12

Spatial Analysis and Digital Urban Animal History 1

Sean Kheraj

As several of the chapters in this volume illustrate, one of the primary challenges of historical scholarship on animals is making animals visible in historical records. People produce all traditional sources used in historical scholarship.2 Of necessity, therefore, historians must find animals in the past through the eyes of people who wrote and thought about animals in the past. This is what geographer Chris Philo refers to as “the distorting lenses of historical documents written by humans.”3 Of course, these lenses have significant limits that constrain knowledge about animal histories. Our challenge is to illuminate where animals appear in inherently anthropocentric sources and find traces of animals in the past. Animal history applies an “animal lens” to the past to help reveal new insights and demonstrate the ways in which animals shape history.4 To do so, however, historians first require methodologies to see animals in the past.5

Spatial analysis using Geographic Information Systems (GIS) software can be a powerful illuminant for animal history. It can bring animals into the foreground from sources that obscure their presence and influence. This new approach draws on the so-called spatial turn, which in recent years has influenced several subfields of historical scholarship. GIS allows scholars to translate textual spatial data into visual spatial data in ways that can reveal previously undetected and even unsuspected geographic relationships among people, non-human animals, and the ecosystems they share that are not obvious in the original sources. In doing so, GIS operates as more than just a visualization for presentation; it also becomes a tool of analysis.6

This chapter provides examples of the use of GIS in the analysis of the spatial relationships between human and non-human animals in nineteenth-century North American urban environments, just as Andrew Robichaud shows in his case study of San Francisco in Chapter 13. It argues that new digital tools and digitized historical resources now available to historians make it possible to yield significant historical insights into human-animal spatial relationships in urban environments through experimentation and simple play with digitized sources. All of this is now possible without the need to rely upon expensive multi-member interdisciplinary teams of scholars and technicians. Individual historians can generate meaningful visualizations and engage in spatial analysis with Web-based GIS tools and digitized historical records. This is possible due to a combination of factors that include the expansion of Web GIS applications, large-scale digitization projects, and crowd-sourced approaches to sharing data.7

To illustrate some of the ways in which historians can mobilize GIS and digitized historical sources to enhance the “animal lens” on the past, this chapter explores two case studies. The first case study examines the transformation of butchering regulations in the City of Toronto in the nineteenth century to understand the spatial relationships between urban development and access to butchered meats. The second case study expands the geographic lens on urban animal history to consider the Great Epizootic of 1872–73 and the interconnections among cities and the horses that powered urban transportation systems in nineteenth-century North America. In both case studies, I use ArcGIS Online as a platform for testing hypotheses about historical spatial relations between people and non-human animals in urban environments. This Web-based GIS tool by the spatial software developer Esri offers solutions for integrating digitized historical sources into maps for spatial analysis. It also supports crowd-sourced sharing of spatial data, which allows for iterative knowledge generation as historians build upon the work of other researchers and GIS specialists.

Butchering and Selling Meat in Nineteenth-Century Toronto

Before 1851, the City of Toronto prohibited the slaughter of live animals and the sale of butchered meats anywhere in the city outside of public market buildings. St. Lawrence Market, the city’s first public market, was once the only place in the city where one could legally purchase fresh produce, including live animals and butchered meats (see Figure 12.1). Signed on 27 May 1834 by Toronto’s first mayor (and future republican insurrectionary), the infamous William Lyon Mackenzie, By-Law 2 first established the rules and regulations that governed the city’s public markets. Across fifteen different sections, this lengthy early by-law laid out the rules for the sale of meat and other produce in the city, limiting the sale of meats to licensed butchers.8

In a manner that was common in other North American municipalities, Toronto’s public market by-law gave the city the power to regulate the sale of meat in order to achieve several goals.9 First, public market by-laws provided a means to ensure the quality of food products. Market clerks and inspectors could check the products sold in the public market to verify that they had not been adulterated. This provided consumers with a degree of protection against unscrupulous sellers who might mix sawdust into a bag of flour or add weight to a cut of meat in order to raise the price. Second, this by-law offered a measure of protection for public health. The city required licensed butchers to furnish their stalls “with a plentiful supply of good meats” and prohibited the sale of “any unwholesome, stale, emaciated, blown, stuffed, tainted, putrid or measly pork, meat, poultry, or other provision.” Third, public market by-laws established licensing fees that provided revenue to the city. Butcher’s licenses were valuable, coveted commodities.10

Restrictions on the sale of live animals and butchered meats (along with other produce) created spatial relationships within the city that reveal some of the close connections between people and domestic livestock animals in urban environments. In an age before artificial refrigeration and railway deliveries, the urban food supply moved by foot and by hoof. Limiting the sale of butchered meats to public markets thus created geographies of access to food that influenced the development of the city. This chapter shows how GIS can reveal those relationships by mapping the spatial distribution of food in nineteenth-century Toronto, allowing us to see the ways in which animals shaped the city.

For instance, in 1858, there were three public markets in the City of Toronto: St. Lawrence, St. Patrick’s, and St. Andrew’s Markets. Using circle vector layers with one-kilometre radii laid atop a georeferenced copy of W.S. Boulton’s 1858 Atlas of the City of Toronto and Vicinity, it is possible see how much of the built-up portion of the city was within a one-kilometre walk from a public market (Figure 12.2). Georeferencing (or GIS rectification) is a process that reshapes a digital image and maps it to real-world coordinates. In this case, a user on ArcGIS Online has taken a digital scan of the 1858 atlas and altered its dimensions in GIS software to correspond with real-world coordinates of the streets and landmarks of the City of Toronto. In doing so, the one-kilometre radii vectors show which parts of the built-up city were within a relatively short walking distance from each public market. The most densely built-up neighbourhoods fall within a one-kilometre radius of at least one public market and imply a correlation between urban development and access to food. The map reveals that residents who lived in the sparsely settled areas north of College Avenue and west of Yonge Street lived furthest from the three public markets. The 1858 Boulton map shows this part of the city to have much less development.

The process of generating such a map even just a few years ago was considerably more labour intensive. A combination of improved Web-based GIS software, historical digitization projects, and data-sharing now make it feasible for solo researchers to compile such spatial visualizations in relatively short order. The underlying historical map was originally compiled and surveyed in 1858 by William Somerville Boulton and Henry Carew Boulton. This was the first large-scale map to depict the built environment of the City of Toronto.11 Toronto Public Library scanned a high-resolution PDF copy of the thirty-sheet atlas, which is available to the public in its online catalogue.12 In November 2017, an ArcGIS Online user georeferenced the image files of this atlas using ArcGIS desktop software and uploaded the georeferenced images as a sharable layer in ArcGIS Online, ESRI’s Web-based version of its GIS platform. The three vectors with one-kilometre radii shown in Figure 12.2 were measured using the measurement tool in ArcGIS Online and laid atop the georeferenced map, which was inserted from the searchable database of shared layers. What would have once required a collaborative team of historians and GIS technicians to produce is now readily accessible to individual historical researchers to generate such maps for experimentation with spatial representations of different forms of historical data. In this example, the result shows what parts of the built-up portions of the city were within walking distance to access live and butchered animals for food. The results help to support the hypothesis that there was a relationship between urban development and pedestrian access to meats.

Online GIS tools can further be used to explore how those spatial relationships between Toronto residents and access to both live and butchered animals changed over the nineteenth century. In 1851, the City of Toronto liberalized its public market regulations to permit the slaughtering of animals at facilities elsewhere in the city, and in 1858, the city began licensing butcher shops outside of the public market buildings. The result of this change in policy was an explosion of butcher retailing across the city. Figure 12.3 shows the number of butcher shops listed in city directories from 1875 to 1890. These directories contain spatial data about the distribution of butcher shops in the form of street addresses, but those data are difficult to interpret unless they can be mapped. Once again, ArcGIS Online’s repository of shared layers provides individual historical researchers simple tools to begin to analyze spatial data to determine how butcher shops were distributed across Toronto in the past.

Figure 12.4 shows the locations of all butcher shops listed in the 1890 city directory. Once again, Toronto Public Library has digitized all the city directories in its collection for the period from 1833 to 1969, making this freely available on its website.13 Each place marker is mapped to the address shown on a georeferenced copy of an 1889 fire insurance atlas of Toronto. This georeferenced resource was similarly uploaded by a user and deposited in the shared repository.14 In this instance, geographers from the Department of Geography at University of Toronto scanned and georeferenced a copy of the original atlas from Toronto Public Library as part of the Georia Project, a joint initiative of l’Université Laval and University of Toronto. The Georia Project’s long-term goal is to coordinate georeferenced databases of environmental, social, and health-related data in Canada “into on-line geographical information systems (GIS) that can be used by researchers and educators alike.”15 The process of plotting each place marker from the addresses listed in the 1890 directory is as easy as dropping a pin on Google Maps. The user can annotate each place marker too, which allows for additional historical context and referencing of source material. In this example, the interactive version of this map allows users to click on each point to see the name of the shop owner and the precise street address, as it appeared in the directory.16

As a tool of analysis, this spatial visualization reveals insights into the shifting landscape of animals, food, and transportation in nineteenth-century Toronto. When the city liberalized its regulations concerning butcher shops, the geography of access to butchered meats changed. Butcher shops appear to have spread rapidly in the period between 1875 and 1890 throughout much of the city, beyond the more limited setting of the three public markets. GIS allows historians to test theories about such spatial relationships. Figure 12.5 adds street railway routes to the map to test the hypothesis that butcher shop locations and transit were related. Visual analysis suggests that, as with other retailers, butchers tended to be located along the city’s main street railway routes, particularly along Yonge Street and Queen Street, two of the earliest developed routes.

In summary, the GIS visualizations shown in Figures 12.2–12.5 partially illuminate one form of historical human-animal relation. I was able to generate each map as a sole researcher to test hypotheses and analyze textual historical data in visual form. I am also able to share my maps and add my layers to the shared database for other researchers to use. Web-based GIS applications provide tools that remove the need for more complicated and resource-intensive computing software and hardware. The integration of shared resources in the ArcGIS Online platform allows users to leverage georeferenced digitized maps from other historical digitization projects. In doing so, the platform opens new possibilities for a wider range of historical researchers interested in exploring other aspects of the spatial relationships between people and non-human animals.

Reconstructing an Epizootic

The examples above show how historians of urban animals can use GIS to illuminate human-animal relations from historical documents at the scale of one city. The same tools can also highlight human-animal relations at the broader scale of urban networks. As the growing body of global scholarship in urban animal history has well established, industrializing cities of the nineteenth century were multi-species environments shaped, at least in part, by the exploitation of livestock animals for food and labour.17 As Clemens Wischermann and Philip Howell claim, “the history of cities should now be unthinkable from the perspective of humans alone.”18 By the 1870s, these animals (and the supplies needed for feeding and sheltering urban livestock) also moved among cities in urban networks connected by railways. That movement of livestock and supplies linked the urban ecosystems of cities across North America, from Montreal to New York City to San Francisco.

No historical case study shows this better than the Great Epizootic of 1872–73.19 In October 1872, an unknown illness struck the urban horse population of Toronto. According to firsthand reports, nearly every horse in the city appeared to be affected. Many believed the disease to be a virulent form of equine influenza. Within weeks, the disease spread to all major cities in the northeastern United States and Canada. By late winter 1873, horses in nearly every city in Canada and the United States had been affected by the epizootic. The Great Epizootic brought cities to a standstill as the sickness incapacitated hundreds of thousands of horses. Even though most horses survived, the temporary suspension of horse labour halted intra-urban transportation, including street railway services and the delivery of goods. The epizootic stranded commuters and left goods piled up at wharves and railway stations. It was made starkly clear how crucial horses were to the functioning of industrial cities in nineteenth-century North America.

But the Great Epizootic also revealed that the ecologies of North American cities were interconnected via animals and railways. Web-based GIS tools, shareable layers and GIS datasets, and large-scale digitization of historical newspapers make it feasible to visualize and analyze these spatial relationships among cities, horses, and railways. Figure 12.6 shows a map of every city in which horses were incapacitated by the Great Epizootic and the railway networks that connected Canadian and US cities. An animated version of this map shows the movement of the epizootic as it spread outward from Toronto (Figure 12.6).20

To create such a map, I analyzed an enormous digital archive of historical newspapers from several different collections. In Canada, there are many digital newspaper archive collections. However, the record of digitized historical Canadian newspapers is incomplete, fragmented, and difficult to access.21 Canada lags behind the US, UK, Australia, and New Zealand in the digitization of historical newspapers.22 There are few national newspaper digitization initiatives. Instead, most digitization projects focus on single newspapers or small collections of regional newspapers. Nearly all digital historical newspaper collections of Canadian newspapers are only available via paid subscriptions. Yet there are some significant collections that are freely available from public archives, libraries, and universities.23 Digitized historical newspapers from the United States are more readily available. Substantial national collections from Library of Congress and private genealogy research companies cover newspapers from most regions of the US.24 The newspaper record provides some of the best evidence to recreate the path of the Great Epizootic because they document when the symptoms of the disease became widely observable in the public. Each place marker on the animated map includes information about the approximate date that reports of widespread illness among horses appeared in newspapers from each city. Altogether, the map includes data from more than 480 newspaper reports and shows the arrival of the Great Epizootic in 164 cities between October 1872 and September 1873.

Once again, ArcGIS Online enabled the creation of this map and animated visualization. The simple place-marker tool (used to mark the locations of butcher shops in Figures 12.4 and 12.5) was used to annotate a base map of North American cities to indicate which cities had been affected by the epizootic and when the disease was first publicly observed. ArcGIS Online allows users to add a date-time field to make a historical GIS layer that is “time-enabled.”25 Such a GIS layer can then be added to a Web application template called “Time Aware” that animates GIS datasets by using the date-time field and a time slider.26 For each city affected by the epizootic, the date-time field indicates the approximate date that reports first documented observations of symptoms among horses in the city. Both the place-marker tool and the “Time Aware” Web application template are relatively simple to use and do not require large-scale research teams.

To compare the movement of the epizootic shown in the time-enabled place marker layer of affected cities with the railway system, I added shared layers of historical GIS datasets of North American railways to the map. An ArcGIS user added a layer representing the railway network in the US in 1870 to the shareable database of layers in ArcGIS online. The dataset for this layer was created by a digital history project based at University of Nebraska, Lincoln, led by William G. Thomas III, Richard Healy, and Ian Cottingham called “Railroads and the Making of Modern America.” This multi-authored digital history project “aims to collect and make available a wide array of materials to shed light on the ways Americans experienced the railroads in the nineteenth century.”27 By making their dataset open-source, it was available to be integrated easily into the time-enabled map of the movement of the Great Epizootic. A second layer featuring Canadian railways was created and shared by ESRI Canada Ltd. based on data derived from several historical atlases of Canada.28 The addition of these two shared layers allows for comparative spatial analysis of the relationship between the movement of the epizootic and the North American railway system in the 1870s. Both historical railway GIS datasets were the products of multi-member digital history projects intended to share data to facilitate future research (much like Georia). The result is that these open, shared GIS datasets combined with Web-based applications make it possible for solo historical researchers to test theories and observe spatial relationships among people, non-human animals, and cities even on a continental scale.

Creating a time-enabled map of the movement of the Great Epizootic in a historical GIS tool like ArcGIS Online illustrates yet again how GIS can illuminate human-animal relations found in historical documents. Like the injection of a fluorescent dye into the bloodstream to highlight blood vessels for medical analysis, GIS highlights spatial and temporal data, making human-animal relations visible in ways that could not be accomplished solely by textual analysis of historical documents. The hundreds of newspaper reports that documented the arrival of the epizootic in Canadian and US cities in 1872–73 contain data about movement and time that cannot be seen unless it is mapped and animated. In doing so, the GIS visualization demonstrates how animal disease flowed through the urban network in a single common ecosystem. Horses working on the streets of Toronto could get sick and eventually infect horses on the streets of Oakland.

The application of GIS as a tool for studying the Great Epizootic not only adds broader scale to the “animal lens” for historians, but the use of time-enabled animation offers a way of observing change over time, fundamental to all historical analysis. The results produced unexpected findings that changed prior understandings of how this disease outbreak moved throughout North America. For instance, at the outset, I had expected to find out where and when the disease crossed the international border from Canada to the US. However, the evidence in the animation revealed that this was not a single event in a single place. Instead, the disease crossed back and forth across the border at least five times passing from Canada to the US and back into Canada at different points along the enormous border between these two countries.

Limits of GIS for Spatial Analysis in Urban Animal History

The examples above show exciting potential for the use of GIS in visualizing and analyzing historical spatial relationships among people and non-human animals in cities. The use of such powerful mapping tools, however, comes with certain caveats and cautions. GIS can be used as an effective tool of analysis for understanding spatial relationships and the shared datasets available to researchers make the use of GIS much easier than ever before. Interpreting spatial meaning from historical texts using GIS still requires skill, patience, and a critical eye. It also requires basic digital literacy and understanding of the processes (and limitations) of digitization of historical records.

Fundamental critical analysis skills are necessary for the evaluation of any shared datasets and GIS layers. When working with a repository of datasets like that available in ArcGIS Online, researchers must carefully choose sources from reliable and credible authors (as they would with the selection of any secondary research materials). Because the ArcGIS repository is open for any user to upload data, verifying the quality of the data is vital. In the examples above, I have tried to use datasets from credible academic research teams and GIS specialists. Basic research skills for evaluating quality and authenticity of sources then can allow animal history researchers to leverage the advantages of crowdsourced data sharing.

Transforming historical documents into digital objects and GIS visualizations also raises questions about how far researchers should go when manipulating data for analytical purposes. Robert Sweeny warns, “the act of warping abstracts the map from its original context. GIS rectification is one of the many ways that historical sources are made to be compatible with computers. It is a process . . . that facilitates ahistorical thinking; it can also allow us to see what might not otherwise be visible.” Sweeny argues that GIS can remove documents from their historical logic. By this he means both the specific conditions under which a source was produced and the circumstances by which the source survived and was preserved into the present. Moreover, in the case of animal history, the logic of the sources is inherently anthropocentric. GIS can illuminate animals in the past, as I have argued here, but it may also obscure the human-centric vision of the sources themselves and the archival conditions under which they were kept as they are transformed from text to digital visualization. In the example in Figure 12.4, the map includes digitized and georeferenced copies of the plates of an 1890 fire insurance atlas for the City of Toronto. The place markers that sit atop that layer are a georeferenced representation of data from a digitized copy of a city directory. These documents were not originally meant to be used in this manner. Digitization and GIS make them legible to a computer and open possibilities for observing one form of human-animal spatial relationship. But this process also decontextualizes the fire insurance atlas and city directory from their own, independent historical logics. Fire insurance atlases were tools of trade for documenting and tracking insurance liability. City directories were commercial products and vehicles for advertising. Neither source was originally intended to reveal insights into how people and non-human animals shared space in the city. Choosing such sources and removing them from their original contexts might obscure omissions and silences in the sources. City directories capture geospatial data about animal trades in urban environments (dairies, butchers, tanners, livery stables, etc.), but they do not capture aspects of the place of animals in the informal economy of cities. Butcher shops were one place that residents of Toronto accessed fresh meat, but some urban dwellers still kept livestock in the city in the 1890s and slaughtered animals in their homes. Fresh meat could still be traded among neighbours in ways that are not captured in sources meant to represent the formal economy. And yet when cross-referenced using GIS, historians can, as Sweeny notes, see what might not otherwise be seen. Given the ongoing challenge of finding non-human animals in inherently anthropocentric sources, this kind of digital manipulation may be worthwhile, but it should be used carefully with a complete understanding of its limits and potential for misrepresenting the past.

In order to leverage new digital mapping tools and understand their limits, historians must possess at least some basic digital literacy skills. As Ian Milligan and others have warned, the revolutionary shift to digital information and online dissemination has meant that historians must acquire and teach new digital skills. “This does not mean abandoning traditional research methods,” Milligan contends, because “historians will long continue to be masters of close reading and parsers of nuance and context—but it does mean that new skills to better contextualize and understand digital material are needed.”29 The resources used in the maps above do not consist of the impossibly large scale of the Web archives Milligan sees as a future challenge for historical researchers, but they do include the ever-growing archive of digitized historical documents available on the Web. Historians need some knowledge and understanding of the digital skills necessary to make use of that archive, including GIS skills.

While it is true that digital history scholarship often demands interdisciplinary collaboration, historians must also acquire the capacity to engage directly with digital tools in order to yield the full analytical benefits of a tool like GIS. Douglas Seefeldt and William G. Thomas see digital history scholarship requiring “interdisciplinary collaboration, the likes of which most historians have yet to embrace; cooperative initiatives that involve historians, programmers, information architects, designers, and publishers.”30 But historians cannot gain the same meaningful insights from GIS without direct engagement with the process of building digital maps; they cannot outsource that work entirely to technicians. Playing with digital maps, manipulating digitized documents, and conducting ad hoc experiments to test theories and hypotheses, I argue, are part of the analytical process enabled by GIS. It is akin to the ideas of William J. Turkel and Devon Elliot concerning “humanistic fabrication.” They find that, “the present conjuncture—of making as a new social movement, of easy-to-use and freely available platforms that invite modification, of detailed online instructions for doing just about anything—makes it almost costless for historians and other humanists to research, teach, learn, play, and experiment with new technologies.”31 This is also true of GIS technologies. ArcGIS Online is one of a handful of new Web-based digital mapping tools that allow for the kind of play and experimentation that Turkel and Elliot argue can be applied to material production using 3D printing technologies. As the examples above show, playing with digitized sources using GIS can help make animals in the past visible to historians in new ways.

Play, in this instance, takes the form of an experiment to test a theory or hypothesis. In the examples above, the final outcome of each spatial visualization was not predetermined. Instead, they were the result of guessing, wondering, and then visualizing. What did the landscape of butcher shops in Toronto look like after the liberalization of public market bylaws? Can I see any relationship to the street railway system if I add another layer to the map? If I place time-aware markers on a map, can I see how the Great Epizootic spread from city to city? Will it show me when and where the disease crossed the international border? Tests, corrections, and iterations drive the process of using GIS as a tool for seeing animals in the past.

Conclusion

In his often-cited essay “Why Look at Animals?,” John Berger acknowledges an immutable distance between people and non-human animals. An animal’s “lack of common language, its silence, guarantees its distance, its distinctness, its exclusion, from and of man.”32 For the historian, that distance also spans the depths of time. To know animals in the past almost seems more impossible than knowing them in the present.

The most significant limit to the use of GIS for exploring historical human-animal relations remains that silence and the anthropogenic character of the sources themselves. In translating historical documents into digital objects and spatial visualizations, it is possible to see a version of animals in the past, but the illumination GIS provides merely casts different angles of shadows on thoughts, ideas, and language that people in the past used to interpret and understand animals. These digital tools must be used with a critical eye so as not to lose sight of the underlying evidence, which consists of fragments of human thought.

Still, GIS has much to offer animal history. In bringing human-animal relations to the surface of sources that relegate animal to the margins, GIS provides historians with new ways of seeing what might not otherwise be observed. By better understanding the relationships between people and non-human animals in space, we might learn more about how those relationships changed over time.

Notes

1 The author would like to thank Jennifer Bonnell and Sandra Swart for their comments and feedback on this chapter. He is also grateful to all the participants of the “Traces of the Animal Past” conference held at York University in November 2019.

2 Beyond traditional textual sources, animal historians may also consider what Etienne Benson calls “material-semiotic traces of the past,” a similar concept to the embodied sources produced by animals themselves that Sandra Swart describes in her chapter in this volume. See Etienne Benson, “Animal Writes: Historiography, Disciplinarity, and the Animal Trace,” in Making Animal Meaning, eds. Linda Kalof and Georgina M. Montgomery (East Lansing: Michigan State University Press, 2011), 3–16.

3 Chris Philo, “Animals, Geography, and the City: Notes on Inclusions and Exclusions,” Environment and Planning D: Society and Space 13 (1995): 677.

4 Joshua Specht, “Animal History After Its Triumph: Unexpected Animals, Evolutionary Approaches, and the Animal Lens” History Compass 14, no. 7 (2016): 328.

5 It is important to note that in this chapter, GIS analysis and techniques tend to privilege the visual over other senses. Indeed, the preceding paragraph is replete with references to sight, lenses, and vision as metaphors for generating knowledge about animal history. As other chapters in this volume show, however, sight is one among other senses that scholars use to generate new knowledge about animals and history. Sandra Swart (Chapter 1), in particular, points to some promising ways in which sound, touch, and even smell could inform analysis and interpretation of animal bodies as sources.

6 Jennifer Bonnell and Marcel Fortin, “Introduction” in Historical GIS Research in Canada, ed. Jennifer Bonnell and Marcel Fortin (Calgary: University of Calgary Press, 2014), xi; Richard White, “What is Spatial History?,” The Spatial History Project (February 2010), https://web.stanford.edu/group/spatialhistory/cgi-bin/site/pub.php?id=29; Anne Kelly Knowles, “GIS and History,” in Placing History: How Maps, Spatial Data, and GIS are Changing Historical Scholarship, ed. Anne Kelly Knowles (Redlands, CA: ESRI Press, 2008).

7 The examples of the use of GIS in historical urban animal history in this chapter rely on ArcGIS online, a commercial Web GIS product developed by Esri. Free, open source solutions are available from other developers, such as Geographic Resources Analysis Support System (GRASS), QGIS, System for Automated Geoscientific Analyses (SAGA) GIS, and others.

8 City of Toronto Archives (hereafter CTA). By-Law 2, “An act to regulate the Public Markets,” May 27, 1834.

9 For more on the role of public markets in early North American cities, see Helen Tangires, Public Markets and Civic Culture in Nineteenth-Century America (Baltimore: Johns Hopkins University Press, 2003).

10 Tangires, Public Markets and Civic Culture.

11 Isobel Ganton and Joan Winearls, Mapping Toronto’s First Century, 1787–1884 (Toronto: Royal Ontario Museum and University of Toronto Library, 1984); “1858 WS Boulton: Atlas of the City of Toronto and Vicinity,” Historical Maps of Toronto, accessed August 18, 2019, http://oldtorontomaps.blogspot.com/2013/01/1858-ws-boulton-atlas-of-city-of.html.

12 The catalogue listing for the PDF scan of this atlas can be found at https://static.torontopubliclibrary.ca/da/pdfs/912_71354_b594_br_fo_oss.pdf.

13 “Digital Toronto City Directories,” Toronto Public Library, accessed September 25, 2019, https://www.torontopubliclibrary.ca/history-genealogy/lh-digital-city-directories.jsp.

14 “1889 Goad Insurance Plan of the City of Toronto” ArcGIS Online, accessed September 23, 2019, http://arcg.is/1bOnuT0.

15 “Georia: Georeferenced Databases for Accessing Historical Data”, accessed September 23, 2019, http://mercator.geog.utoronto.ca/georia/home.htm.

16 To access an interactive version of Figure 12.4, visit: https://arcg.is/1qrCWq.

17 There now exists a broad international scholarship in urban animal history that highlights the degree to which industrialized cities were built to support livestock husbandry and accommodate populations of a common assemblage of domestic livestock animals including cows, horses, pigs, and chickens. Some prominent examples include Andrea Gaynor, Harvest of the Suburbs: An Environmental History of Growing Food in Australian Cities (Crawley: University of Western Australia Press, 2006); Clay McShane and Joel Tarr, The Horse in the City: Living Machines in the Nineteenth Century (Baltimore: Johns Hopkins University Press, 2007); Peter Atkins, ed., Animal Cities: Beastly Urban Histories (Farnham: Ashgate, 2012); Catherine McNeur, Taming Manhattan: Environmental Battles in the Antebellum City (Cambridge: Harvard University Press, 2014); Sean Kheraj, “Urban Environments and the Animal Nuisance: Domestic Livestock Regulation in Nineteenth-Century Canadian Cities,” Urban History Review/Revue d’histoire urbaine 44, nos. 1–2 (Fall/Spring 2015/2016): 37–55; Dean, Ingram, and Sethna, eds., Animal Metropolis; and Frederick L. Brown, The City Is More Than Human: An Animal History of Seattle (Seattle: University of Washington Press, 2017); scholars of contemporary urbanization have reached similar conclusions to the historical scholarship using a framework that blends urban studies and animals studies that Tora Homberg calls “zoocities” in Urban Animals: Crowding in Zoocities (London: Routledge, 2015).

18 Clemens Wischermann and Philip Howell, “Liminality: A Governing Category in Animate History,” in Animal History in the Modern City: Exploring Liminality, ed. Clemens Wischermann, Aline Steinbrecher, and Philip Howell (London: Blomsbury, 2019), 1.

19 For a complete account of the Great Epizootic of 1872–73 and its effects on urban environments and horses, see Sean Kheraj, “The Great Epizootic of 1872–73: Networks of Animal Disease in North American Urban Environments” Environmental History 23, no. 3 (July 2018): 495–521.

20 An animated version of this map is available at bit.ly/greatepizootic.

21 Sean Kheraj, “Canada’s Historical Newspaper Digitization Problem, Part 2,” Active History, http://activehistory.ca/2014/02/historical-newspaper-digitization-problem/.

22 Richard A. Hawkins, “Digitised Newspapers” Historical Insights: Focus on Teaching (February 2011), https://www.advance-he.ac.uk/knowledge-hub/historical-insights-digitised-newspapers.

23 The largest digitized Canadian newspaper collections available by paid subscription include ProQuest’s historical newspapers, Early Canadiana Online’s periodical collection, and Newspapers.com. Some public archives, libraries and universities collections include the digital magazine and newspaper collection at Bibliothèque et Archives Nationales du Québec, the British Colonist collection from University of Victoria, the Simon Fraser University digitized newspaper collection, and Island Newspapers by the Robertson Library at University of Prince Edward Island. The most comprehensive index of digital historical newspaper collections in Canada is “Digitized Newspapers and Magazines” compiled by Kiera Mitchell, Brandi Adams, and Donica Belisle, accessed September 27, 2019, https://www.donicabelisle.com/digitizedhistoricalperiodicals.

24 For this map, I made extensive use of “Chronicling America” a mass newspaper digitization project of Library of Congress and GenealogyBank.com, a large private genealogy company that offers an extensive digitized historical newspaper collection with access by paid subscription. See “Chronicling America” at https://chroniclingamerica.loc.gov/.

25 In order to activate the date-time field, users must first publish their layers as “Featured Layers.”

26 For more information about Time Aware, visit https://arcg.is/0u1b9i.

27 “Railroads and the Making of Modern America: A Digital History Project,” accessed September 27, 2019, http://railroads.unl.edu

28 “US Railroads 1870” ArcGIS Online featured layer, accessed September 27, 2019, http://arcg.is/08DTP00; “Canadian Historic Railways,” ArcGIS Online featured layer, accessed September 27, 2019, http://arcg.is/1O0D4a.

29 Ian Milligan, History in the Age of Abundance? How the Web Is Transforming Historical Research (Montreal: McGill-Queen’s University Press, 2019) 7.

30 Douglas Seefeldt and William G. Thomas, “What is Digital History?” Perspectives on History (May 2009), accessed October 18, 2019, https://www.historians.org/publications-and-directories/perspectives-on-history/may-2009/what-is-digital-history

31 William J. Turkel and Devon Elliot, “Making and Playing with Models: Using Rapid Prototyping to Explore the History and Technology of Stage Magic” in Pastplay: Teaching and Learning History with Technology, ed. Kevin Kee (Ann Arbor: University of Michigan Press, 2014), 176.

32 John Berger, About Looking (New York: Pantheon Books, 1980), 6.