This is a review of
The Ghost Map by Steven Johnson.
Johnson's account of John Snow, Henry Whitehead, and other key figures in the story of a nineteenth century London cholera epidemic is an engaging narrative with insights about scientific investigation and persuasive argument. Starting with an orientation to London c. 1854, the heart of the book explains how a particularly perspicacious investigator developed and tested hypotheses to reach correct and influential conclusions. Along the way we see how somewhat different approaches led others to different outcomes. Of course, the process of scientific discovery remains relevant to us in the 21st century, as are many of the issues concerning urban design and public policy.
The Broad Street cholera epidemic of 1854 was in large part a consequence of urbanization with inadequate public works infrastructure, at a very early stage in the development of medical science and microbiology. Particularly in the last chapter, "Broad Street Revisited," Johnson discusses the relation between 1854 London and present-day first-world city planning. Citing
Jane Jacobs and other contemporary sources, he discusses the architecture of cities, the role of government organizations such as the CDC, and even the War on Terror and nuclear non-proliferation. I think he overreached in this attempt to reinforce the continuing relevance of the Snow/Whitehead collaboration and to advocate for positions on tenuously connected modern issues.
Readers of Tufte's
Visual Explanations will recognize this epidemic from several pages in that book (it was also mentioned in
The Visual Display of Quantitative Information). While Tufte focuses on an analysis of Snow's compelling maps as designed artifacts, Johnson describes in detail the scientific inquiry that led to the development of those maps. Johnson also offers a short critique of Tufte's treatment. It is disappointing that after describing
Snow's second-edition Voronoi-enhanced map (
see also) as "Snow's most significant contribution to the field of disease mapping" and calling out Tufte on having overlooked it, Johnson omits the diagram from his own book.
The Ghost Map includes end notes, but the note numbers appear only in the notes and not in the text. The absence of a visual indication that more detail is available makes the end notes less likely to be used and made me less trusting about some of the author's assertions.
I have pointed out some weaknesses in The Ghost Map, but on the whole I think it was well worth reading. If you are interested in Snow's maps, the history of epidemiology, or the process of scientific inquiry, I heartily recommend this book.