The following essay emerges from my work on the Grub Street Project, a digital social edition of eighteenth-century London (on the social theory of text and editing, see McGann 1983, McGann 1991, McGann 2010, and McGann 2014). By mapping the city's print culture, literature, and its networks of people and their trades, the project aims to create both a historically accurate visualization of the city's commerce and communications and a record of how its authors and artists portrayed it. The Grub Street Project also aims to demonstrate how we might reconceive the notion of scholarly editions, in particular, how a digital environment for reading, annotating, and establishing connections between books, images, and maps published and sold in London during the long eighteenth century can be read as a social and topographical network. It hardly needs to be said that digital methods demand new practices and ontologies in definitions of texts, and accordingly, in approaches to editing. As D.F. McKenzie famously defined them, in the digital age "texts" might "include verbal, visual, oral, and numeric data, in the form of maps, prints, and music, of archives of recorded sound, of films, videos, and any other computer-stored information, everything in fact from epigraphy to the latest forms of discography" (McKenzie 1999, 13). Accordingly, for the Grub Street Project, the challenges of digital editing include not only the usual considerations for textual scholarship but also a process of determining how to integrate facsimiles, scripts, database entries, texts and markup, and meta-data with eighteenth-century maps to create a new form of scholarly edition based on the social networks and spatial-temporal relationships of literature and print culture, rather than solely on individual books or works. 

The Grub Street Project (in progress) comprises maps, plans, and views of the city of London in the long eighteenth century correlated with a relational database of place names, place descriptions, and place name alternates, trade names and locations, London's people, especially those involved in the book trade as authors, publishers, printers and booksellers, and bibliographical data for approximately 240,000 works printed and sold in London from 1660 to 1830 and approximately 120,000 others printed and sold elsewhere that would have been available to London readers during this period. It will incorporate 2,000 xml texts encoded in TEI-Lite from the ECCO-Text Creation Partnership and several thousand more released from the EEBO-Text Creation Partnership, converted to HTML5 with microdata markup. This markup will in turn enable visualizations of people, businesses, places, and events on maps, graphs, or timelines. The intent is that these editions-within-the-edition, along with the maps, images, and other entries in a relational database will demonstrate how a digital scholarly edition can present many linked works, events, places, and people within multiple complex spatial and social environments. This edition of London is intended to allow statistical visualizations of many books associated with map points, but also to enable visualizations and readings of a particular culture and its representations—or heterotopias, to use Foucault's term (1984)—of the eighteenth-century city. The process has entailed significant time devoted to design and development, particularly in working through how to visualize the city and its environs in a combination of quantitative and qualitative views, and in assembling a dataset large enough to provide a spatialized view of the city's trades (particularly the book trade) alongside the inexact heterotopias inherent in literary representations of its people and society (on the heterotopia of Grub Street, see Muri 2011). This development is ongoing, and so certain search and visualization tools are still pending. The following pages summarize some early, exemplary, approaches to manually mapping historical topography and narrative. In particular, I will address some of the limitations of using spatial data to reveal networks of intellectual, social, political, economic, cultural, and spatial elements of eighteenth-century history and narrative, and the approaches taken to address these. Applying Mikhail Bakhtin's theory of the chronotope to studies of Alexander Pope's The Dunciad Variorum (1729) and William Hogarth's series Industry and Idleness (1747) provides some insights into the challenges of visualizing and analyzing qualitative and quantitative historical spatial information.

Mapping narrative: Challenges

The application of digital mapping and geographic information systems (GIS) to research in history and literature has risen dramatically in the past fifteen years, influenced by several developments that have converged in new methods variously called historical geographic information systems (HGIS), the spatial humanities, or geohumanities (see, for example, Bodenhamer, Corrigan, and Harris 2010; Ell and Gregory 2001; Knowles 2000; Gregory and Geddes 2014; Von Lünen and Travis 2013). The so-called spatial turn in the humanities is now a well established movement, largely influenced by Mikail Bakhtin's theory of the chronotope (1981; written 1934 and first published in 1975), Michel Foucault's concept of heterotopia (presented in 1967 and published in 1984), and Henri LeFebvre's theory of the social production of space (1974). Few involved in that first wave could have imagined the extent to which digital humanities techniques might inform spatial analysis. Powerful systems and vast quantities of data are now available to individual scholars on their desktops: in just the past decade computer processing speed, display resolution, and colour accuracy have advanced dramatically; a suite of increasingly intuitive software tools has emerged; and high-resolution satellite imagery has become publicly available online along with application programming interfaces (API') to enable data use and web application development. Over a relatively short period GIS and spatial analysis have spread across disciplines and have proven to be particularly effective in historical studies (Knowles 2002; Raven 2010).

The field variously called literary cartography, literary geography, or literary GIS, however, has remained a marginal and still relatively unexplored avenue of the digital humanities. This is hardly surprising for what is still a new approach to a discipline long rooted in print. Even for historians, where the spatial humanities are more widely applied, spatial studies pose a daunting challenge "for those who are not too savvy in geography/cartography, database technology and statistical analysis," as co-editors Alexander von Lünen and Charles Travis observe in their preface to History and GIS. They explain that although geospatial questions and visualizations are more popular than ever before thanks to Google Earth and GPS technology, recognition and acceptance of GIS in mainstream history is lagging because all too often "GIS and its proponents have so little to offer in terms of intellectual merit, but are asking for so much in terms of learning curve." One "cannot help, but to have sympathy with those historians remaining unconvinced," they continue, "given that many HGIS are about digitization rather than analysis" (Von Lünen and Travis 2013, vi–vii).

To a certain extent this is true in studies of creative works as well. The challenges posed by mapping the imagined spaces of literary and artistic works—which make little concession to measurable coordinates of latitude and longitude—are even more difficult to transcend. How does one create such a map that offers more to scholars than a mere literary tourist's guide or a coffee-table literary atlas in digital form? As Franco Moretti has asked,

There is a very simple question about literary maps: what exactly do they do? What do they do that cannot be done with words, that is; because if it can be done with words, then maps are superfluous. Take Bakhtin's essay on the chronotope: it is the greatest study ever written on space and narrative, and it doesn't have a single map.... Do maps add anything, to our knowledge of literature? (Moretti 2005, 35)

While Moretti's theories of distant reading have been influential in the digital humanities, his famous book has not convinced detractors that maps do contribute a constructive new method to the study of literature. Timothy Mennel, for example, noting that "Moretti seems to be largely unfamiliar with geography as a discipline," calls attention to the work's "less-than-rigorous statistical techniques," and characterizes Moretti's process as "stumbl[ing] from intuition to analysis" (Mennel 2006, 684-686). David Matless comments on the irony in Moretti's earlier Atlas of the European Novel 1800–1900 (Moretti 1998) that "Literature on literature from the discipline of geography is ignored throughout the book" and asks: "Is this book profound, obvious yet profound, simply obvious, or banal? ... Is it surprising that 'The London of silver-fork novels, 1812-40' is focused on the West End, or that 'Jane Austen's Britain' is predominantly southern English?" (Matless 1999, 659-660). While such critiques take issue with what they see as incomprehension of geographical methods or theory, they fail to take into consideration the nature of the questions and valid qualitative interpretations that characterize literary scholarship. As Germaine Warkentin remarks in a more charitable review, Moretti's theory stimulates "pondering of the diverse yet marvellously productive ways in which humans structure knowledge," (2006, 344) despite its limitations in practice.

Certainly as Ian Gregory acknowledges, "there are limits to what we can expect GIS to achieve," and, although some techniques to show temporal change (such as animations) cope with space and time, qualitative issues such as theme are more difficult to address: "the degree to which [GIS] offer new insights into the topics under study has yet to be established" (Gregory 2010, 72). A case in point might in fact be the pilot project Gregory developed with David Cooper, which maps literature of the English Lake District. Using GIS techniques to compare Thomas Gray's account of his tour of the area in 1769 and Samuel Taylor Coleridge's account of his "circumcursion" in 1802 shows that "the places they talk about are largely mutually exclusive" (Gregory and Cooper 2010, 10). The authors integrated a Digital Elevation Model (DEM) with their point data on place-names to allocate a height to every mention of places in Gray's and Coleridge's narratives, distinguished visited places from mentioned places, and then graphed the results. Their findings showed, among other observations, that:

A clear pattern is apparent. [Gray] spends all of his time at low altitudes, with over 60% of visited places being under 100m and all being under 1000ft. Most of the places he mentions without visiting are similarly low although there are also mentions of high places, particularly those over 600m, the higher Lake District peaks. He almost completely ignores places in mid-altitudes. This pattern seems to fit well with the concept of Gray as a Picturesque tourist: he spends his time in the valleys and passes, describing the areas around him and looking up to the high peaks.

Coleridge's pattern is striking for both its differences and its similarities. Like Gray he spends much of his time at lower altitudes but not to quite the same extent. Coleridge also visits places across the height range including a cluster of mentions in the very highest intervals, over 800m when he climbs Sca Fell. While his account is famous for this ascent, it only occupies a relatively small proportion of the heights of the places that he visits. It is also noticeable that Coleridge does not ignore mid-height places. (Gregory and Cooper 2010, 12-13)

In a later article, the authors note that the paths of the two authors overlapped:

there were also points of geographical intersection.... there was a solitary, yet significant, line of overlap on the route between the town of Keswick and the village of Grasmere….The single point of overlap immediately gestures towards the prominent role that this central route played, and continues to play, within the spatial history of the Lake District (Cooper and Gregory 2011, 98).

Technically both sophisticated and innovative, this pilot project for what the authors call a humanities GIS highlights the undeniable difficulties in generating meaningful new discoveries by applying quantitative methods such as GIS to the study of narrative. A sceptic might wonder, why does Coleridge's elevation in metres matter? Do we need the application of GIS to these works to confirm that the major road between Keswick and Glasmere was an important route through the area? Cooper and Gregory articulate a desire to "use GIS technology as a tool for critical interpretation rather than mere spatial visualisation," (Cooper and Gregory 2011, 90) but acknowledge that there are limitations to literary cartography models; however, they also emphasize the foundations these early studies lay for developing methods to facilitate the work of spatial critique (Cooper and Gregory 2011, 104-105). This conclusion must be endorsed: in the digital humanities experimental studies are important, valid, and necessary trials as we test new methods in a still nascent field. We cannot proceed without experiments and testing of hypotheses. We also need to ask—and answer—(to recast Alan Liu's (2013) important question about the Digital Humanities), what is the meaning of a literary GIS to literary studies and textual scholarship? This is no straightforward task, however, and particularly in studies of early modern works, which lack the specificity of precise addresses, consistency in place names and spelling, and completeness of data.

Visualizing eighteenth-century spaces and the Respublica literaria

In 2008 the Mapping the Republic of Letters project at Stanford University experimented with computer-generated visualizations of eighteenth-century communications, based on metadata provided by Electronic Enlightenment. The links to visualization maps and tools have been removed from the Mapping the Republic of Letters site. For a demonstration of the digital mapping technology under discussion, see the embedded video here. The Flash visualization tool is still accessible here. The visualizations attempted to address the question of how we analyse correspondence networks (in this case, letters digitized from modern editions) as mappable data: using the texts and structured metadata from the Electronic Enlightenment consisting of, at that time, roughly 50,000 letters exchanged between 6,400 correspondents, project members set out to answer such questions as "What did these networks look like? How geographically extensive were they in reality? Did these networks connect or overlap? Was there any evolution in the configuration of scholarly networks over time, from the beginnings of the Republic of Letters in the Renaissance to its flourishing in the Enlightenment?" (Edelstein and Findlen 2011)

The resulting visualization tool demonstrates the difficulty of successfully achieving such prospects. To begin with, the data is incomplete, as the maps are able to display only those letters where locations are known, and the dataset itself implies that intellectual communities can be visualized via mapping available letters. As Dan Edelstein and Paula Findlen (2011) acknowledge in their white paper on the project,

One of the first discoveries was actually not what we visualized, but what we could not visualize. There are about 9,000 letters in Voltaire's correspondence, but we are only able to map about 13% of them. The rest simply don't appear because our data is so incomplete: we do not have geographical or chronological metadata for the vast majority of the letters in these correspondences. From a visual perspective, this meant that we need to indicate more clearly what was not being seen and we needed to develop a visual language for representing data that is uncertain.

Johanna Drucker had earlier pointed out that "there is a statistical naiveté that's at work" in the visualizations, which she argued "reifies a notion of correspondence networks according to a completely anachronistic model, as if letters simply travelled from one point to another without circuitous routes, without all kinds of encounters, without all kinds of difficulties that have to do with the experiential reality of the space-time continuum within a historical frame" (Drucker 2010, 20:34) Drucker's critique focuses on assumptions about quantification and visualization, emphasizing that the "reifying rhetorical force of visualizations is so powerful that we look at this and go 'Ahh! That's the correspondence network.' And it so isn't," she states (Drucker 2010, 25:46). It is unclear what the visualizations mean: to play the selection called "Connections," which shows the traffic of letters between correspondents in European cities over a selected period of time, is to view something like a laser light show, with lines in varying intensity and thickness of orange and yellow shooting across the map. Figure 1 shows Swift's correspondence, a graphic ostensibly demonstrating that "Swift's occasional letters to Paris falsely suggest a cross-pollination with French thinkers. (He was writing to British friends there)" (Cohen 2010). This graphic ultimately demonstrates very little about the role that Swift's correspondence played in the intellectual community known as the Republic of Letters; however, it prompts important questions about literary/historical mapping, crucially: What does this experiment tell us about what digital mapping could do, and what it cannot do? Clearly, before we can make sustained arguments about the value of mapping literature, history, and culture in the early modern period, we must find ways to address such limitations as the availability or inaccessibility of complete sets of data and the problematic design of visualizations created from this data. For Edelstein and Findlen (2011), an important outcome of their initial prototype was that "Visualizing degrees of uncertainty has since become a major research question." Drucker calls for an even more formidable enterprise, that is, for a shifting of epistemological ground. Humanistic interpretation, she suggests, "is always about observer-dependent and co-dependent relations between phenomena and their constitution, and a constructivist model. ... [W]e really have to question certain fundamental assumptions about how we know what we know" (Drucker 2010, 24:10–25:46).

Figure 1: Visualization of Jonathan Swift's correspondence as represented by the Republic of Letters visualizaton tool (http://www.stanford.edu/group/toolingup/rplviz/rplviz.swf)

One challenge for building applications in the digital humanities (spatial or textual), then, is to find ways to communicate results that go beyond merely counting—that is, to determine how to present or visualize the data available when the technologies of gathering, storing, retrieving, and viewing them can obscure or distort meaning even while they carry the deceptive implication that empirical knowledge has been demonstrated through computational means. For the Grub Street Project, different layers of information can be mapped, both qualitative (predominantly interpretive), and quantitative, empirical data. The current map viewer, for example, allows historical data such as streets, buildings, and so on, to be layered with tours of literary or artistic works and series. Another view will display statistical and spatial information about London's print culture as represented by the locations of printers and sellers indicated on the imprints of books and pamphlets. The known limitations of the imprint information on eighteenth-century books and pamphlets mean that the data collected for this project provide a very good general representation of London's print culture, but the precise counts of books printed and sold in certain locations must always be considered somewhat imperfect and provisional: often the imprint information does not provide a location, and the booksellers named on imprints was not always a complete list. Further studies can provide corroborating evidence of a bookseller's location in a particular year, but there is much more research to be done to find exact locations for all the booksellers in question.

When I first proposed the digital visualization of eighteenth-century literary networks in 2006, I suggested that a digital edition of eighteenth-century London could help to understand the spatial, social, and material conditions of eighteenth-century London's literary commerce (Muri 2006; subsequently published in Muri 2009). What became clear through the work of theorizing and building this edition was that the potential power of digital maps for spatial studies of pre- or early modern literature, art, and culture might well lie in abandoning many of the tools built for geographers, city planners, industry, and twentieth-century travellers: using ArcGIS, Google Maps, Google Earth, even much of twentieth-century spatial data and graphics, have real limitations when attempting to visualize eighteenth-century London, where a serviceable address was "at his shop over against the Town-House," where Grub Street still existed, and where the London Wall and Temple Bar were still physically and culturally potent liminal barriers between the City of London and "the Town" and Westminster beyond. While Google Maps and Google Earth are becoming increasingly popular tools for "GIS-lite" models, mashups, and applications in the humanities, and while ArcGIS has shown real promise for certain historical questions, mapping historical narrative poses significant challenges. Figure 2 shows the topography of Alexander Pope's "Grubstreet race" (Pope 1729a) using a map from John Strype's two-volume A Survey of the Cities of London and Westminster (1720). This map presents a visualization of the city as the environment with which Pope and his readers were familiar. The heavy black line indicating the London Wall bears witness to the importance of that barrier around the City, while buildings rendered three-dimensionally—St. Paul's Cathedral and other churches, Guild Hall and the Royal Exchange, the Tower, the Inns of Court, Whitehall and St. James's Palace—mark the architecture of consequence and power. Such representations of historical topographies are critical to understanding history and historical literature but lack the geographic accuracy of modern maps and accordingly make it extremely difficult to analyse with precision changes to the topography over time, or often to make anything but qualitative comparisons between maps.

Figure 2: The topography of the "all the Grubstreet race" (Book I, l. 42) in the Dunciad Variorum (1729) shown overlaying A New Plan of the City of London, Westminster, and Southwark, in A Survey of the Cities of London and Westminster by John Strype (1720). Shown here, from left to right: St. James's Palace, Hay Market, Covent Garden, St. Mary le Strand Church, Lincoln's Inn, Fleet Street, Fleet Ditch, Ludgate, West Smithfield, Grub Street, Bethlehem Hospital (Bedlam), and Rosemary Lane. Courtesy of Special Collections, University of Saskatchewan Library.

Typically, historians will address such limitations through a process of rubbersheeting or warping a digitized historical map to aid in georeferencing or georectification (shifting pixels on a raster image of a map to align points to geographic coordinates). Using this method, control points are added to the maps in software such as ArcGIS to align them to latitude and longitude points on a contemporary map. Once the control points are designated, the software warps the original map image to fit the contemporary map projection as closely as possible. This is an imperfect process. For example, the technique was used by the Locating London's Past project (Davies, Hitchcock, and Shoemaker 2011b), which "provides an intuitive GIS interface enabling researchers to map and visualize textual and artefactual data relating to seventeenth and eighteenth-century London against John Rocque's 1746 map of London, published in twenty-four separate sheets, and the first accurate modern Ordnance Survey map" (Davies, Hitchcock, and Shoemaker 2011a). The web-based user interface was created using a Google maps container that mapped data as layers onto the 1746 Rocque map, the 1869–80 OS maps, and Google maps. This approach was complicated by the imprecision arising from eighteenth-century techniques of surveying and printing, in addition to differential paper shrinkage of the twenty-four plates. Pen width of 1 mm on the map was calculated to be approximately equivalent to 2.4 m, and in the cases of buildings, "very little error in initial measurement or final engraving would result in edges wobbling by twice or even three times this amount" (Rauxloh 2011a).

Using forty-eight common points between Rocque's original map and a modern Ordnance Survey map in GIS software to warp and stretch the original image, the process resulted in the typical irregular shape of georeferenced historical maps. Notably, Rauxloh comments that a guiding principal of this project was that its data sets were to be "embedded in a geographical reality," but the survey techniques and coarse representation of structures in the engravings meant that, even though the georeferencing was fairly accurate, "an alley way shown on the Rocque map would not exactly overlie the same alleyway shown on the far more accurate ordnance survey data of some 50 years later" (Rauxloh 2011b). The problem of accurately representing spatial history is compounded by the significant amount of data without spatial information. Searching for the name "Roberts" in the dataset for fire insurance in Locating London's Past, for example, results in 327 hits, of which just 39% can be mapped (Locating London's Past. Click on "Add some data"; then "Fire Insurance"; then search for a name). Viewing the mappable hits on the Rocque map shows the not inconsequential limitations of distorting the visual dimensions of the original map. The first mappable record,

"THE CORNER OF SUN STREET, LONG ALLEY, MOORFIELDS,"
Name: ROBERTS, MARY Insured value: 100 Occupation: SPINSTER

is associated with a point in the middle of the fields behind Bethlehem Hospital (Figure 3). The mapped location is far from the likely location on the north or south corner of Sun Street where it meets Long Alley (Figure 4). Much of the data available to us today is simply too coarse to get a better sense of spatial history than these very general addresses. The Locating London's Past site provides a page with information on interpreting results and several important caveats about potential sources of error, including an incomplete set of place names, automated tagging processes, missing data for some parishes in some datasets, and the proportion of place names successfully mapped. Developments in open source software such as Bernard Jenny's MapAnalyst (Jenny and Hurni 2011; see also www.mapanalyst.org), which identifies inaccuracies, grid distortions, and wrongly assigned control points resulting from georeferencing, show promise for the creation of more accurate maps that could be added to Grub Street resources in the future. Begun in 2011, the British Library's georeferencing project is crowdsourcing location data to make over 50,000 of its historic maps searchable and viewable online using Georeferencer and MapAnalyst. While historical GIS has the potential to produce reasonably accurate results for nineteenth-century maps, at present the process does not consistently produce reliable spatial coordinates for seventeenth- and eighteenth-century maps and, depending on the application, may not add significant value for the necessary investment in time and resources. Indeed, James Raven's admirable and precise mappings of London's eighteenth-century book trade has achieved tremendous results not through automated calculations, but through painstaking studies of the annual London Land Tax Records (London Metropolitan Archives) and historic maps to determine tax collectors' routes and identify places of business by building number (see Raven 2010 and Raven 2014).

Figure 3: The corner of Sun Street, Long Alley, Moorfields, Locating London map (Rocque 1746).

Figure 4: The corner of Sun Street, Long Alley, Moorfields, in A Plan of the Cities of London and Westminster, and Borough of Southwark London: John Pine and John Tinney, 1746, by John Rocque. Courtesy of the Beinecke Rare Book and Manuscript Library, Yale University.

Additionally, the process of distorting maps through georectification/georeferencing may have limited applicability for qualitative visualization or analysis—especially when considering the inexact and equivocal nature of narrative and image. The processes of georectification and georeferencing is based on an ideal of aligning historical representations of topography and geography with twentieth-century projections in order to be presumed scholarly or meaningful, and this ideal in turn can tend to undervalue the qualitative aesthetic and cultural content of the maps. For these reasons I have chosen not to distort the source maps in the initial stages of development for the Grub Street Project; instead, the maps are assigned spatial references as place names and map coordinates stored in a relational database where each unique place ID may also be associated with contemporary latitude/longitude values if known. James Raven (2010 and 2014) has accomplished some very thorough and painstakingly precise mappings of London's eighteenth-century book trade by studying the annual London Land Tax Records (London Metropolitan Archives).

Within and without the city wall: Mapping chronotopes of progress in eighteenth-century London

In his celebrated essay "Forms of time and of the chronotope in the novel: Notes toward a historical poetics," Bakhtin suggested that the chronotope (literally, time-space) is "the intrinsic connectedness of temporal and spatial relationships that are artistically expressed in literature":

spatial and temporal indicators are fused into one carefully thought-out, concrete whole. Time, as it were, thickens, takes on flesh, becomes artistically visible; likewise, space becomes charged and responsive to the movements of time, plot and history. This intersection of axes and fusion of indicators characterizes the artistic chronotope. (Bakhtin 1981, 84)

Bakhtin's "chronotope of the road" as a path-of-life metaphor generally applies to the early modern "progress narrative" (John Bunyan's  The pilgrim's progress (1678); Daniel Defoe's Moll Flanders (1722)), where significant turning points in the course of events are fused with a spatial course or road. Intersections might signify turning points in characters' lives; road markers might indicate their fates, and so on. The hero might be defined by the sequence of guilt→ retribution→ redemption→ blessedness, but might also, in the eighteenth century narrative, be defined by an ironic or didactic denial of the sequence leading to redemption. Where the progress narrative is emblematic, maps are unlikely to illuminate our reading, but where they are topographically specific, we can be attentive to how time, plot, and history intersect with or interact with historical topographies. I propose to adopt Bakhtin's premise as an important foundation to interpreting historical narrative (fictional or otherwise) through contemporary maps, which represent a material history of space with particular social, political, and cultural meanings.

As Bakhtin argued, when "meanings … enter our experience (which is social experience) they must take on the form of a sign that is audible and visible for us (a hieroglyph, a mathematical formula, a verbal or linguistic expression, a sketch, etc.)" (Bakhtin 1981, 122)—a temporal-spatial expression. Maps, then, presented in their original projection and without distortion, provide readers and viewers with glimpses of the lived spaces that early modern writers and cartographers have made "artistically visible" to us. To view literature and visual narrative through contemporary maps allows for analysis of the social foundations of these creative works. Lefebvre's speculation that the production of space takes place in terms of a triad of dialectically connected dimensions—the spatial practices of a society, the conceptualized representations of space by "scientists, planners, urbanists, technocratic subdividers and social engineers," and representational space as "directly lived" through associated images and symbols that overlay physical space (Lefebvre 1991, 38–39)—resonates with this reading of Bakhtin's chronotope. The interconnections cannot be ignored, Lefebvre argues:

Representational space ... embraces the loci of passion, of action and of lived situations, and thus immediately implies time. ... We should have to study not only the history of space, but also the history of representations, along with that of their relationships—with each other, with practice, and with ideology. History would have to take in not only the genesis of these spaces but also, and especially, their interconnections, distortions, displacements, mutual interactions, and their links with the spatial practice of the particular society or mode of production under consideration. (Lefebvre 1991, 42)

The concept of the chronotrope provides theoretical ground for reading the literature of London through the lens of the city's spatial and social history in the eighteenth century. Expanding on Jerome McGann's social theory of text (McGann 1983; McGann 1991; McGann 2001; McGann 2006; McGann 2014), the Grub Street Project as a digital social edition attempts to present the relationships of books and society in a particular period and in a particular space. This means that the project must map both historical facts and creative representations of the city. Here I take as examples the representations of the London Wall in Alexander Pope's The Dunciad Variorum (1729) and William Hogarth's twelve-plate series Industry and Idleness (1747). This is not to say that Pope wrote The Dunciad or Hogarth drew Industry and Idleness with maps in hand; rather, that studying their narratives through eighteenth-century maps reveals assumptions and views of the city they shared with their contemporaries.

Temporal-topographical readings of these two early modern narratives of London demand particular attention to the cultural configurations overlying the geography of the metropolis. Though the Cities of London and Westminster had long been joined into a single entity by the Strand and the northward expanse of streets and buildings of the "Town" (the fashionable West end), Joseph Addison in 1712 described it as an "Aggregate of various Nations":

The Courts of two Countries do not so much differ from one another as the Court and City in their peculiar ways of Life and Conversation. In short, the Inhabitants of St. James's, notwithstanding they live under the same Laws, and speak the same Language, are a distinct People from those of Cheapside; who are likewise removed from those of the Temple on the one side and those of Smithfield on the other, by several Climates and Degrees in their way of Thinking and Conversing together. (Addistion 1712, 403)

While not all such distinctions are made visible in a map such as John Strype's 1720 representation of the city, the signs of the division between City, Court, and Town are discernible in the Temple Bar marking the outer boundary of the City, in the vast park spaces and large gardens and squares in Westminster compared to denser concentrations of buildings in the east, and in the clear barrier of the City Wall dividing that centre from the rest of the city as a whole (Figure 5). The old Roman defences of the city, though still a physical and psychological presence until mid-eighteenth century, receive surprisingly little attention in histories or studies of literature and art of the early modern period. No longer a fortress, the Wall around the City of London encircled a powerful commercial and financial enclave within the rapidly expanding limits of the city as a whole. Here was to be found Guildhall, the centre of civic government; the Royal Exchange; the new Custom House; the new East India House; the new South Sea House; the Excise Office; the large Stocks Market; and the Bank of England. In Strype's 1720 map, the frequency of three-dimensional buildings concentrated in the City of London and Westminster contrast tellingly with the two-dimensional streets and flat shaded areas representing blocks of buildings in the areas to the north and east outside the Wall. The Wall and the power of the enclave within assumes great significance in the narratives of the City by Hogarth and Pope, and its representation in maps serves to illuminate its meaning in their chronotopes of progress through the City. I now turn to an exercise in static mapping of Hogarth's twelve-plate series Industry and Idleness before concluding with an examination of the The Dunciad to suggest how such chronotopes might be visualized using digital mapping and dynamically generated content.

Figure 5: A clearly demarcated City Wall, from A New Plan of the City of London, Westminster, and Southwark, in A Survey of the Cities of London and Westminster by John Strype (1720). Courtesy of Special Collections, University of Saskatchewan Library.

In 1747 when Industry and Idleness was published, the Wall was undergoing a gradual process of disintegration. An anonymous "Citizen" five years previously had described a barrier still largely intact with one breach caused by collapse from houses being built too far into the wall, and a few others resulting from general decay or deliberate construction of passages through houses or rooms, bulwarks, doors, and openings. "Little Mooregate," for example, had been "lately, (A.D. 1741.) by Order of the City taken down, a Bridge build over the Channel, and the Foot-Way greatly enlarged and posted, and rendered most convenient for Foot Passengers," while next to "Moorgate to the West, the Wall is broke down for the Convenience of the adjoining Coffee House" (Citizen 1742, 41). The demolition of the one-time fortress's perimeter was a development reflected in Rocque's map based on surveys done from 1741–5 (Figure 6), where the Wall has all but disappeared. In 1752, one J.M. wrote that the City Wall "is now mostly decay'd, Dwelling Houses being now built upon the Foundations of that which is wanting, or upon the Top of that little which remains" (J.M. 1752, 200-201), with the gates alone remaining in good repair; a year later David Henry also commented that "From Ludgate there is little of the Old Wall remaining" (Henry 1753, 54) Hogarth's series shows how, though the Wall itself had crumbled away to barely visible traces in the lines of remaining streets and building foundations, the space that had once been circumscribed by the great barricade remains as a virtual citadel in narratives of the City's commerce and industry.

Figure 6: The City Wall now largely invisible, in John Rocque's A Plan of the Cities of London and Westminster, and Borough of Southwark London: John Pine and John Tinney, 1746. Courtesy of the Beinecke Rare Book and Manuscript Library, Yale University.

Hogarth's series, which depicts the industrious apprentice Francis Goodchild marrying his master's daughter, becoming a partner in his master's business, and eventually becoming Sheriff and finally Mayor of London—in contrast to the idle apprentice Tom Idle who forfeits his indentures, is sent out to sea, becomes a thief and possibly a murderer, and eventually ends up on the gallows—seems to provide a simple moral tale. Ronald Paulson suggests a more complicated irony: "If you are a poor ugly boy who lacks the favor of his master, the ability to please him, the energy of industry, and you have a dark spot for your loom, then you are doomed never to rise—indeed to descend lower" (Paulson 1992, 311). Paulson argues that "Goodchild's world of enclosed architectural spaces is plainly delimited as the City of London" and that the values Hogarth represents, perhaps ironically, are "Protestant and economic," potentially allowing readers to detect the series of ironies that subvert an overly simplistic reading of the images.

Examining the mappable points of Industry and Idleness on John Rocque's 1746 Plan of the Cities of London and Westminster corroborates such a reading. Hogarth's chronotope of progress and prosperity depicts the lives of his characters as journeys comprised of particular stopping points through and around the City of London. The clear pattern of Francis Goodchild's centripetal gravitation toward wealth and prestige within the City Wall, explicitly contrasting the centrifugal passage of Tom Idle propelled to corporeal punishment and death far beyond the bounds of the greater city (Figure 7), shows a subtly ironic view of prosperity and industry in eighteenth-century London. The first plate shows the two apprentices and their master, a weaver, in his workshop in Spitalfields. The area was associated with the Huguenot silk-weaving trade, renowned as a center of production for fine quality silks. Silk merchants, master weavers, and related business owners and retailers in the industry were prosperous, and Spital Yard was one of the best and most affluent addresses. Other areas in Spitalfields, however, were overcrowded and largely impoverished. Riots and other criminal acts contributed to a long history of disputes between apprentices, journeymen, and the prosperous master weavers and other successful members of the trade. Tom Idle, asleep at his loom, has a beer tankard close at hand and a ballad of Moll Flanders, based on Defoe's The fortunes and misfortunes of the famous Moll Flanders, tacked to the beam behind him. The tankard may suggest either immoral drunkenness or a means of escape. The ballad might remind the plate's viewers that it is possible to overcome the barriers to success that are an accident of one's birth (Moll was born in Newgate, lived as a "whore" for twelve years, and a thief for twelve years, was a transported felon in Virginia for eight years, and "at last grew rich, liv'd honest, and died a penitent"). Another interpretation for some viewers might arise from a central passage in Moll Flanders, which suggests that money and not exemplary character is key to societal approval and belonging: "and if a young Woman have Beauty, Birth, Breeding, Wit, Sense, Manners, Modesty, and all these to an Extream; yet if she have not Money, she's no Body, she had as good want them all" (Defoe 1722, 18).

Figure 7: William Hogarth's chronotope of prosperity, showing the movements of Tom Idle overlaid on An Exact Survey of the City of London Westminister the Borough of Southwark and the Country Near Ten Miles Round Begun in 1741 and ended in 1745 (1746) by John Rocque. The lines of the City Wall shown here do not appear on the original map. 1. Plate 1: "The Fellow 'Prentices at their Looms" in Spitalfields; 2. Plate 3: "The Idle 'Prentice at Play in the Church Yard During Divine Service," likely at St. Paul's Shadwell; 3. Plate 5: "The Idle 'Prentice Turn'd Away, and Sent to Sea," at the Isle of Dogs near the pirate's gibbet; 4. Plate 9: "The Idle 'Prentice Betray'd by his Whore & Taken in a Night Cellar with his Accomplice," at the Blood Bowl House in Hanging Sword Alley; 5. Plate 10: "The Industrious 'Prentice Alderman of London, the Idle One Brought Before Him & Impeach'd by his Accomplice," in Guildhall; 6a. Newgate prison; 6b. St. Sepulcher: on the day of their hanging the condemned prisoners of Newgate were placed in a horse-drawn cart, often along with their own coffins, for their journey; the bells of St. Sepulcher tolled on execution days, and outside the church a nosegay was presented to each condemned criminal; 6c. Snow Hill and Holborn Bridge: the procession then went down Snow Hill, and across Fleet Ditch via Holborn Bridge; 6d: Holborn Street: from thence they proceeded along Holborn Street; 6e. past St. Giles, where prisoners had historically stopped to drink their last bowl of ale; 6f. along Tyburn Road; 6. Plate 11: "The Idle 'Prentice Executed at Tyburn."

From the first plate, the paths of the two apprentices diverge, with Goodchild attending service at St. Martin in the Fields in the fashionable west end of London in plate 2 and Idle playing hustle-cap on an altar-tomb in a churchyard cemetery during the service in plate 3. The church has not been identified, but seems likely to be St. Paul's Shadwell far to the east (Stephens 1877; see also Paulson 1956). Notably, Shadwell Parish was impoverished, the home of many unskilled immigrant workers, and many of its cheaply constructed houses had degenerated into slums. From this point, whether Hogarth was deliberately commenting on the cycles of poverty or solemnly providing a moral tale, we see Goodchild move within the City Walls, where he remains, ascending to positions of greater prosperity and increasing power. Idle, meanwhile, finds himself propelled outward, far past the city. In plate 5, "The Idle 'Prentice Turn'd away, and Sent to Sea," he is rowed past the Isle of Dogs toward a moored ship. He has forfeited his indentures, and his papers float away in the water; the waterman points to a pirate's gibbet on the shore, displaying the body of a hanged pirate. These repeated images of corporal punishment characterize Idle's journey: the man behind him taunts him with a rope, suggestive of the cat-o'nine-tails that will be the punishment for miscreants on the ship, and foreshadowing Idle's final journey out of London to Tyburn to be hanged. Rocque's map shows these two emblems of punishment or "justice" for those condemned to steal for survival at the western and eastern extremities of his map (1746). Tom Idle appears inside the City Walls just once in Hogarth's narrative, in the Justice Room at the end of the Matted Gallery in Guildhall, where he is tried, found guilty, and sentenced to death by Francis Goodchild, now an alderman taking his turn sitting as magistrate. The former borders of the City Wall for Hogarth and his contemporaries, then, might be seen as a tangible psychological and social barrier to power and prosperity within the larger city, however pragmatic and unproblematic the ethic of "industry" might seem.

Almost two decades earlier when The Dunciad Variorum was published, the Wall was still a physical barrier as well as a social and political one, and it must have seemed especially so for Alexander Pope born at Plough Court just off Lombard Street in the City, whose family had been forced out shortly after he was born by a new statute barring Catholics from living anywhere within ten miles of London and Westminster. The poem begins with the Goddess Dulness, representing the decline of taste and literary standards in the metropolis, at the "Cave of Poverty and Poetry" in or near Rag Fair close to the Tower of London, a place dearer to her than "her own Guild-hall" (Pope 1729b). Rag Fair was indeed a space of poverty, held in Rosemary Lane just outside of the City and its Liberties; the eighteenth-century social commentator Ned Ward laughingly suggested the fair attracted "Ill-favoured Maukins" who seemed to be suffering from the "terrible Persecution of Rags, Lice and Poverty" picking through the old rags, "as so many Cocks upon a Pile of Horse-dung, scraping about the Filth to find out an Oat worth picking" (Ward 1703). Here outside the City are born the works of Dulness' sons, "all the Grubstreet race" of dunces, literary hacks, writers for hire, and pompous dull fools (Pope 1729b). These are the "Smithfield muses," whose low forms of entertainment – "formerly agreeable only to the taste of the Rabble" – had migrated from Bartholomew Fair to the "Theatres of Covent-Garden, Lincolns-Inn-Fields, and the Hay-Market, to be the reigning Pleasures of the Court and Town" (Pope 1729b). Dulness is inherently associated with the mercantile City, however, as the death of the City poet Elkanah Settle, whose office it had been to "compose yearly panegyricks upon the Lord Mayors, and verses to be spoken in the Pageants: but that part of the shows being frugally at length abolished, the employment of City Poet ceas'd; so that upon Settle's demise, there was no successor to that place," thus resulting in the present "Crisis of the Kingdom of Dulness" (Pope 1729b).

After Dulness crowns Tibbald (the pedantic scholar, critic, and playwright Lewis Theobald) the new King of Dunces, she and her sons (poets, critics, and booksellers) hold heroic games to celebrate, in an elaborate progress toward the City within the Wall. They begin just outside the City without at St. Mary le Strand Church, and travel eastward through Faringdon Without, which forms the southwestern extremity of the City. Along Fleet Street the dunces hold a race (wherein Edmund Curll slips in a puddle of urine close by his own shop), hold a pissing contest, move on past Bridewell, the women's prison, and hold a diving contest in the sewage of Fleet Ditch; from here they finally arrive at the City and proceed through Ludgate, but they are barely inside when they lull themselves to sleep with their own great works (Figure 8). For more information on the mapping of the games see Muri (2011) and Muri (2015). Inside the Wall is Stationers Hall, legal centre of the book trade, and the long-established locale of the estimable book trade in the areas around Stationers Hall, Paternoster Row, and St. Paul's Cathedral. The Stationers' Company had controlled the profitable business of printing and had assisted with pre-publication licensing and censorship until the abolition of the Star Chamber and Court of High Commission in 1641, followed by the lapse in the Licensing Act 1679-85 and, finally, Parliament's failure to renew the Act in 1695. The dunces represent the negative aspects of the new order of press freedom instantiated by the lapse of the Act: they are the greedy booksellers and the new breed of "Hackney-Scriblers": "a Set of Fellows, who make a Trade of Defamation and Scandal, who infamously hire themselves out for Bread, to pelt with Filth and Ordure all Persons of distinguish'd Rank and Eminence … an abandon'd Crew of Scribblers, Poets, Pamphleteers, and Journal-Writers" who, "from their Garrets and lurking Holes, or the Corners of Lanes and Allies, pelt even Persons of the highest Rank with all the Dirt and Excrement that they can rake out of their Dunghills, or sweep out of their Kennels" (Daily Gazetteer 1739). While the writers were associated with Grub Street, booksellers specializing in particular genres located their businesses in districts determined to some extent by the traffic of their audiences: the works produced by Pope's dunces were not the estimable "Antient Books in all Languages" sold in Little Britain and Paternoster Row; nor were they the "Divinity and Classicks" sold on the north side of St. Paul's Cathedral (Macky 1714; see also Raven 2007, especially chapter 6). They were more likely to be writing the plays that were to be found in Fleet Street near Temple Bar (Macky 1714), or novels, magazines, and fashionable titles to be found in Covent Garden, Fleet Street, the Exchange in the Strand (Raven 2007), and of course, the newspapers found all over the city – that is, all the new forms invented in a new commercial society that catered to modern public taste, with little regard for classical or religious tradition. Fleet Street, where the dunces hold their games had, of course, long been home to respectable booksellers and stationers, but by the eighteenth century, it was also known for the rapid change and growth of the print trade and for its piratical practices (Raven 2007). The dunces' "Guild Hall" was no venerable building inside the City Walls.

Figure 8: The heroic games of the Dunces in Alexander Pope's Dunciad Variorum (1729).

Does the topography of dunces have meaning beyond what Pat Rogers so skillfully explained four decades ago in his ground-breaking study Grub Street: Studies in a Subculture? (Rogers 1972). I have observed elsewhere that a combination of statistical mapping and visualizing the eighteenth-century city illuminates Pope's self-representation in the topography of dullness from which he, adopting a voice of omniscience from far above the excremental mire where he positioned his dunces, excludes himself (Muri 2011).Visualizing the dunces' progress through the city conveys one aspect of Pope's chronotope (Figure 9), but another layer of meaning can be added by searching a database of booksellers' and printers' addresses I have compiled from ESTC records. In this process, I limited the search to the years of the decade up to the year before The Dunciad was first printed in 1728. I then displayed the streets and locales where the works of Pope and his "dunces" were printed or sold as points on a map. As in a bubble chart, the area of each circle corresponds to a numeric value (here, the number of books printed or sold in a given location). The results suggest that Pope's own market was largely located in the area around Temple Bar (Figure 9). This dataset (books printed in London 1720–1727) is courtesy of the Center for Bibliographic Studies & Research, University of California at Riverside. Note that the data I present here provides a very good general representation of London's print culture, but nevertheless the precise figures presented here must be considered somewhat imperfect and provisional. The ESTC provides a description of the ideal copy for each edition; the visualizations do not provide any indication of sales of copies; additionally, the imprint information often does not provide a location for a printer or bookseller, so if one record in a particular year names a bookseller's location, and another does not, or if a location is named over a span of years but is missing in a record within that timeframe, I apply that location to the record(s) where there is none named; finally, I include printer locations in these maps although printers may not have sold the works. The trade for the works of Daniel Defoe, on the other hand, was largely located within the City Wall but also appeared in significant numbers in Town and Court (Figure 10). Trade in the works of Eliza Haywood, notably, most resembles Pope's characterization of the taste of the rabble, with most of her business with printers and sellers located in the area of Temple Bar and Temple Gates, and to the east in Covent Garden and Pall Mall (Figure 11). The trade in Edward Ward' works least resembles Pope's imagined "Respublica Grubstreetaria," as Jonathan Swift called it: one of the most famous, and most successful, of "Grubstreet" hacks, Ward sold his works to an audience mainly inside the wealthy City (Figure 12). Statistical mapping demonstrates that Pope's self-construction in The Dunciad as somehow above or apart from the dunces competing for book sales in the Strand and Fleet Street is clearly rather hypocritical, situated as he was along with everyone else in the crass market of the popular book trade expanding out of the City along with other commercial ventures encroaching upon the Strand. To see the book trade of Pope and the dunces in this light demonstrates how Pope was firmly positioned in the market he so despised, in the midst of the filthy, money-grubbing market of Fleet Street; indeed, along with a few of his dunces clustered around Temple Bar or Gates, his trade was situated at the egress between the City without the wall and the Court, and far from Ludgate and the powerful enclave of the City within.

Figure 9: Printers and booksellers for Alexander Pope, 1720–1727.

Figure 10: Printers and booksellers for Daniel Defoe, 1720–1727.

Figure 11: Printers and booksellers for Eliza Haywood, 1720–1727.

Figure 12: Printers and booksellers for Edward Ward, 1720–1727.

Conclusion

The study of history, space, time, and narrative through digital mapping techniques demands a layering of both qualitative and quantitative information and a recognition that digital collections are not only datasets to be mined for quantitative analysis and display, but also demand traditional close readings of the texts that digitization has made accessible in abundance. The visualizations presented here certainly make no claim to certainty, whether in terms of geographic positioning, or in terms of the number of any particular author's imprints sold at the locations indicated. The unreliability of imprint information as the sole indicator of where all books were printed and sold is well known (see Muri 2011 and see Raven 2007 on the complexity of sorting out these issues in his mapping of more precise addresses for booksellers in particular regions of London), but nevertheless, in terms of relative ratios, we can get a sense of where Pope's imprints were available compared to other writers in the years leading up to the publication of The Dunciad. In terms of the position of the London Wall and its relationship to a community of wealth and power, we can see the portrayal of the City's insiders or outsiders by both Pope and Hogarth, and we can gain insights into their idiosyncratic and complex social critiques. Mapping these creative works alongside other historical information is just a beginning to a larger enterprise where eighteenth-century books, images, and maps create a better understanding of the city and its texts as a social-spatial network.