmaking in this way. But while Cohen, Frank, and Ippolito keep their maps "actual size" as it were, Lombardi takes on byzantine cases typified by astronomical numbers—of players, of dollars, of references. The Inuk elder who draws his map and then discards it is similar to Cohen, Frank, and Ippolito in that his process retains an intimate scale. He is comparable to Lombardi in that he is managing a gigantic fund of information—a whole region, with its bays and inlets, weather-patterns and wildlife. The difference, obviously, is that the Inuk mapper has no reason to "subvert" a process that supports his existence.

When Wood says that mapmaking arrives with the founding of the centralized state, he implies that the map splits away from the body via the alienation caused by the growth of industrial capital. In capitalism (so goes another master narrative of culture), the work of art becomes commodified, causing a loss of aura that corresponds to the alienated worker's loss of identification with his or her production. The Inuk elder is imagined as existing in visceral proximity to his landscape—in conversation with it—but organized, science-based mapmaking inspects and records that land while no longer depending on real-time, back-and-forth survival on it: In other words, mapmaking is no longer conversational. In the mass-produced map, a represented landform loses "its presence in time and space, its unique existence at the place where it happens to be."3 Subversive mapmaking tries to reassert the map as a dialogic tool, tries to revive the aura of presence and integration symbolized by the Inuk elder's memory. But artists like Cohen, Frank, Ippolito, and Lombardi (or Arno Peters, for that matter) can only do this in a self-conscious way. The exchange enacted by subversive map-making no longer takes place between inhabitant and landscape, but between reader and reader, both homeless to a certain degree, sojourners in alien terrain.

The "Argument Drawings" recuperate presence through insouciant personality, by privileging the artists' own more-or-less unremarkable experience. Lombardi's handling of the problem of alienation and aura is more complicated. Near the end of his life, he was experimenting with using the Internet as a research aid, and with computer-generated versions of his drawings. The exponential increase in archive-access and the corresponding facilitation of copying and adding to existing pieces would have altered his process significantly.4 But perhaps more importantly, computer-assisted versions of Lombardi's works would have eroded their nature as unique, handmade interventions into the web of sinister transactions that have become synechdochical for the powerlessness of average citizens. Lombardi's drawings testify to the overwhelming task of continually reasserting personal perspective in a political and social environment that is not only rapidly expanding, but increasingly imagined as totally digitized, decentered, and manipulable. By illustrating the difficulty one person has in mastering the movements of global capital, his drawings also remind us that the huge distortions of scale endemic to such wealth tend to scramble into nonsense, unless a single interested intelligence sorts them out.

The Conversation Map system is a Usenet newsgroup browser that analyzes the text of an archive of newsgroup messages and outputs a graphical interface that can be used to search and read the messages of the archive. The Conversation Map system incorporates a series of novel text analysis procedures that automatically compute a set of social networks detailing who is responding to and/or citing whom in the newsgroup; a set of discussion themes that are frequently used in the newsgroup archive; and a set of semantic networks that represent the main terms under discussion, and some of their relationships to one another...The Conversation Map system computes and then graphs out who is "talking" to whom, what they are "talking" about, and the central terms and possible metaphors of the conversation.
—Warren Sack, "Conversation Map: An Interface for Very Large-Scale Conversations"

Gigabytes of streaming data and considered, individual opinion; global networks and personal voice: The Interface for Very Large-Scale Conversations (VLSCs) developed by Warren Sack through the Center for Advanced Visual Studies at MIT attempts to order the cacophony of enormous listserv discussions. The interface consists of four parts, or rather, it offers four interrelated filters through which to sort the accumulated messages of a given newsgroup. The social, thematic, semantic, and "message thread" filters isolate and foreground different aspects of the VLSC content, making it possible for a researcher to zero in on chosen facets of the material while de-emphasizing others.

For example, the archive profiled in Sack's introductory web page, soc.culture.albanian, analyzes more than 1200 messages posted between 16 April and 4 May 1999, a period in which the war in Kosovo was at its height. In the Social Network, we see a graphic model of the frequency with which individual posters spoke and were responded to. With a click of the mouse, we can see who was posting opinions that drew flurries of response, and who was crying in the wilderness at the periphery. Since the Social Network recognizes reciprocal statements only (A quotes B; later B refers to A), those whose messages were not reciprocally engaged do not appear. Basic social math applies—if no one picks up on what you say, you tend to drop out of the conversation. Soliloquy is not "social."

The menu of Discussion Themes and the Semantic Network sort for what is called "cohesion"—instances of verbal adjacency and association that suggest thematic and semantic import. Algorithms for distilling these "lexical cohesions" are derived from computational linguistics and quantitative sociology:

Specifically, two terms are "talked about in similar ways" if they are often used with the same verbs, appear together with the same nouns, and share a large number of adjectives with which they are both modified... If [two terms] are used in similar ways by the discussants (e.g. "You're wasting my time," "You're wasting my money," "You need to budget your time," "You need to budget your money"), then the two terms will show up close to one another in the graphically displayed Semantic Network, and so indicate the presence of a literal or metaphorical similarity between the terms (e.g. "Time is money").5

In the thematic and semantic panels, the problem of schematizing meaning is solved via simple grammatical parsing, where the equation "time is money" takes as its least common denominator the modular interchangeability of the two nouns. In the last of Sack's four display panels, the Message Thread Archive, the schematic operation is more complicated. Message Threads symbolize the linked progressions of statement, response, and counter-response that compose evolving segments of the VLSC—this panel offers a chronological overview of the entire archive. Here the mapping procedure moves beyond linguistic adjacency—which comes bundled, as it were, with its own application for sorting, i.e. the rules of English grammar. Diagramming the relationship between successive messages, Sack finds himself fully immersed in the unpredictable realm of actual conversation, and the organic nature of his explanatory metaphors reflects this:

Conceptually, a thread is a "tree" in which the message is the "root" and links between the responses are "branches." Graphically, a "thread tree" can be plotted as a "spider web" in which the initial post is placed in the middle, responses to the initial post are plotted in a circle around the initial post, responses to the responses are plotted around that, etc. One of the nice features of plotting the "thread trees" as "spider webs" is that, at least in theory, any size "tree" can be plotted within a given amount of space.6

In other words, by playing on the formal similarity between different kinds of ramifying structure (root and branch system; spider web; they-told-two-friends-and-so-on) very large-scale meaning can be shrunk to fit the screen of your Palm Pilot. Underlying this concept is not only a practical methodology for translating prose into symbol, but an implicit assertion about communication as a sequence of traceable pathways, issuing in more or less orderly fashion from a single, knowable starting point.

Sack does not offer a rationale for the practical applications of his program—perhaps because, to anyone who has tried to follow a newsgroup, its uses should be obvious. Critics, theorists, and students of any stripe who want to digest the content of a particular VLSC must quickly be drowned in data without some formula for prioritizing what they read. Net users may be pioneering technological practice, but they are also in the same position as every adventurer in new territory since the Sumerians. Without a means of recording acreage and bushels harvested, the management of a population explosion across a fertile delta is impossible—hence clay tablets, algebra, and (it is postulated) some of the first maps. With only sail and sextant, the ocean must remain for generations splotched with blank swaths, where there may be monsters. Imagine mapping the entire Atlantic coastline from Maine to Florida, traveling on foot, with handheld instruments—as the US Army Corps of Engineers did in the 1850s—to such a degree of accuracy that satellite images laid over them fit precisely. Consider the dizzying number of measurements, calculations, and notations made by surveyors that did not need to be made again in future years by naval officers, fishermen, shipping merchants, and wayfarers—travelers whose transit along the eastern seaboard had stimulated, and now benefited from, the development of the Coastal Survey maps.

The polydirectional exchange of listserv conversation is a comparably emergent phenomenon: Sack's project unfolds a tiered grid on which this collective polemic can be tracked. Electronic communication is often theorized in terms of a return to epistolary or conversational consciousness, and the opportunity of discussing, say, the Kosovo situation with political scientists, Balkan historians, NATO-watchers, Albanian teachers, and Serb journalists represents a previously unimaginable crucible for spontaneous intercultural and inter-disciplinary debate. The existence of such a collective is so fascinating that the interface seems transparently beneficial, a labor-saving device without which important knowledge would smear into static. Sack's high-tech browser and Mark Lombardi's painstakingly low-tech works on paper thus perform similar procedures on the information glut, but their interventions point to opposite feelings about that information. The group Lombardi examines is a suspect elite, and the conversations are presumed to be exploitative and self-serving, ripe for the whistle-blower. News-group and chat-room speech, in contrast, is imagined as vox populi in action. The VLSC map does not expose a closed coterie; it expands an egalitarian fellowship.

Nevertheless, the much-touted democratic connectivity of cyberspace is a prime example of Denis Wood's rapidly expanding society—that is, a society in which mechanisms for stratification develop apace with the logarithmic expansion of enfranchisement. We are already entering a period where ubiquitous access, nonstop e-commerce, and increasing regulation encourage nostalgia for the rough-and-ready days of geek/hacker prospectors and cyber-cowboys. New frontiers do not resolve into habitual settlement without new maps, and the despoliation brought by crowds is only lamented after a site is on the map so that crowds can find it. Whether you view new maps as helpful tools, as Sack does, or weapons of control, like Wood, or even as organs of resistance à la Lombardi, depends on your perspective as a dweller in the opening environment.

Utterance is place enough.
—Ralph Waldo Emerson

As Lombardi's phrase "narrative structures" tacitly admits, the mapped conversation is a constructed fantasy, a quixotic and intentional mix-up between quantitative and qualitative analysis. Maps belong (or seem to belong) in the domain of numbers and objective physicality, while conversation is a quintessentially subjective, immaterial process. It may be that projects juxtaposing the two invite failure, since they attempt to index the ineffable. But it might be said that we try to do that all the time. Mapping conversation is more unusual, but no more absurd, than composing written descriptions of smells or taking photographs of the Grand Canyon. In a sense, the scalar audacity with which Mark Lombardi and Warren Sack seek to bring their vast fields of study into tangible proximity with individual readers exemplifies the hubris and pathos of all signifying—art, politics, and utterance included. If representation solidifies one and dematerializes the other, what is the difference between a moment and a place?

  1. Janet Koplos, "'The Argument Drawings' at Wynn Kramarsky," Art In America, November, 1997, pp. 130-31.
  2. Wood, p. 40. The author is quoting research conducted by Robert Rundstrom.
  3. Walter Benjamin, "The Work of Art in the Age of Mechanical Reproduction," in Illuminations, (New York: Schocken Books, 1968), p. 220.
  4. Lombardi discusses these concerns in a videotaped studio visit with Andy Mann, February 28, 1997.
  5. Sack, www.media.mit.edu/~wsack/toc.html (link defunct—Eds.):"Detailed Introduction"
  6. Sack, ibid.

Frances Richard is a writer who lives in Brooklyn. She is a frequent contributor to Artforum and the non-fiction editor of the literary journal Fence.

Cabinet is a non-profit organization supported by the Lambent Foundation, the Orphiflamme Foundation, the National Endowment for the Arts, the New York City Department of Cultural Affairs, the New York State Council on the Arts, the Katchadourian Family Foundation, and many generous individuals. All our events are free, the entire content of our many sold-out issues are on our site for free, and we offer our magazine and books at prices that are considerably below cost. Please consider supporting our work by making a tax-deductible donation by visiting here.