Knowledge in the Time of Cholera (22 page)

Beginning in the 1850s, the association of cholera and filth was given a concrete form by three new techniques—dot maps, sanitary surveys, and the statistical artifact of “normal mortality.” In many ways, this association required no mean feat of investigation. The copresence of squalor, filth, and disease was apparent to all familiar with certain urban streets. The
New York Times
(June 25, 1856, 3) noted that just walking down the street was enough for someone to “consent that the
sense of smelling
itself is a nuisance.” What the new techniques offered sanitary reformers was the elaboration and systemization of such anecdotal, sensory observations. Less important as sources of new insight, they served to illustrate existing ideas, providing visual representations that could be rhetorically deployed in the advocacy of sanitary reform. And sanitarians used them to great effect.

Mapping Disease—
Even though the perceived link between environment and disease was not new to the nineteenth century,
²
doctors and sanitarians began to illustrate this link in the mid-1800s using maps. While most early maps of cholera were European (especially English) in origin, by the 1850s, Americans were involved in mapping disease (Barrett 1996). The growing prominence of medical mapping received great impetus from epidemic diseases that emerged during this period (Jarcho 1970, 138), especially cholera (Vinten-Johansen et al. 2003, 322). This interest in mapping epidemic diseases coincided with technical advances in printing, specifically the shift from copperplate to lithography, which allowed for more elaborate maps and more efficient dissemination of these maps (Koch 2005, 41). In this “golden age of medical cartography” (Gilbert 1958, 173), sanitary reformers adapted maps not only to make sense of cholera but also to illustrate particular arguments as to its nature.

During this period, two types of cholera maps were produced. The earliest maps were progress maps that traced the spread of cholera across large areas (Vinten-Johansen et al. 2003, 323), visually representing the move
ment
of cholera
through
space. They emphasized the mobility of disease. As such, progress maps promoted a view of cholera more favorable to contagion theories. Temporality was an important dimension, as dates of outbreaks in specific areas were often noted along the routes that cholera took. However, because they covered wide geographical spaces, progress maps were limited by certain informational barriers; it was difficult to accumulate accurate information from disparate geographic locations. Since the United States lacked an adequate infrastructure for the collection and aggregation of vital statistics, the data behind progress maps lacked the requisite reliability, and as such, the maps had limited impact.

Of much greater importance were the dot maps, or spot maps, of cholera. In the mid-1800s, cartographers began to focus on the local circumstances of areas with the highest incidence of cholera. The most famous of these nineteenth-century dot maps was John Snow's map of the 1854 cholera outbreak in London.
³
Unlike progress maps, the intent of dot maps was to show the clustering of disease
in
space. Cartographers mapped out a geographical area, and then using “dots” or some other marking device, noted the incidence of disease. Clusters of disease reflected unhealthy local conditions.
⁴
As techniques developed, they used more sophisticated techniques for “spotting” their maps, such as shading techniques to visually relate the incidence of cases to population density. Interpretively, dot maps suggested a robust correlation between place and disease.

Like all maps, the dot maps were not mere presentations of facts. They marshaled selected propositions into arguments about the nature of the disease (Koch 2005). In other words, the cholera dot maps were arguments in visual form, and the argument they sought to convey was that there was a relationship between disease and some local factor (e.g., nuisances, cesspools, inadequate sanitation). For example, Snow's map linked disease to contaminated water sources, most famously the Broad Street pump (Johnson 2006). Indeed, it was this ability to visually convey information that was most alluring about the maps. Complex, poorly understood disease properties were reduced to dots on a map—dots when aggregated conveyed a pattern not readily “seen” otherwise.

For sanitarians maps performed the crucial tasks of simplification, reduction, and translation. And they interacted with the viewers in a seemingly transparent way. Knowledge claims were made visible and legible to the public. Like statistics for homeopaths, maps made arguments through demonstrations of knowledge, albeit in visual form. In this respect, they sug
gested
a more democratic form of knowing, even if this democratic commitment existed more in appearance than substantively. Indeed, as arguments in visual form, the maps excluded and masked as much as they included and revealed. Still, they made knowledge
seem
transparent and open to public evaluation. It is the appearance of transparency, and the unconscious way in which maps insinuate arguments (Boggs 1947), that make them effective
rhetorically.
Dot maps took complex arguments about the nature of cholera and transformed them into simple visual representations, rendering them perceptible to all.

This dot map indicates every cholera case (represented as black boxes).
Map of Lancaster, Kentucky, Showing the Location of Each Cholera Case in 1873
. Courtesy of the National Library of Medicine.

This spot map illustrates the more advanced mapping technique of shading, showing the varying incidences of cholera outbreaks through shading (the darker the shading, the higher the incidence) in Nashville, Tennessee, from Ely McClellan, 1873.
Map of the Cholera Epidemic in Nashville
, Tenn., in 1873. Courtesy of the National Library of Medicine.

The message dot maps conveyed was clear: when it came to disease, place mattered. While neither inherently contagionist nor noncontagionist, dot maps fostered a certain localism by offering a static picture of cholera (Stevenson 1965). By encouraging an “inherently ecological” way of thinking (Koch 2005, 2), they focused viewers' attention to the environmental context of disease. They established a spatial relationship of disease and some other local factor (a source of putrefaction, a ship or a pier), rooting a disease in place by relating it to fixed characteristics of the environment. Consequently, while the maps did not exclude any notion of contagion, these notions would have had to be actively read onto the visual representation. They were not inherently noncontagionist, but they made contagion more difficult to see.

Normal Mortality—
A great obstacle for medical cartographers was gathering accurate knowledge to collate and fashion into useful maps (Osborne 2000). Maps were joined with and backed by vital statistics. Indeed, historically the emergence of medical mapping coincided with the development of medical statistics, as both shared a common intellectual base as part of the slow progression of medical science (Koch 2005, 8). But more than just gathering raw data, cartographers had to construct statistical techniques and measures to impose order on such data.

Sanitarians drew on statistics to support the visual arguments of the dot maps. It was not enough to show that diseases clustered in some areas more than others; cartographers also needed to show that this clustering was somehow exceptional. After all, people naturally had to die somewhere. To indicate the presence of preventable deaths, sanitarians developed a statistical artifact—the “uniform law of mortality” (Shattuck 1850, 95) or the “normal death-rate” (Smith 1911, 119). This measure was used to support the argument that certain sanitary locales were unnatural as they produced outcomes that violated the normal incidence of mortality. In essence, American sanitarians cobbled together the limited available vital statistics and calculated the mortality rate due to “inevitable causes,” like old age, accidents, and endemic disease. This was deemed the normal death rate. Deaths from epidemic diseases fell outside the normal death rate, and, in turn, were by definition preventable. Through neat, obfuscating data management, sanitarians reframed deaths from cholera as aberrations to be prevented by
the
proper application of sanitary reforms, rather than part of the normal course of nineteenth-century living.

With the concept of normal mortality underlying their construction, dot maps became visual representations of nonnormal mortality that was linked to filthy local conditions. Rhetorically, this statistical ratio possessed all of the benefits—reduction, simplicity, and mobility—that the homeopathic statistics had. The measure effectively transformed the central message of the sanitary movement—sanitation could prevent unnecessary death—into a simple, seemingly transparent numerical ratio.

Sanitary Surveys—
The normal death rate supplied an abstract, rough numerical indicator of the pernicious effect of poor sanitation on health. It quantified disease abnormality but did so via abstraction from the reality on the ground. Behind this abstraction, however, was the mundane activity of data collection. To create dot maps and to calculate normal death rates, reformers had to acquire hard data on local conditions that would explain the variation in mortality rates and the patterns observable on dot maps, to pinpoint the conditions of filth that lead to nonnormal death. Sanitarians conducted sanitary surveys that not only provided important information but also became their most effective rhetorical tool.

Sanitary surveys required good old-fashion “shoe-leather” research to gather local intelligence. Indeed, America's most important contribution to medical cartography was the idea that the local conditions of disease could not be gleaned through correspondence or from decontextualized statistical tables. Researchers had to go to the place to understand the relationship between disease and place (Barrett 1996). To produce their sanitary surveys, sanitarians divided a geographic locale into discrete sections and then sent surveyors to note the conditions on the ground and take inventory of a plethora of environmental factors. The comprehensiveness of the sanitary surveys reflected sanitarians' desire to reconstruct the total environment of a specific locale, because the total environment might be relevant to the prevalence of disease. These diverse conditions were then correlated with “prevailing sickness and disease” in all the specific locales that when amassed provided a comprehensive sanitary picture of the city. The most famous of these surveys was the
Report of the Council of Hygiene and Public Health of the Citizens' Association of New York upon the Sanitary Condition of the City
(1866), which would a play key role in securing a permanent board of health in New York City.

Sanitary surveys provided a “ground level” view of local conditions. In a
sense
they rendered the city legible, as they were the primary instrument by which sanitarians transformed sensory observations into first, systematic data and second, visual dot maps that spatialized the data of disease. Such a “thorough and systematic sanitary inspection by competent experts” (Citizens' Association of New York 1866, xxii) would confirm:

what reason should have taught every person, however uneducated, that filth, overcrowding, bad drainage, excessive humidity, imperfect supply of air and sunlight, neglect of excrementitious and decaying material, and the putrid exhalations from sinks, sewers, gutters, and dirty streets, will both produce and perpetuate disease; and that whatever sickness occurs in such localities will be more virulent and destructive than the same or similar maladies when occurring in places where such conditions do not prevail. (Citizens' Association of New York 1866, lxiii)

Through the sanitary surveys, local knowledge and anecdotal observations were systematically amassed so as to transform them into the generalized, abstract representations of dot maps and normal morality ratios.