Case Study 1: John Snow's Cholera Map and the Birth of Epidemiological Cartography

Case Study 1: John Snow's Cholera Map and the Birth of Epidemiological Cartography

In September 1854, a cholera outbreak killed hundreds of residents of the Soho neighborhood in London. A physician named John Snow, skeptical of the prevailing "miasma" theory that cholera was caused by bad air, mapped the deaths block by block. The map showed a cluster around a public water pump on Broad Street. Snow convinced the local authorities to remove the pump's handle, and the outbreak subsided. The map became one of the most famous data visualizations in history — and one of the first uses of geographic visualization to establish a causal argument about disease.

The Situation: Cholera in Victorian London

London in the 1850s was the largest city in the world, and it was filthy. The population had exploded during the Industrial Revolution, outgrowing the city's sanitation infrastructure. Human waste was dumped into cesspools under buildings, which leaked into the ground and contaminated wells. The Thames, which supplied most of the city's drinking water, was fouled with sewage from every neighborhood upstream. The water that came from street pumps was variably clean, depending on where the pump was located and what was leaking into the groundwater nearby.

Cholera had arrived in England in 1831-1832, brought from India via trade routes, and had caused several major epidemics in London through the 1840s and early 1850s. The disease was terrifying: healthy people could die within hours of symptoms onset, and mortality rates in outbreaks were 20% or higher. The medical establishment's prevailing explanation was the miasma theory — the idea that diseases were caused by "bad air" emanating from decaying matter. If you lived near a slaughterhouse, a tannery, a cesspool, or a dumping ground, you breathed the miasma and got sick. The theory had been held for centuries and seemed plausible: diseases did cluster in dirty, smelly neighborhoods.

Not everyone believed it. A few physicians and sanitation reformers had argued that cholera was transmitted through contaminated water, not bad air. One of them was John Snow, a London physician who had been studying cholera since the 1840s. Snow had published a paper in 1849 arguing that cholera was a waterborne disease, but he lacked definitive evidence. The medical establishment dismissed his argument as eccentric.

In late August 1854, a severe outbreak began in the Soho neighborhood. Within a week, over 500 people were dead. The epidemic was concentrated in a small area around Broad Street (now Broadwick Street), and Snow lived nearby. He decided to investigate the outbreak in person.

The Data: Counting Deaths, Knocking on Doors

Snow's investigation was methodical and time-consuming. He went door to door in the affected neighborhood, asking about deaths, symptoms, and water sources. For each confirmed cholera death, he recorded the address. He visited the local water pumps and collected samples. He interviewed the families of the dead and asked where they had obtained their drinking water.

The data he collected was not a public health dataset in the modern sense. There was no standard format for infectious disease surveillance. There were no health department databases. Snow worked with the General Register Office, which recorded deaths by date and location, and he supplemented those records with his own door-to-door survey. The whole effort was driven by one physician with a hypothesis and time to walk the streets.

After weeks of work, Snow had:

The locations of every confirmed cholera death in the Soho neighborhood (over 500 cases).
The locations of every public water pump in the area (about 13 pumps).
Interviews with victims' families about which pump they had used for drinking water.
Background information about pump water sources — which pumps drew from deep wells, which from shallow ones, and which had known contamination issues.

This was enough to build a map.

The Visualization: Dots and Pumps

Snow's map of Soho, published as an illustration in his 1855 book On the Mode of Communication of Cholera (second edition), was simple by modern standards. A hand-drawn street map of the neighborhood served as the base. On this base, Snow marked:

Each cholera death with a small black bar drawn at the address. Houses with multiple deaths had stacked bars. The bars were dense at addresses with many deaths, sparse at addresses with few.
Each public water pump with a small dot or X at its location.

The map had no color (it was printed in black and white), no legend beyond a simple key, and no statistical ornamentation. It was essentially a dot map — a specific kind of geographic visualization where individual events are placed at their physical locations and the reader sees the pattern of where they cluster.

The pattern was immediately visible. The deaths formed a dense cluster centered on one specific pump: the Broad Street pump. Houses close to the pump had many deaths; houses farther away had fewer. The pattern radiated out from the pump like a target. Some anomalies — a house near the pump with few deaths — turned out to have explanations (the residents drank from a different source). Other anomalies — a death far from the pump — also had explanations (the victim had worked near the pump and drunk from it).

Snow used the map as evidence for his argument that cholera was waterborne. The visual pattern was too compelling to ignore: the cluster of deaths around the pump was not what you would expect from a miasma that should have affected the entire neighborhood roughly uniformly. It was what you would expect from a contaminated point source that people drank from.

The Action: The Pump Handle

Snow presented his findings to the Board of Guardians of St. James Parish, the local administrative body responsible for public health. He argued that the Broad Street pump was the source of the outbreak and that closing it would stop the epidemic. The board was initially skeptical — Snow was making an unconventional argument against the dominant medical theory — but eventually agreed to remove the pump's handle as a precautionary measure on September 8, 1854.

The outbreak subsided soon after. Snow's argument was vindicated, though not immediately — the medical establishment was slow to accept the waterborne theory, and the miasma theory continued to dominate public health discourse for another decade. Snow himself died in 1858, before his ideas were fully accepted.

The delayed vindication came in the 1860s and 1870s, as further epidemiological investigations confirmed the waterborne nature of cholera, and as Koch's later work on germ theory identified the specific bacterium (Vibrio cholerae) responsible. By the end of the 19th century, the waterborne theory was accepted, sanitation reforms had been implemented in major cities, and cholera had ceased to be a major threat in developed countries. Snow is now credited as one of the founders of modern epidemiology, and the Broad Street pump episode is taught in every public health program as a foundational example.

The Map as Argument

Snow's map is famous not because it was technically sophisticated — it was a simple hand-drawn dot map — but because it was rhetorically effective. The visual argument was what convinced the Board of Guardians to remove the pump handle. A table of addresses and death counts would not have been as persuasive. A verbal description of the spatial pattern would not have been as immediate. The map made the argument visible at a glance.

This is the chapter's threshold concept applied to a historical case: maps are arguments about space. Snow's map was an argument that the deaths clustered around a specific pump. The argument was correct, but it could have been wrong — if the deaths had clustered elsewhere, the map would have shown that instead, and the conclusion would have been different. The map did not manufacture the argument; it revealed the pattern that supported an argument. But the pattern was only visible because Snow made the map. Without the map, the door-to-door data would have been a list of addresses without a coherent shape.

The map also illustrates several of the chapter's design principles:

Unit of analysis matters. Snow's map showed individual deaths (dots), not regional rates (choropleth). This was the right choice for the question: "is there a cluster at a specific location?" A choropleth of cholera rates by London neighborhood would have obscured the point-source pattern.
Normalization was not needed. Because the unit was individual deaths and the reader was looking for spatial clusters, raw counts were appropriate. A per-capita normalization would have been confusing without adding information.
The base map carried context. Snow's dots were drawn on a street map of the neighborhood. The streets and buildings gave readers a familiar reference frame. Readers could see which dots were near which buildings and water sources without additional annotation.
The pumps were highlighted. The water pumps were the potential causes Snow was investigating, so they were drawn prominently. A map that omitted the pumps would not have supported the argument, even with the same death data.
Simplicity beat ornamentation. Snow's map is striking partly because it is so plain. No colors, no decoration, just dots and pump locations. The data-ink ratio (which Tufte would formalize over a century later) was high: every mark on the map was meaningful.

Theory Connection: Why This Map Still Teaches

Snow's map is the single most-cited historical example in visualization pedagogy for several reasons.

It shows visualization solving a real problem. Most visualization examples in textbooks are clean datasets that illustrate techniques without consequential stakes. Snow's map saved lives — it led to closing a contaminated water source and stopping an outbreak. The stakes make the example memorable and underline why visualization matters.

It challenges the prevailing theory. Snow was arguing against the mainstream medical establishment, and the map was his primary evidence. This reminds students that visualization can be a tool for dissenting arguments, not just for confirming established views. A visualization can challenge received wisdom by making a pattern visible that the wisdom cannot explain.

It uses no advanced techniques. The map is a hand-drawn dot map with a street base. No color, no statistics, no interactive features. This shows that the core of effective geographic visualization is not technical sophistication but a good match between the data, the question, and the visual representation. The tools in this chapter (Plotly, geopandas, Folium) can produce Snow's map in seconds, but they cannot replace the decision about what to map and why.

It has a clean causal story. Remove the pump handle; outbreak subsides. The cause-and-effect is direct and memorable, which is rare in epidemiology (most disease clusters have ambiguous causes). The clarity of the example lets teachers focus on the visualization without getting bogged down in epidemiological caveats.

The takeaway for practitioners: a simple map of the right data can be more powerful than a sophisticated chart of the wrong data. The technology does not matter if the design is wrong, and the design cannot be right without careful thought about what question the map is answering. Snow's map is 170 years old and still teaches more about geographic visualization than most modern examples.

Discussion Questions

On data collection. Snow's data came from door-to-door interviews conducted by one physician. Modern epidemiological data comes from systematic health department surveillance. What changes in visualization practice when the data collection is centralized and automated rather than artisanal?
On simplicity. Snow's map is technically primitive but rhetorically powerful. Could a modern Plotly or geopandas version improve it? Would the interactivity add value, or dilute the clarity?
On establishing causation. The map showed a spatial correlation between deaths and the Broad Street pump, but correlation is not causation. What additional evidence would be needed to establish that the pump caused the outbreak, and did Snow have that evidence?
On the medical establishment's resistance. The miasma theory dominated even after Snow's map was published. Why do visualizations sometimes fail to change minds, even when they clearly support an argument?
On modern analogs. The COVID-19 pandemic produced many geographic visualizations tracking outbreaks, hotspots, and transmission patterns. Which of these were most effective? How do they compare to Snow's map?
On Plotly reconstruction. Using the tools in this chapter, how would you rebuild Snow's map in Python? What data would you need? Which library is the best fit?

John Snow's cholera map is the founding example of epidemiological cartography and one of the most famous visualizations in history. It demonstrates the core lessons of this chapter: that a map is an argument about space, that the right data plotted the right way can be more persuasive than any volume of words or tables, and that the design choices — what to include, what to omit, what to emphasize — matter more than the technology used to draw the map. Every modern tool in this chapter can reproduce Snow's map in seconds, but the discipline of asking the right question and trusting the pattern in the data is the lesson that endures.