It is surely true that not everything that stinks kills, and not everything that kills stinks, but many late nineteenth-century post-bacteriological-age French hygienic pundits continued to exploit the association of deadly disease with the ‘filth’–stench–human-excrement nexus. They established what David Barnes calls the ‘sanitary–bacteriological synthesis’, or SBS (p. 135), a relationship of reciprocal support between lay sensibilities (an array of bourgeois revulsions, of which fear of the occult transmission of excremental effects was the keystone) and the expert knowledge of the new profession of bacteriologists. This SBS conquered educated taste; we justify sensibilities by pretending to a bacteriological rationale for them but on our own terms, and we feel much freer to dismiss the experts' warnings where the modes of disease transmission do not involve excrement, odours or imaginations about them. And even where they do, we disregard the experts when their recommendations are too inconvenient.

Barnes finds the origin of this synthesis in France. It was in consolidation mode during the ‘great stink’ of 1880. Perhaps Paris stank unusually during the late summer and autumn of 1880, though Barnes finds it impossible to identify any particular cause. Certainly there was a spasm of literary disgust, with journalists and government officials taking steps to avoid being seen as soft on smells. Drawing support from experts, including Louis Pasteur, those concerned insisted that the smelly emanations which outraged citizens (surely from human excrement) were deadly, perhaps because the emanations carried deadly germs. In fact the smells did not manifest themselves on mortality tables. When, fifteen years later, Paris again stank, it was not necessary to claim that emanations carried germs. The particular association could be dispensed with, so secure was the general trust that bacteriology and popular sensibility marched in lockstep against dirt, superstition and barbarousness.

The two ‘stinks’ mark the beginning and end of this book; its four middle chapters explore the changing motifs of sanitation and meanings of filth in nineteenth-century France. The second chapter moves back to the well-known hygienists of the early century, L. R. Villerme and Alphonse Parent-Duchatelet. The short and superb third chapter lays out (some) of the several non-exclusive modes of aetiological accounting: the local, personal and folk aetiologies that may coexist with academic aetiology. The fourth chapter treats the translation of the germ theory into existing modes of response to epidemics, and the fifth chapter charts the theory's insinuation into national and colonial discourses. Whether or not a sanitary–bacteriological synthesis is a uniquely French invention (I suspect not), nineteenth-century France does serve as an excellent locus for exploring the meanings of insanitation, precisely because sanitation was never effectively incorporated into a legal or a technical framework. Public health debate was ‘loud, contentious, remarkably unproductive’ (p. 8) – and ongoing. At century's end, systematic urban piped sewerage remained a controversial idea. Small, local and private nuisances were preferable to large and costly public nuisances, many argued. Hence health concerns could not so readily be translated into pipe dimensions, professional credentials or impenetrable bureaucracy, but instead were persistently public – matters for mayors and sub-prefects, for journalists and local doctors, and for the ‘epidemic doctors’, sent by the state to investigate and recommend, but with no powers actually to change things.

Echoing Charles Dickens, Barnes asserts that ‘the germ theory of disease changed everything and nothing at all’ (p. 2). The antithesis of this fine book is a view in which bacteriology, the medicine of the laboratory, rapidly displaces a mix of miasmatism and moralism. While no ‘straw man’, such a view is less prominent than it once was, but no clear replacement has emerged. Here, the moralism never gives way; public health campaigns remain social movements. At many levels, in many ways, and towards many ends, writers creatively combined elements of a hygienic cosmos. They enlisted these to praise or condemn governments, to assail the persistent if declining Catholic Church, to turn backward Bretons into civilized French, to manifest class and so on. By century's end, germs – tethered, though on a long leash, to the new bacteriologists – are among the most important of elements, but they, and the others, are more important as idiom than as colonies isolated in a growth medium.

This book should not be read solely, or even mainly, as a contribution to the history of French public health. The ‘sanitary–bacteriological synthesis’ which Barnes proposes is at once an event occurring in France, a historical generalization of prominent features of modern medicine, an entity acquiring a certain inertial moment in the microphysics of power, and an explanation. It is also, perhaps, a broader assertion about the status of medicine as culture, and a manifesto about how to understand medical events and carry on the business of medical history. While it should be all of these things, sometimes, these several senses jangle. Thus the SBS is declared ‘not a thing’ and then reified as a ‘framework that structures’ (p. 262). It works best as descriptive generalization. Barnes eschews any master explanation in favour of ‘a workaday, pragmatic bricolage of methods and analytical perspectives’ (p. 5), yet Michel Foucault, Mary Douglas, Norbert Elias and Pierre Bourdieu do not blend smoothly without further work.

Barnes's focus on layers of meanings (p. 8) works well, most strikingly in elucidating the multivalent potential of the term ‘infection’, but it comes with costs. We get little sense of the importance of professionalized power, for example – of the enormous impact credentialled technical authority had come to have in France (on which matter, reaching back to read the nineteenth-century campaigns in terms of late eighteenth-century urban eudiometry would be illuminating). Mostly, despite village-level vignettes, this is a book without protagonists. Finally, notwithstanding an intriguing interpretation of Pasteur's panic-mongering in 1880 (p. 48), there is little here about aetiological science, whether bacteriology or its predecessors. A more careful study of the many varieties of so-called miasmas (a concept often oversimplified) might clarify. So too, one might argue that a key part of the story should be the accumulation of empirical studies of microbe dispersion during the 1880s and 1890s, particularly from sewer water to air. The SBS becomes all the more remarkable when one realizes that such studies often undermined its premises.

To state limitations is to characterize, not criticize. Great Stink is one of the most important books in the history of public health in recent years. It is a must-read for historians of public health of whatever period or place.