BACTERIOLOGY AS A CULTURAL SYSTEM: ANALYSIS AND ITS DISCONTENTS

Hist. Sci., xlix (2011)

BACTERIOLOGY AS A CULTURAL SYSTEM: ANALYSIS AND ITS DISCONTENTS Christopher Hamlin University of Notre Dame A cultural system in two senses, bacteriology is underrepresented in the history and philosophy of science (significantly, however, it was the home science of the great prophet of underdetermination, Ludwik Fleck).1 Pasteur and Koch garner attention, but more as founders than practitioners.2 A quarter century ago, in 1987, Olga Amsterdamska called attention to the peculiarity of the enterprise, a biomedical praxis tied only loosely to natural history, to biology, or even to science.3 In this paper I shall examine the bacteriology presented in American textbooks between 1935 and 1950 in the terms of John Pickstone’s Ways of knowing: A new history of science, technology, and medicine (2000), and supporting work (2007, 2009).4 In the book, Pickstone identified world-reading (natural philosophy), natural history, analysis, and experimental synthesis as distinct modes of knowing, which had arisen in a broadly sequential relation. Later writings would extend the concept to complementary ways of working and make clearer the concurrence of multiple ways of working and knowing in particular times and domains — there was no teleology here, nor were the ways of knowing/working mutually exclusive. From a Pickstonian standpoint, bacteriology is exemplary in two respects. First, there, the categories of natural philosophy, natural history, analysis, and rationalized synthetic craft are almost always conflated though in changing ways. Second, any working–knowing divide dissolves. There is an enormous emphasis on the pragmatic and operational: theories may underwrite practice, but practice is more important. Bacteriology may fit Pickstone, but does Pickstonizing bring anything to it? Some years ago as a reviewer, I suggested that Ways of knowing had a Kuhn-like potential to inform our understanding of science (and technology and medicine). I thought it might link Kuhn to the emerging literature on technoscientific practice and enrich our characterization of large scale units of activity (i.e., paradigms). That enrichment in turn, might help to clarify their dynamics. Kuhn had seen the overturning of paradigms primarily in epistemic terms. Accumulated anomalies brought revolution. He had not gone far into distinguishing multiple ways of knowing and doing, or to exploring factors that might give a technoscientific enterprise extended and trans-epistemic viability. Pickstone’s analytic, by contrast, is applicable not only to diachronic changes within a technoscientific enterprise, but also to synchronic differences within and among the enterprises in a particular time and place: it offers, in short, a complement not only to Kuhn, but to J. T. Merz, whose classic History of European thought in the nineteenth century was equally about ways of knowing. By offering a richer set of categories for appreciating what sorts of work ‘working 0073-2753/11/4903-0269/$10.00 © 2011 Science History Publications Ltd

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

270 ·

CHRISTOPHER HAMLIN

knowledges’ might do, Pickstone’s approach eased burdens of epistemic accountability without sacrificing the benefits of a systematic comparative framework. representations of bacteriology

By the 1930s, a third generation of bacteriologists practised a discipline distinct — in terms of institutional organization, textbook traditions, modes of training, and career possibilities — from botany and zoology; equally from the other main biomedical laboratory science, pathology; and from the applied, and significantly colonial, enterprises of genetics, parasitology and veterinary medicine. They had, as we shall see, subverted an incipient biochemistry, which challenged the sanctity of the bacterial species concept. They had spawned as a parallel endeavour, immunology, and its chief practical arm, serology: the relation of immunology to bacteriology would be ambiguous and often contested — serological specificity was a means of defining bacterial specificity, and yet many serologists leaned dangerously toward general physiology and physical chemistry. Any picture of a community of happy equals, with each bacteriologist investigating the wee beasties of his (or rarely her) chosen domain is misleading. Study of human pathogens was the discipline’s flagship area; the roles of microorganisms in sewage purification, soil nitrification, or cheese-making, were marginalized as agricultural or applied bacteriology.5 Strikingly, such investigations were taken most seriously by Russians or Russian émigrés, like Selmen Waksman and Serge Winogradsky.6 But if the Russians were on the margins, the Germans (secondarily the French) were at the centre: Berlin was bacteriology’s capital. All this would soon change. After 1950, the normality of that normal science began to crumble. Within laboratories and beyond, in connection with biochemistry, genetics, ecology, and population biology, the status of bacteria, their utility, and the disciplinary ownership of bacterial phenomena would all change. Viruses displaced bacteria at the vanguard of infectious diseases research.7 The previously distinct bacteriology would be swallowed by a broader microbiology, more reflective of the integrative agendas of general biology. Remarkably, these changes did not render bacteria themselves unimportant. But greater and changed relevance came at the cost to independence and professional distinctiveness. Histories of bacteriology often reflect the strains that pre-existed that change. Some, writing in the tradition of William Bulloch in 1930, have highlighted the achievement of the “orthodox” medically-oriented bacteriology. What is called (generally by critics) Cohn-Koch monomorphism (faith in a correlation between distinct bacterial species and distinct diseases) was presented as the apotheosis of progress in etiology through the ages.8 But others, following Fleck, for whom the peculiarities of his home discipline were fodder for the anti-epistemic overtones of Genesis and development of a scientific fact (1935), highlighted the persistence of tensions within that dogma from the very beginning of institutionalized bacteriology in Paris and Berlin around 1880: laboratory objects (“pure” cultures) refused to be stabilized; laboratory truths were defied in the field. Each new round of resolving methodologies brought only a new wave of anomaly. With respect to the immunological/serological side of the

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

271

enterprise (Fleck’s home turf), Pauline Mazumdar has elegantly shown the repeated iterations of a single grand dialectic — insistence on the one side that bacteria existed as Aristotelian natural kinds; on the other that bacteriology’s units were fictitious and its settled truths artifacts of the laboratory.9 Others too (e.g., Rob Kohler, Olga Amsterdamska, Thomas Brock, William Summers, Andrew Mendelsohn, Ilana Löwy, and Warwick Anderson) raised questions about greater complexity within the domain, and about incipient or marginalized endeavours, prescient paths not taken.10 Whether or not revisers of the Bullochian heritage were explicitly Kuhnian, their work was framed by The structure of scientific revolutions: indeed, there was no adequate alternative.11 But there were problems. While commentators noted the centrality of practice in bacteriology, they found it hard to avoid assessing it in the predominantly cognitive terms of paradigms (or Fleck’s “thought collectives”). A tipping point of anomaly, or a sophisticated falsification of a Lakatosian research programme, was to trigger change, but what happened after 1950 was less refutation than reconfiguration of ways of knowing/working as bacteria acquired new significations. Preoccupation with stability became fascination with variability.12 Rather than fixating on the bacteriology that wasn’t, Pickstone’s approach promises a better sense of that which was. It does so by escaping both a hierarchical theory–practice–institution evaluative framework and the motivational complexities of technoscience, by focusing instead on the complex of working relations and the meanings they carried. The complex of telling stories, classifying, taking things apart, and the rationalized making of new things, cuts across theories, practices and institutions. All were epistemically significant activities without necessarily being reducible to epistemology, long-reigning queen of science studies. Thus Pickstone might help us understand sympathetically Bulloch’s bacteriology in terms other than Bulloch’s triumphalist autobiography or Kuhnian paradigm formation. I begin with bacteriology’s public face — representations of its territory, cosmic significance, arcane practice, and role in ordering the modern world. I then move inside to examine the bacteriologists’ central problem of finding stable units of knowledge, a taxonomic (natural historical) endeavour that came to rest on analytical, synthetic, and narrative ways of knowing/working. I explore also bacteriologists’ borrowings from other sciences — chiefly chemistry: a borrowing of rationales, protocols, and confident ways of moving forward. At the end of the essay I touch briefly on more recent, and quite different, ways of knowing bacteria. Textbooks, my main sources, are units of Kuhn’s “normal science”. They are sites for managing confusion. Bacteriology was taught at many levels and to many audiences; texts with similar titles differ in depth and coverage. Here I use three introductory American texts. Although not exclusively directed toward training researchers, the 9th (1948) edition of Hans Zinsser’s Textbook of bacteriology, revised by several faculty members at the Duke University Medical School, is effectively a literature review of medical bacteriology. The 3rd (1937) edition of Fred Tanner’s Bacteriology emphasizes medical and non-medical bacteriology. Tanner taught at the University of Illinois, where applications in agriculture and industry, as well as in public health,

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

272 ·

CHRISTOPHER HAMLIN

were important. Finally, the 3rd (1944) edition of Charles Carter’s Microbiology and pathology was addressed to nursing students, who might play roles as bacteriological technicians or public health diagnosticians and practitioners. Carter himself taught bacteriology in Dallas and served as a consulting pathologist.13 Fully persuaded of bacteriology’s importance and promise, these authors are nonetheless remarkably critical assessors of the current state of their enterprise. They admit disunity, controversy, rapid change, and puzzlement. They acknowledge the incoherence of bacterial taxonomy, wonder if the problem is soluble. They understand the artifactuality of laboratory work, but see no alternative. Indeed, so candid are they in their misgivings that one may ask where they find countervailing confidence. The case they make to the public (embodied as the beginning student) for a special science of bacteriology rested on the distinctiveness of its domain, that domain’s vital importance, the uniqueness of the bacteriologist’s art, and the magnitude of the potential payoff. I take each in turn. Who We Are? Mighty, Mighty Microbes Was bacteriology simply a division of biology, defined by a class of beings, or something other? Authors sought a biological warrant. In his 1937 preface, Fred Tanner urged his students to see bacteriology as “one of the biological sciences … [its subject being] a group of living organisms subject to all of the fundamental laws of biology as are other groups”. Charles Carter, in a section on “The bacteriology of Nature” [italics mine], also worried at the neglect of biology — “classification …, cytology … nutrition” — in bacteriological curricula. These protests suggest bacteriology’s ambiguous status. Central biological questions had to be smuggled aboard a domain that had developed around medical applications. A textbook writer must avoid “a popular superficial discussion of applications of microorganisms”, Tanner asserts: applications occupy over half his text (though the treatment is hardly “superficial”).14 Ostensibly, bacteriology belonged to biology because its subjects were living. Exactly what they were was less clear. “Are Bacteria Plants or Animals?” Tanner asks, then drops the question. It made no difference, “as long as … fundamental characteristics are understood”.15 But those characteristics were not understood. The preKochian study of bacteria had occurred in a mid-nineteenth century milieu of interest in fundamental life units and processes — cells, urschleim, cytoplasm, spontaneous generation. Bacteriology’s two great progenitors, Ferdinand Cohn and Carl von Nägeli, had both approached as botanists of the microscopic. Cohn, patron and mentor of Robert Koch, would impose on bacteria the template of natural history, designating fixed types. But, rather than serving as a means to broader ends, description and classification quickly became self-justifying. Central to this “monomorphist” doctrine was the limited variability of each type. Bacteria came as true and stable bacterial types; though methods at any given stage of the science might not be adequate for distinguishing one type from its imposters. Against this “conservative” approach (Zinsser’s term), Nägeli, and, even more radically Ernst Hallier, held that form (and other elements of identity) were

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

273

ephemeral: in their radical pleomorphism, all these were products of function, and in turn of environment. Underlying essence was unknowable and unimportant, for we could not tell what things were, only see what happened; biochemical or pathological activity were simply empirical classes of events. On this issue, followers of Pasteur, while perhaps making greater room for variation, generally sided with Koch. The bacteriology of the half-century after 1880 was the product of their joint influence, though, with the exception of vaccine preparation and virulence, Koch’s is the heavier hand. And yet, positivism (and pleomorphism) persisted, notwithstanding relegation of Nägeli and Hallier to the circle of curmudgeon philosophers (Nägeli was “a dialectician and a verbose philosopher rather than an accurate observer”, sneered Bulloch).16 For, however much bacteriologists might want to take up the natural historian’s project of distinguishing natural kinds, they found that they could do so only by seeing what these cells did. Size and shape were not enough: all comma-shaped vibrios, it was found, did not cause cholera. The new dyestuffs might further distinguish morphologically similar microbes; it did not follow that staining discriminated important natural kinds. For organisms that seemed, at best, merely to split (they were, after all, the fission fungi), life histories were little help. Variable response to media became central in defining species. Bacteria might be more or less adequately distinguished via a motley of unrelated factors, but such a practice did not meet expectations of essentiality by which a form of life could be properly placed within the well-ordered domain of living things.17 The plant–animal distinction might be a false dichotomy, yet it was embarrassing to let rules of thumb substitute for an understanding of what, in effect, the objects of one’s study actually were. Ultimately, Koch’s practice of equating species with diseases did not rest on fulfilment of Cohn’s taxonomic program: rather it bypassed the problem of species altogether by presuming congruence between the domain of bacterial differences and Koch’s methods of studying them.18 That stratagem would make anthrax, the organism-disease on which Koch first worked, paradigm exemplar: all other microbes would be subjected to anthraxbased expectations. What’s Our Story? We Are Your Killers — Though Believe Me, Some of Us Are Rather Useful By the 1940s, bacteriology’s primary intellectual identity — its narrative, natural philosophy, or mode of world-reading in Pickstone’s terms — came from pathology, not botany. For many, it was more than incidentally medical. That context shaped central concepts. First, bacteria were not simply objects, but agents. If we humans were somehow protagonists in our own cosmic stories, they were antagonists. Their arrival was an “invasion” of an unwilling “host” whose response, subject of the parallel science of “immunology”, was “resistance”. At times, the dualism was almost Manichean: “The properties of aggression on the part of the microorganisms are the subject matter of bacteriology. The mechanisms of defence are the materials of immunology. The two disciplines deal with reciprocal reactions of parasitism and

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

274 ·

CHRISTOPHER HAMLIN

cannot be separated.”19 More broadly, agency implied change over stasis, aberration over normality. There was also a counter-narrative, equally anthropocentric. It emphasized the world’s dependence on the continual hidden labour of bacteria. It did not in general serve to enclose or displace the pathological frame, but was rather its ironic inversion. In his single short chapter on “useful” bacteria, Carter felt obliged to italicize his second sentence: “The majority of bacteria are helpful to man, animals, and plants, all of whom are dependent on bacteria for their very existence.” He went on: “From a broad point of view it may be said that the disease-producing bacteria form a small and unimportant group.”20 Mostly, such holism was pious meditation. Textbook diagrams of elemental cyclings were fetishes: they celebrate, rather than inviting research, and even harken back to a long heritage of natural theology. In an end of chapter test, Carter catechises nursing students on the “purpose” of decay with true–false questions: “Bacteria have been said to be Nature’s garbage-disposal system” (T) and “The majority of bacteria are helpful to man” (T). A chapter on “The work of useful bacteria” (ecological functions and industrial applications) gets seven pages in an almost 800-page text. Tanner too appeals to the teleologist’s “Nature”, devoting a subsection to The rôle of bacteria in Nature’s plan. The reflective student would intuit that “nature is a very delicately equilibrated system … [in which] Each group of organisms seems to have a definite function and may be necessary for the existence or growth of other groups”.21 Many of those who perversely emphasized that broader view were primarily agricultural bacteriologists; often it was they, spokespersons for the microbe in the wild, who preached heterodox doctrines of variability.22 Represented as loyal opposition, they might be accommodated — irritating, but not corrosive. Perhaps the most consistently troublesome question in the medical side of bacteriology was the co-adaptation (or even co-evolution) of microbe and host.23 The view of pathogencity as a contingent, epiphenomenal, and, presumably, transitory, mode of relation between microbe and host was recognized most clearly by Theobald Smith. Writing of Smith’s views under the topic of commensalism, the Zinsser authors note that commensal relations “affect both organisms in the association, provoking adaptive changes, and perhaps cyclic and mutational variations”. They admit that the pathological consequences of some microbe infestations may represent “a bungling parasitism”, in contrast with the “skillful or well adapted parasite … [which] resides for long periods in the tissues at the expense of the host, and may produce destructive lesions only as a means of securing an exit …”.24 But such considerations, they appreciate, are curiosities of the margins. Even less palatable was the fitting of pathogenicity into population biology, including the regulation of the human population. If, in Tanner’s delicate system, bacteria kept the world on an even keel, how could one object to their ending of the lives of higher beings? The Zinsser authors noted that the common distinction between parasite and saprophyte (often represented as between bad and good bacteria) was hardly absolute.

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

275

But, by and large, the gulf between narrative and counter-narrative was a no-go area. That the great evil of disease might be inseparable from the maintenance of life was a truth that might occasionally be vented, but, as Fleck makes clear, irony, like biology itself, threatened. Such conundrums contributed nothing to bacteriology’s seek-and-destroy orientation; better just to treat pathogens as malign essences from beyond.25 That master narrative had multiple implications for bacteriology’s relation to surrounding sciences. Not only did phylogeny get short shrift, so too did physiology and pathology, important research fronts in the pre-bacteriological era. The emphasis on matching disease with microbe, and building a dossier on each member of the league of invisible demons, deflected attention from underlying pathology. The units of Koch’s postulates were experimental animal deaths and pure culture recoveries: one might theorize about toxins and their effects, but the form of chaos a microbe caused within was less important than its fact.26 Therapeutics too was implicated: block pathogen access, and there is no need to worry about pathology-informed therapeutics. To represent pathogenesis in terms of an invading horde also diverted attention from environmental, somatic, social, and psychic co-factors, which had often been key concerns in the decades before an agent-based pathology had taken hold. With the partial exception of Carter’s text for nurses, which treats bacteriology in conjunction with pathology, authors made little effort to interface with clinical matters. How Do You Know Us? In the Lab … Though many of its standard operations were novel, bacteriology drew heavily on chemistry — Pickstone’s model science for substantive analysis. Both enterprises were jointly matters of knowing and of doing, and centrally of analysis and rationalization, as Pickstone’s framework suggests . Events in nature were hopelessly messy, but much practical power could accrue from recognition and manipulation of the simple and pure essences that underlay them. Sometimes, e.g., in the reliance on titration to determine potency — terms, tools, and concepts are close analogues. The depth of the borrowing is evident in a casual textbook allusion: “the intestinal canal is a large test-tube from which bacterial products can be absorbed….”27 Whether in selectively taking up dyestuffs, combining with antibodies, or acting as organized ferments, bacteria were effectively reagents. As we shall see, transferring ways of knowing and doing exposed differences too. Often chemistry’s glorious achievements blinded early bacteriologists to the depth of those differences: chemistry was largely quantitative — summative and combinatorial; bacteriology significantly qualitative.28 The profundity of the borrowings was often unrecognized. They occurred on many levels: in experimental design, the employment of wholly operational entities, modes of communication, and the assessment of results and judgement of competence.29 But perhaps the greatest was technique. The novice would find the link between chemistry and bacteriology at the laboratory bench. Both were glassware enterprises; in bacteriology, fetishized technique would overshadow all: more than instrumentation, observation and measurement,

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

276 ·

CHRISTOPHER HAMLIN

mathematical analysis, or conceptual manipulation, the well-made culture was the epistemic guarantor. Thinking distracts from the skilled deployment of rote methods. So teaches Dr Max Gottlieb, the forlorn German émigré immunologist-bacteriologist who has washed up on the medical faculty of a Midwestern state university around 1905, in Sinclair Lewis’s classic medical novel Arrowsmith (1925). Sensing some gift of “craftsmanship” in his pupil Martin Arrowsmith, he asserts the “art in science” (“for a few”), and extols “the beautiful dullness of long labors”. Anticipating the philosopherchemist Michael Polanyi’s privileging of the tacit, Lewis, who was advised by the bacteriologist-science writer Paul de Kruif, writes that “Gottlieb’s wise fingers knew when the peritoneal wall was reached” (italics mine). Teaching his students to flame test tubes, Gottlieb presents laboratory practice as an end in itself: “‘it is a necessity of the technique, and technique, gentlemen, is the beginning of all science.’”30 How Can You Use Us? Lots of Ways … Bacteriology was an industry; routine (presumably) guaranteed its product. Relatively quickly culturing became a regular part of public health administration. There, and, more slowly in biotechnical production industries, the determinative bacteriologist would complement the analytical chemist — many ‘analysts’ would have training in both. Even the medical GP would be trained to perform elementary bacteriological operations, though how many actually used that training is less clear.31 That high demand for instruction is reflected in the explosion of bacteriological research in the post-Koch decades, as teaching was institutionalized in colleges, universities, and technical institutes. Most instructees would not have been researchers, but the fictional career of Martin Arrowsmith (and Fleck’s real one) show how readily the practitioner–researcher border might be crossed. Bacteriology was dangerous and exciting; a way to fame if not fortune. Research required a minimal ante; the circulability of cultures, slides, or photomicrographs gave isolated researchers access to the mainstream, though any remarkable findings were apt to be doubted or dismissed. Indeed, textbook authors worried that easy circulability encouraged too many to play researcher. “New” species were often simply odd appearances of old ones; investigation of bizarre phenomena wasted time and compromised credibility: disciplining bacteriologists was no less important than disciplining bacteria.32 Cohn-Koch monomorphism, bacteriology’s dominant cognitive construct, reflected its practical habitus, Amsterdamska makes clear. “Although, strictly speaking, monomorphism was not a logical consequence or a necessary prerequisite of the germ theory of disease, it guaranteed that the results of laboratory investigations of aetiology would be medically relevant.… [T]he heuristically adopted assumption of monomorphism legitimized certain laboratory methods, implied that others were inefficient, and, at the same time, provided standards and criteria of evaluation.”33 The most important use of that product was to reinforce authority itself. Notwithstanding conceptual and technical limitations, stool cultures for the cholera bacillus were used to regulate travel. Wasserman (syphilis) and Widal (typhoid) tests founded on dubious investments in serological authority, dictated sanity, honesty, and chastity,

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

277

and regulated marriage, employment, and even citizenship.34 At a broader level, these ways of working reflected an age of militarism, colonialism, and racism, with which bacteriology shared categories, rationales, and vocabulary. All belonged to an anxiety about accountability, resolvable by “rationalization”. The “resistance” studied by the immunologists was to “a living foreign protein”.35 “Colony” was the central construct of the culture plate — not only analytic, but, as in the macroscopic world, diagnostic, prophylactic, managerial, and industrial. The “pure” culture of the colony produced the desired chemical or pathological species.36 Discussion of eliminating harmful races or resisting their invasions went well beyond the practical problem of sterilization — without apology or irony. Even Carter’s short chapter on “The work of useful bacteria” foregrounds bad bacteria: “food bacteriology … treats of the prevention of the contamination of foods by harmful bacteria. It also treats of the methods used to manufacture certain foods….” Capitalism put an additional premium on species. “Some patents are founded on the activities of bacteria and it is necessary to describe an organism carefully”, Tanner wrote: “Patent litigations may center about the bacterium involved.” There must be species; only if ownable things can be distinguished can there be property.37 These aspects cohere, if not without tensions, inconsistencies, or even contradictions, as a cultural system. Bacteriological analysis inhabited an image of nature, whose poles were malign entities and productive possibility, both of which might be regulated by identification and specification of the agents concerned. Overseeing all were the bacteriological sages, like the fictional Gottlieb, persons at once meditative, aesthetic, and “philosophical”, spiritual leaders imposing the zen of the bench: “Before the next lab hour I shall be glad if you will read Pater’s ‘Marius the Epicurean,’ to derife [sic] from it the calmness which iss [sic] the secret of laboratory skill”, Gottlieb tells his students. Fetishized technique could bypass moral questions; yield instrumental ends without appearing instrumental. Mediators of life and death, the bacteriological priesthood can dabble in the unthinkable and inexpressible. “The trypanosomes of sleeping sickness”, which young Arrowsmith’s prowess has earned him a right to study, cause “quite a nice disease”, Gottlieb explains. “In some villages in Africa, fifty per cent of the people have it, and it is invariably fatal.” In their well-bounded world, bacteriologists played God. Perhaps, muses Gottlieb, we medical bacteriologists should become philosophical biologists. “Why … destroy these amiable pathogenic germs?” Was it so obvious that we should protect “these oh, most unbeautiful young students … from the so elegantly functioning Bacillus typhosus with its lovely flagella?” Bacteria, after all, solved unemployment.38 Bacteriology’s cultural power extended beyond the laboratory. By book’s end, Gottlieb has pushed ahead with a controlled trial of a therapeutic and prophylactic phage in the midst of a plague epidemic, a project which epitomizes the colonizer’s power of life or death.39 Textbooks of course may not address such matters directly, and yet, as Fleck observes, the authority they project cuts through all questioning and doubt.40

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

278 ·

CHRISTOPHER HAMLIN

fixing bacterial subjects

I move now from out to in, from public representations of bacteriology to problems of its practice, following Pickstone’s fourfold way: classification (natural history), analysis, rationalization through synthetic means, and creation of a technoscientific narrative to order the enterprise and take care of residual anomalies. But each, it is worth reiterating, was implicated in the others. The Natural Historical Approach Textbook authors and historians alike have located the stabilization of its objects as classical bacteriology’s fundamental problem — stable facts, at some level, are a presumption of a science; they are both bearers of natural order and units of scientific communication. In botany and chemistry, the principles for identifying and cataloging species were familiar, if not wholly unproblematic. But on what basis did one declaim “Vibrio cholerae” and what exactly did one mean by it? Bacteriology kept threatening to become a science with no nouns, only verbs. Cohn had appreciated the problem, insisting that there were distinct species, and offering a classification admittedly provisional.41 But things had not gotten better. Classification was a domain of “great confusion”, noted Tanner — evident in “the number of classifications which have been published and the storm of discussion which each new one stimulates”. Ultimately he could only implore: “Biologists and bacteriologists should be interested in attempts to systematize knowledge; … should support any effort to organize their chosen fields.” While convention would have to be enough, the Zinsser authors acknowledged that no existing convention was “satisfactory”. Their detailed diagnosis makes the problem clear: A morphological system is not sufficient … because many organisms with totally different properties are identical morphologically. Biochemical, antigenic, and pathogenic properties have been drawn upon to provide the additional criteria needed for identification…. [M]odern systems of classification utilize the knowledge of morphology, staining reaction, physiology, metabolism, chemical composition, antigenic structure, pathogenicity, and virulence of the bacterium. Morphology still is accorded primary importance in the classification of large groups, but form, like the other criteria, fluctuates in “weight” in different categories. In some cases, … fermentative reactions are of minor importance…. In others … fermentative differences and antigenic structure are more important than morphology for differentiation .42 The taxonomic problem was viciously circular: a classification representing the relations of the several kinds of bacteria to one another required a concept of bacterial species. But that in turn required a classification of similarities and differences on the basis of some robust principle of prioritization. Even before Linnaeus, mode of reproduction had been a primary principle of botanical classification; but these fission

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

279

fungi did not (apparently) have sex, leaving them beyond the pale. The Zinsser authors lament that ambiguous status: “that bacterial species are reproducible signifies that a genetic mechanism must exist, although the lack of sexuality … suggests that the mechanism of inheritance is different from that of the higher plants and animals.”43 Thus there was no ready analogue to the branching tree or the periods of the periodic table. At best, bacteriologists might hope that the marshalling of characteristics would uncover a single hierarchy of distinctions. It did not. The regularly revised and argued-about classifications did not aspire to the status of a natural system, or even an artificial one. They were no single system at all, but at best a matrix of apparently unrelated properties that would allow a practical differentiation of kinds, so long as properties were to be stable (a dubious assumption).44 Such matrices would be the basis of an essentially analytical approach to taxonomy (a thing is the sum of its properties) that I shall consider below. Had such problems been raised in some Prolegomenon to a future science of bacteriology, the result might have been paralysis. But the inertia of practice revealed in the textbooks reflects the suppression of such concerns. Confidence came from chemistry.45 The chemists’ fundamental unit, the element, was a similarly operational entity. Chemists tolerated some anomalies: imprecision in atomic weights would be explained by discovery of unknown elements clinging to more common species, or later by isotopes. Analytical chemists pragmatically employed multiple axes of differentiation — one might want to know acidity or elemental composition; or by combustion distinguish inorganic salt-formers from organic matter; or by fractionation, isolate crystallizable organic substances; or discover fat content. Sometimes chemists were really interested only in one species: a pragmatic division might be “arsenic” and “not arsenic”. So, too, an exhaustive analysis was usually unimportant to the bacteriologists who served as a public health early warning system. Instead, their concern was limited to a few pathogens of interest.46 Evidently, the world’s business did not require well-warranted concepts of species or ultimate systems of classification; surely, bacteriologists too could simply get on with things. But again, in the face of a strong demand for authority, the profundity of the disanalogies could not fully register. In chemistry, the joint assumptions of the inviolability of elemental identity and the conservation of matter (as measured by mass) allowed a hypothetico-deductive combinatorial science to flourish.47 Elements could be recovered from compounds, then recombined in simple, whole-number ratios into new compounds. One could confidently deploy multiple schemes of analysis because one understood how they derived from a single system of elements. In bacteriology, however, no single set of operations commanded full confidence. A more specific basis for distinguishing bacteria was to know them by their works. The specificity of microbes might be discovered in (or defined as) the specificity of their effects — on the farm or in the body. Science could replicate craft. As Bruno Latour has elegantly pointed out, Pasteur’s success came from moving back and forth between laboratory and field or area of commerce. The foundational concept for equating microbe species with sauerkraut, beer, or disease was the process/metaphor/

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

280 ·

CHRISTOPHER HAMLIN

practice of fermentation, an ancient mode of chemical industry which had evolved into a broadly applicable mode of chemical change.48 Long before Pasteur, Francis Sylvius and Thomas Willis had made fermentation central in physiology and pathology; Pasteur, recognizing microbes as the agents of fermentations would extend that equation to the new bacteriology.49 One then worked backwards from the specificity of fermentation products to the specificity of ferments. One could even, like the English chemist Edward Frankland, go further. Interested in the chemical operations carried out by microbes in nature, Frankland proposed in 1885 that microbes be identified wholly in terms of the chemical operations they performed. Biology, in effect, would reduce to chemistry; Frankland was outlining a program of enzymatic specificity that would be the agenda for the first generation of biochemists.50 For reasons still unclear, Pasteur moved in the opposite direction, transferring fermentation from chemistry to biology. The assertion that each fermentation was the irreducible attribute of a unique life form would help to separate bacteriology from the new biochemistry while bringing little clarity to bacterial taxonomy.51 Nor was it ultimately a successful gambit. Some enzymes were widely shared. Unusual enzymatic capacities would take their place among species characteristics, but they were not necessarily a stable factor. The Zinsser authors noted the regularity (1:5000 cells) of a mutation in a glucose-limited form, allowing utilization of lactose.52 But disease was more important than vinegar. Remarkably, the erosion of confidence in enzymatic specificity did little to shake faith in the microbe-disease equation, the cornerstone of bacteriology’s authority. One may well ask why. Before 1850, the hypothesis of contagion was often used to account merely for the occurrence of disease, not for its particular clinical character (i.e., it belonged to etiology rather than diagnosis). There was therefore no necessary presumption of an equation of unique species with unique disease. As a matter of pathology, that association came mainly from Justus Liebig’s zymotic theory, developed around 1840, which attributed clinical distinctiveness to distinct fermentations. Pasteur’s subsequent attribution of distinct fermentations to distinct microbe species completed the link-up.53 In fact, however, post-Pasteurian practice was less a mapping of the unknown (microbe species) onto the known (varieties of disease), than a reciprocal redefinition of two dubiously distinguished domains in terms of one another.54 As it became habitual, this daring commerce of mutual definition no longer required the fragile bridge of zymotic theory. By 1876, Koch was employing distinctiveness and constant presence arguments in his anthrax work.55 If the two taxonomies (bacteria and diseases) were one, profound problems of species definition and of pathological specificity could be sidestepped. Ultimate explicability in either domain was of little practical importance so long as each corresponded to the other. Disease was analysed in terms of causative agents, agents specified in terms of the diseases they caused. The text book authors do not unambiguously endorse the clinical-microbial equation, but their admission of anomalies is perhaps indication of their general confidence in this remarkably attractive oversimplification. They understood that clinical

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

281

presentations of illnesses frequently did not translate into any clear microbe-defined diseases; also that infection by specific microbe did not translate into identical clinical conditions — and might not produce apparent disease at all.56 At the macroscopic level of clinical signs and symptoms, the human (or guinea pig) repertoire for manifesting infection was not especially large. Nor did it follow that any particular case of infectious disease need come from a single species: “either/or” could be “both/ and”. Recognizing synergy as the default state of natural bacterial operations, the Zinsser authors deplored “the trend … since Pasteur and Koch … toward the monoetiology of disease” as “too narrow, … there are many infections which result from the synergistic activities of several varieties of microbes”. Sequential infections might be an equally important feature of clinical disease.57 The Analytical Approach A plausible metaphor — disease as microbial fermentation — did not garner such power without a strong methodological foundation, one which was, at root, both analytic and synthetic. Two great working assumptions, exhaustibility and exclusivity (purity), underwrote the central practice of the bacteriologist, the making of cultures. That is, from some suspect sample brought into the laboratory, the bacteriologist should be able to find all forms present and isolate each to learn its powers. Both paralleled assumptions of chemical analysis: an elemental analysis gives us all the pure substances a sample contains. We will confirm purity through the sharp definition of each element; exhaustiveness via conservation of mass. For the approach to work in bacteriology one must assume that differential culturing can discriminate pre-existing, clear, distinct, and stable natural essences, and that a medium (available for our use) exists for growing each microbe. However familiar, the craft of culture making is a curious and complex epistemic practice. In Pasteurian terms, a wort or mash, kept under certain conditions, fostered growth of particular microbe(s) whose modes of living in that medium manifested mysteriously as the desired metabolites — beer, vinegar, or sauerkraut. Pasteur’s experience with “diseases of wine and beer” — where disease was defined as failure of the customary product to freely flow — reinforced recognition of the unique power of medium and conditions to ensure the flourishing of the unique creature whose telltale metabolic signature was the desired product. That each creature had its proper habitat, food, and role to play in nature’s economy was a longstanding trope of natural theology (and an intuitive lay articulation of the competitive exclusion principle). Assuming relatively free access of microbes to nutrient media (including tissues and fluids of the human body), it was tautological that types most suited to the conditions of growth would flourish over others less well suited. By a combination of re-culturing of the flourishing matter (more easily done with the solid media that Koch would pioneer) and refinement of culturing conditions to discourage unwanted hangers-on, one could, presumably, approach purity — make the sort of beer (or infection) one sought. Thus the consistently good lager became the model epistemic guarantor for an enormous terrain beyond (and “in vino veritas” understated: wine was truth).

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

282 ·

CHRISTOPHER HAMLIN

Differential culturing was an analytical operation in a broad sense, inasmuch as it allowed one to say what bugs a sample contained, but it also promulgated an analytical conception of bacteria as composites, reducible via the matrices of cultural (and other) characteristics noted above. In the sciences of matter in the nineteenth century, taxonomy had rested on the assumed adequacy of analytical operations: a natural history of material things was based on understanding them as composite. Some hoped to find a distinct chemical basis for each biological attribute of bacteria (though recognizing that each chemical constituent might be expressed in more than one way). Thus Ehrlich would famously link stain specificity to serological specificity: both presumably reflected the operation of surface receptors. While other characteristics too — like metabolism and cultural specificity — were presumably similarly determined, that presumed truth did not translate into a practical program of biochemical distinction, prior to maturation of molecular genetics in the 1990s. Generally bacteriologists resisted rather than welcomed biochemists.58 “It has been found”, Tanner laments, “that the chemical composition of bacteria cells does not vary greatly from that of other cells” (or, presumably, from one another).59 Hence the reliance on an empirical approach to differentiation. Build a matrix with enough dimensions and each bacterial species will have its own coordinates. Culturing was the main means of differentiation simply because it offered so much room for variation. In their 1930 Compilation of culture media for the cultivation of microorganisms, carried out at the behest of the Society of American Bacteriologists, Max Levine and H. W. Schoenlein listed 2543 media — the count did not include multiple recipes for preparation or physical conditions of use. Beyond morphology, basic habitat (aerobic-anaerobic), response to stains, enzyme repertoire, ad hoc chemical tests (e.g. the Voges-Proskauer test for Vibrio cholerae) could be deployed. A response to a single stain, the Gram test, took on great importance as the first division in bacteriology: but whether because it reflected real and basic divisions of biological kinds or because those kinds were defined in terms of it, was less clear. It had been used from the earliest days of bacteriology — it was, one may say, a Koch habit — in an enterprise that was essentially accretive.60 Thus, a species definition in an authoritative work such as Bergey’s manual of determinative bacteriology, a product of the same Society, would be a set of facts that together comprised a unique profile.61 On the basis of such catalogues of characters, one might claim that the microbes thus differentiated were essentially different. If the scheme made sense in principle, in practice it was asymptotic — the course of differentiation was always in progress; it was not clear how close it was to completion. One might hope that varieties of culture media could be made congruent to the array of microbe species and needs, but one could not claim exhaustiveness from trials with a finite number of media. One could not say that a particular set of discriminations demarcated a single species : it was possible, at least with regard to the media offered, that different forms might be equally happy eaters of whatever came their way. Repeatedly, new modes of differentiation led to one becoming many; odd “behaviours” of cultures presumed pure were later re-explained in terms of mixes of

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

283

kinds.62 Thus there was in bacteriology nothing akin to the regulative role that mass played in chemistry, as growing awareness of filterable viruses made abundantly clear. Nor could one be sure that differences were stable. “Few forms of life are sharply separated from other forms”, Tanner admitted. “Usually, intergrading forms exist which seem to connect them to other forms. This may be the result of evolutionary processes.”63 One did not have to worry about chemical species evolving . Anxiety that cultural distinctions were neither exclusive nor exhaustive, frustration that collections of characters yielded nothing like a Linnaean taxonomic system led to repeated overturnings of the dominant analytical approach. Morphological discrimination yielded to cultural, and, in the period of these texts, the latter was yielding to “physiological or functional characters”, notwithstanding that such an approach united “forms … quite unlike with respect to other characters well established in taxonomic work”. 64 Perhaps the best way to distinguish bacteria was in terms of pathology. “Subdivision into pathogenic, nonpathogenic, and the various intermediate varieties of microorganisms is a very convenient means of classification”, noted the Zinsser authors, even though many strains of common pathogens were not virulent. Serological specificity held out the greatest hope for these authors. Evidently, animal bodies sensed differences far more precisely than chemists and bacteriologists could. “Serologic methods often are so delicate that they can be used to identify and detect variations of bacteria which appear to be identical both in form and in physiologic activity”, noted the Zinsser authors.65 That serological classifications often did not cohere with other traditional classifications; that serological properties might not always be stable did not matter overmuch. Indeed, the Zinsser authors saw the new serological age as restoration of “a kind of modified monomorphism … bacteria do not change from cocci to bacilli or mutate from genera to genera but profound changes in biological activity, antigenicity, and virulence occur within the framework of the species themselves”. That a changeable entity could underwrite a renewed faith in stability suggests how desperate things had become.66 But here too was an iteration of reciprocal definition. Serological response would displace clinical manifestation and microbial presence alike. Appraisal of microbe (as suite of antigens) by body might offer precise distinctions, but surely it also privileged a “paranoid style”, by defining identity in terms of what another organism feared an entity might do to it. Bodies might be fooled: antigens wrongly read resulted in allergic reactions. Interestingly, the texts do cover hyper-sensitivity (though not auto-immune diseases). The Rationalized Production Approach Just as developing a natural history of bacteria involved analysis — the translation of a sample into an array of microbe types, each defined as a composite of characters — it also involved a rationalization of the fermenter’s craft, the multiplying under ideal conditions of one or a few cells to create new artificial array of “pure” cultures. With classification tied so closely to characteristics in media, purity in turn often was

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

284 ·

CHRISTOPHER HAMLIN

measured in terms of performance: bacteria became what they could be made to do. The product of pure cultures in turn became the basis of other products. Some were physical and directly saleable: serological tests, vaccines, and antitoxins. Others — e.g., instruction, public health advising — were symbolic or indirectly saleable. Again, chemistry was the model. Chemists unproblematically took messy natural things apart and recombined the parts into useful goods and forms of authority. Transplantation of ways of knowing and doing was, again, both compelling and problematic. Like many operations in chemistry, culturing transcends what is often presented as a sharp division between the made and the found, between “natural history” (a form of science) and technics. Culturing epitomizes knowing as doing — by privileging products of metabolic and, particularly, enzymatic, activity, early culturing fostered a natural history founded on enzymatic specificity, which, early on, had seemed to cohere with other factors to allow an easy means of natural classification. Culturing had been industrial craft long before becoming tool of science, as central a biotic conversion technology as was selecting seed or any mode of processing harvest.67 Unique to bacteriology were the multiple, and potentially incompatible uses of culturing. It was a threefold enterprise: a way to catch and thus reveal a natural entity; to magnify it for study and production; and to transform it. The last, intended or not, was a consequence of altered environment. Discrimination of natural kinds assumed transparency and stability; production of artifacts assumed transformability. For the Petri dish did not merely fatten or force fission. “It has been shown that … [bacterial] composition, physical properties, metabolism, growth, death and morphology are influenced by chemical substances in the natural or artificial media in which bacteria occur”, note the Zinsser authors. “This influence is so profound and widespread that it is, in fact, impossible to deal adequately with any property of bacteria without taking into consideration, directly or indirectly, the factors of chemical environment.”68 Not only did the microbe find itself isolated from other species, in an uncommonly concentrated nutrient medium, under specific conditions of temperature, oxygen, etc., but in an environment of its own (increasingly concentrated) wastes. The dimensions of change were many. Beginning 312-page review article on dissociation (one central mode of variation), Philip Hadley offered a succinct list: Cocci become rods and rods cocci or spirals; forms of growth change overnight; motility is lost and regained; fermentation reactions are modified by time and opportunity; spore formers become sporeless; hemolytic activity comes and goes; capsulated bacteria lose their capsules, and capsules are gained by noncapsulated forms; antigenic power vanishes and reappears; cultues become spontaneously agglutinative or fail of aggultination, virulent cultures become harmless and harmless cultures virulent.69 Isolation too, might represent perversion rather than purification. The re-cultivation generation after generation in these unnatural media, undertaken to secure purity, was, in effect, imprisonment in solitary. It treated identity as alienation: a species is truest when longest alone. That assumption worked in chemistry; not so clearly with

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

285

bacteria. Isolated chemical substances sat happily on laboratory shelves. Though most downplayed its seriousness, a problem was plain. It was widely recognized that culturing changed the cultured entity; the conclusion that such alteration undermined the applicability of the truths of the culture plate to the world beyond was generally resisted. For there was no alternative. However long their lists of the changes that may occur, authors stop well short of Jakob Moleschott’s materialist dictum: “you are what you eat.” Instead, the recourse was to a loosely Lamarckian admission of the power of environment to modify existing form over time — a less brazen view of the malleability of identity. Pushing further risked crossing from practitioner to philosopher — what Nägeli and Hallier were accused of. A more radical stance would be incompatible with the sine qua non of replicability; also, it seemed, with production — whether of vaccines or authority. You could not study bacteria without isolating them and multiplying them; Nature’s promiscuous mixes of bacteria were effectively beyond science. The gap between laboratory and field was a trope of bacteriology. Often bacteriologists seem ambivalent to the question “Which is truer, laboratory or field?” Field failures of vaccines which had excelled in vitro need not imply poor laboratory work — perhaps it was nature, or test subjects, whose performance was wanting. Recognition of variability problematized normality. Often the hope was to represent variability from stability, understood as the natural, or even essential state. But the power of the medium to change the mediated threatened any such concept. “As do all forms of life, bacteria vary in both shape and function”, the Zinsser authors wrote: the many terms used for describing different forms were relative to a so-called “normal form, designation of which may be quite arbitrary”. Regularity itself might be artifactual — “the apparent stability of our standard strains is enforced by specific, artificially produced, environmental and cultural conditions”.70 We might know bacteria only in terms of what they did, but they could be made to do many things. Puzzling over the general problem of which medium would best allow a particular microbe to flourish, the Zinsser authors acknowledged that the problem could not be answered until one translated “flourishing” into a mode of production. What did one want it do? “While it is customary to speak of an optimum temperature for bacterial growth, it is necessary, in the final analysis, to define the phase or characteristic of bacterial activity to which the term is applied”, they wrote. “The optimum temperature for rapid multiplication is not always the optimum temperature for the greatest crop yield, because injurious waste products accumulate most rapidly at this temperature…. The optimum temperatures of fermentative, proteolytic, and synthetic processes are not always the same as those of growth and differ among themselves. The optimum temperatures for constancy of form in relation to phases of the growth curve vary.” The ironic result was that any natural had to be defined in terms of an artificial.71 As Mendelsohn has shown, variation of virulence, either to attenuate or to accentuate, was a key locus of that ambiguity. Variable virulence undercut monomorphism (to the extent that species were defined by their pathogenicity), and thus threatened the simple equation of microbe with disease. At the same time, it represented a

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

286 ·

CHRISTOPHER HAMLIN

powerful technology (that relied on that equation).72 A common resolution was to imagine virulence as stable in nature; modifiable only in the Promethean laboratory. A consequence of the dominance of the context of production was the discomfort with concepts of adaptive pathogenicity, and general lack of interest in microbial ecology.73 The qualities bacteria exhibited in the field, where they interacted with other species and responded to physical and chemical environment could not readily be summed. Phenomena later explained by transduction or complex processes of gene expression were sometimes encountered, but there was no institutional frame for responding to them, or, in many cases, even acknowledging them. The simple equation of agent and product ruled, because the product, be it a human disease or a human fermentation, remained the key subject. The Narrative Approach: Forced Stories Bacteriology was an ordering endeavour, a search for simplicity behind chaos; the bacteriologist’s stock rose as the bacteria’s fell.74 The telling of stories is Pickstone’s first way of knowing; it is last here because it was the most comprehensive recourse to bacterial unruliness — “when all else fails, make up a story”. The stories are less theories than taming analogies which transform aberration into order. Broadest was the overlay of chemistry. Chemical species were simple and passive; bacteria, being small, should be equally so. But bacteria regularly exhibited what appeared to be spontaneous (i.e. uncaused) variation, inviting the term “behaviour” — usually used ironically. Over time, cultures varied. As Carter told the nurses, “bacteria of the same species”, even under “identical conditions”, were not always “exactly alike”. Only sometimes was this variation traceable to conditions of culturing; “it may be due to factors inherent in the bacteria themselves”.75 The first response to apparently spontaneous variation was to assume laboratory error. As well as accusation and apology, admissions of cultural contamination often convey relief. But analogy could also serve as anodyne. Gradual directional change was easiest to accommodate; the gaining of new characteristics through acculturation to an artificial medium was known as the “training” of bacteria, an overtly Lamarckian concept.76 The obverse to such tutelary “training” was to see change as deterioration. Even if the kitty litter had been regularly changed, an existence of many generations away from nature or without a revivifying trip through an animal immune system, might account for degenerative change. Closely related were “cyclogenetic” or life course models — sequential change of form might not be induced but inherent. This trope domesticated bacteria; it drove off the joint imputations of determinism and of randomness (a Darwinian selection of random variation threatened monomorphism; major [de Vries-style] mutations were less problematic so long as they occurred with regularity).77 “Instead of a mobile field, wherein the transformations were determined by chance, we are perhaps approaching a belief that these changes follow one another in orderly manner”, noted Tanner. But biological orthodoxy remained bacteriological heterodoxy. “The life-cycle hypothesis

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

287

for bacteria has not been generally accepted, despite the fact that as we come up the evolutionary line of living beings to forms which are easily studied, even certain microscopic forms, complicated life cycles are recognized.”78 More troublesome were sudden changes. Again, denial often seemed the best option. Since there could be no uncaused change, one was not seeing change. Perhaps a hitherto hidden alien strain had emerged at a late cultural stage, disclosing the fiction of peaceful racial purity. But, as with cyclogenicity, merely to give a name to a phenomenal change, even if its cause remained inexplicable, was to bring it within the realm of moral order. Earlier naturalists had referred to “sports” of nature. Bacteriologists tamed change with “involution” and “dissociation”. Neither was a well-defined concept, but each was more than a neutral descriptor. Bacteria sometimes took on “irregular and bizarre shapes, stain irregularly, … [are] swollen, shrunken, or granular”. This is involution, declares Carter. Calling involutes “deformed” and “aberrant”, Tanner stressed that other life forms exhibited analogous variation. Involution changed, rather than impaired function — it was a concept more of perversion than pathology per se.79 Some, like Carter, would confine the phenomenon to old cultures, seeing it as bacterial senility. The involute thus became the microbial equivalent of a dirty old man.80 But others protested both the accuracy of the relegation and the implicit value judgement. “Involution”, Tanner recognized, was simply “a convenient refuge for explanation of forms which differed from what was accepted as the normal”. But similar abnormalities occurred in “fresh young cultures”. Some species (e.g., Azotobacter) were particularly prone: “it is sometimes almost impossible to secure cultures free from those abnormal, aberrant forms commonly referred to as involution forms. Under such optimum conditions, is it right to consider them as decrepit, diseased, or weakened forms?”, Tanner asks but does not answer.81 Worse was “dissociation”. A colony, ostensibly the product of a single cell, suddenly shape-shifted into two (or more) radically different forms — known (without allusions to Jacob and Esau), as “rough” and “smooth” forms. The differences were more than morphological; the colonies, which bred true after the schism, differed serologically and in virulence (the smooth were more virulent). Surely, “dissociation” too, had overtones — during this golden age of Freud, it put the “schizo” back in Schizomycetes. The phenomenon was treated as mutation, but still baffled. It occurred relatively regularly in many bacteria. As with involution, the magnitude of change exceeded normal inter-species differences. Dissociation undermined faith that capacities like pathogenicity could adequately be understood in terms of fixed types.82 Even more jarring was an eerie apprehension of individuality among bacterial cells. An atom possessed nothing approaching individual uniqueness — interchangeability was a foundation of chemical knowledge. If all cells of a particular species were merely chemical, all ought all to die or be neutralized under identical disinfectant conditions. But they did not. Disinfection was “gradual … a minority will survive very much longer than the majority”. Thus, for each so-called species, one had to find a neutralization curve for every disinfective regime.83 Reliance on the laws of

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

288 ·

CHRISTOPHER HAMLIN

large numbers or empirical rates of reaction could effectively deflect attention from individual cells and even from species. In noting that “the analogy between disinfection and a chemical reaction indicates the fundamental nature of the process”, the Zinsser authors were modelling bacteriology on not organic chemistry, but the new positivistic and stochastic physical chemistry. The frequency, magnitude, and persistence of variability can seem in retrospect incompatible with the premises of monomorphism. Yet textbook writers manage to frame variability as part of business as usual; accommodation is refinement not refutation. Acknowledging the pleomorphist margins reified their marginality, making clearer the coordinates of the centre.84 CONCLUSION

Surely bacteriologists recognized the weaknesses in their ways of knowing — indeed, in these textbooks they broadcast them to novices. To a critical reviewer, this pastiche of multiple ways of knowing can seem disturbingly arbitrary — unable to stabilize its fundamental units, late Kochianism is a hollow institution, ripe for revolution, clinging to dogmas daily violated, repeating the liturgy of the pure culture, circumventing the anomalous and inexplicable with empty verbalisms. And yet, as Brock notes, “although this rigid viewpoint of the constancy of species was wrong, it was in the early days of bacteriology an idea that was essential, because only in that way could sense be made out of the research”.85 To focus only on epistemic risks fails to recognize both technoscientific stakes and limited alternatives. Those become clearer when we turn from a Kuhnian heritage to the Pickstonian framework. Kuhnian revolutions are chiefly intradisciplinary epistemic events. From the prospect of the present we see paradigm succeed paradigm, each comprehending its predecessor’s achievements but also some of its anomalies. Pickstone avoids that implicit teleology. Kuhnian scientific revolutions are in idealized disciplines — astronomy, chemistry, or modern physics — which are embodied more or less adequately in real institutions. Pickstonian ways of knowing/doing, by contrast, are effectively trans- or non-disciplinary, even if they may be particularly well expressed within disciplines at particular times and places. There is no Platonic bacteriology to be institutionalized; the bacteria themselves come with no instructions for how to do bacteriology; discipline builders must scrounge methods, techniques, concepts, and marketing practices. No surprise then, that they produce a motley — traditions from botany, many (and varied) importations from chemistry, bits of medical theory and physiology (fermentation), out-dated biology (Lamarckianism), metaphors from psychology (involution), even, possibly demonology (though I think Fleck overstates).86 Moreover, Pickstone suggests, these elements are no more inherent to their source disciplines: confronted with any new domain of entities with social-technical-medical significance, one seeks some system of natural kinds. This involves a natural historical-analytical representation of each kind as a composite of properties; also a rationalization operation to enhance regularity (kinds that do not come naturally must be made so); and recourse to theoretical and conceptual language to order ­complexity.

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

289

It represents also a plausible response to high stakes issues — especially clear in medical matters. If diseases are singular events attributable to single causes, magic bullets may suffice. Isolation and control of responsible agents is a good first-order problem-solving strategy; overreacting to variability will undermine medical and public health applicability. A standing back from the quality of the knowledge may also help us better know the knowers. The freeze-frame of “paradigm” fails to catch the anxiety of science in process, the hope that the epistemic duct tape can hold another day.87 We get a better sense of the difficulty of creating a new science, the power of habitual ways of thinking/doing, even when they are not working perfectly, and the risks of worrying overly much about (transitory?) anomalies. A Pickstonian approach clarifies Fleck’s ambivalence to his own science. Fleck complains of bacteriologists’ intransigent assumptions, protests the pretension of bacteriology’s persona, yet suggests that its methods are usually good enough. Greater rigour, coherence, justness, efficiency, or accuracy may be possible, but they have significant opportunity costs; for the purposes at hand, the juryrig may be good enough. postscript

A final way of representing the unique assembly of ways of knowing and doing of the bacteriology of the 1930s and 1940s is to contrast it with what followed. The renaming in 1961 of The Society of American Bacteriologists as The American Society of Microbiology represents a broadening, an explicit integration with biology, and reflects the arrival of new contexts of working and knowing. To biochemists and geneticists bacteria were of interest not in and of themselves but for the fundamental processes they housed. Rapidly dividing bacterial cells proved a good experimental system for geneticists, though they often emphasized genomes and mutations over complexities of gene expression. Biochemists often took the opposite tack.88 Molecular genetics, and, in particular, cheap and quick gene-sequencing, seemed like the analytical end game. Finally, it seemed, chemistry had delivered: polymerase chain reaction would finally sort out the vexing problem of bacterial identity. But these techniques too disclosed variability; processes of horizontal gene transfer, curiosities in the late 1940s, became important.89 It seemed too that both uptake of genetic material and the variability of gene expression were not random, but interestingly adaptive — in Frank Schätzing’s eco-thriller, The swarm (2004), algae use gene transfer to wrest control of the world. Just as the study of bacteria was no longer compartmentalized, set off from biology generally, it was less restricted to manipulation of laboratory entities. A bevy of new analytical techniques has allowed the tracking and mapping of codons of interest, facilitating research in bacterial ecology. Practices of isolation and assumptions of stability remain useful, but the former is less an end in itself, and the latter belongs to means of working rather than to dicta about nature.90 The shifts also reflected changed institutions of technoscientific work — greater ease of world travel, a more global distribution of expertise, willingness of research

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

290 ·

CHRISTOPHER HAMLIN

funders to meet the expense of field work. As bacteriology ceased to be at the forefront of public health (giving way to viruses, carcinogens, and lifestyle factors), they became part of broader modes of biological production. While a burgeoning biotechnology industry would continue to keep an eye on the synthetic powers of particular species, their activities in concert with other species were increasingly important in the maintenance of what would be called “ecosystem services”. Bacteriology was also implicated in changing sensibilities — an unease with laboratory ways of knowing expressed across the spectrum of the life sciences. Conventional zoos no less than pure cultures were challenged on epistemic grounds; simplification and control were distortion. That distrust had conceptual and ethical implications. It involved reassessment of distinctions between studier and studied, human and non-human, that paralleled (and reflected?) the rise of feminist philosophy of science.91 I wrote above that Schätzing’s algae “use” horizontal gene transfer. With greater recognition of the complexity of bacterial lives, bacteria ceased to be mere determinate objects, passive and dumb. In earlier decades such anthropomorphic allusions would have clearly placed a text on the popular side of a science-popularization divide (though less so, were one writing of the malignity of pathogens). These new model bacteria, on the other hand, seemed quasi-cognitive, almost to be thinking and choosing in responding to (and shaping) their environments. With cognitive science models comprehending logic, machines, minds, and brains, it became harder to justify subject–object antitheses, in which bacterial objects required to be represented in wholly different terms than their human observer-creators. No longer was bacterial “behaviour” overtly ironic. Unwilling to objectify ourselves, it seemed increasingly anomalous to reduce bacteria to some essential racial identity. If anything, they were more malleable than we: the gene swapping that went on in their wet places — soils, ponds, or seas — made these a big bathtub of Giddensian role-playing, information-exchange, or, depending on one’s lingering loyalty to natural kinds, polymorphous perversion.92 None of us, after all can master facultative anaerobism, even though there are occasions when it might be useful. Shifts in representation did not require new discovery. What had been called “training” — acclimatization to a Petri-dish prison, might equally have been “learning”. Many borrowings were now from the human sciences and the helping professions — psychiatry, sociology, and social psychology. Symmetry of explanatory language brought symmetry in moral assessment. Even pathogens benefited. Merely a curiousity in the bacteriology of the 1930s and 1940s, Theobald Smith’s agenda of evolutionary epidemiology became dominant. Pathogens were now legitimate life forms; disease an incidental (and, in the long durée) transitory aspect of their association with humans. Bacteria were important members of biotic communities and communities were good. Even the cholera vibrio was a “benign” resident of estuaries and good citizen among the zooplankton.93 To represent this post-modern bacterial technoscience as a new disciplinary paradigm in bacteriology would be both limiting and misleading. It remains an ­assemblage

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

291

of ways of knowing and doing. With identity more fluid and complexity more conspicuous than simplicity, taxonomy has receded. Analysis, in terms of genetic units, remains important and significant as a foundation of commercial synthesis. The stories we tell are very different, however. Bacteria are less exclusively the agents of events in which they are implicated, like cases of disease; rather they are tragic figures caught up in emergent composite events, which displace the entities involved in them. Enigma replaces clarity; interaction has become more interesting (and, in some sense, more truth-like) than isolation; variability has switched from noise to signal. ACKNOWLEDGEMENTS

My thanks to Mick Worboys, John Pickstone, Ian Burney, and others at Manchester University workshop, June 2010. Also to the late Harry Marks, who commented on an earlier draft, and was an ever-critical and inspiring friend and colleague. I also wish to remember the work of the late Olga Amsterdamska. As I reflected on bacteriology texts, I found myself at every stage illuminated by her seminal articles in the 1980s and 1990s. REFERENCES 1.  I have not used “paradigms” or Fleck’s “thought collectives” here to deflect attention from the primacy of the cognitive. Instead, warranting the pun is Clifford Geertz’s 1964 essay, “Ideology as a cultural system”, in D. Apter (ed.), Ideology and discontent (New York, 1964), 47–76. Rejecting the view that ideologies were merely distortions, Geertz emphasizes their utility as simplifying systems in efficiently mediating the practical work of communication, allocation, categorization, and public action generally. Geertz’s concept, like Herbert Simon’s “satisficing” in economics, recognized the impracticality of being robustly rigorous in all directions simultaneously. Cf. Ludwik Fleck, Genesis and development of a scientific fact, ed. by T. J. Trenn and R. J. Merton, transl. by Fred Bradley and T. J. Trenn (Chicago, 1979); Herbert Simon, The sciences of the artificial (Cambridge, MA, 1969). Similar themes arose in Thomas Kuhn’s paradigms about the same time, though the emphasis was on cognitive matters (Thomas Kuhn, The structure of scientific revolutions (Chicago, 1962)). Geertz’s essay was the jumping off point for a descriptive approach to technoscientific ideologies which I developed with Philip Shepard (Deep disagreement in U.S. agriculture: Making sense of policy conflict (Boulder, CO, 1993)). 2.  This has begun to change. See Andrew Mendelsohn, “Cultures of bacteriology: Formation and transformation of a science in France and Germany, 1870–1914”, unpublished Ph.D. dissertation, Princeton University, 1996; Christoph Gradmann, Laboratory disease: Robert Koch’s medical bacteriology, transl. by E. Forster (Baltimore, 2009). 3.  Olga Amsterdamska, “Medical and biological constraints: Early research on variation in bacteriology”, Social studies of science, xvii (1987), 657–87. 4.  John Pickstone, Ways of knowing: A new history of science, technology, and medicine (Manchester, 2000); idem, “Working knowledges before and after circa 1800: Practices and disciplines in the history of science, technology and medicine”, Isis, xcviii (2007), 489–516; idem , “Commentary: From history of medicine to a general history of ‘Working Knowledges’”, International journal of epidemiology, xxxviii (2009), 646–9. 5.  Keith Vernon, “Pus, sewage, beer, and milk: Microbiology in Britain, 1870–1940”, History of science, xxviii (1990), 289–325; Daniel Schneider, Hybrid nature: Sewage treatment and the

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

292 ·

CHRISTOPHER HAMLIN

contradictions of the industrial ecosystem (Cambridge, MA, forthcoming). 6.  A popular Soviet history of bacteriology reflects this orientation (see Lev Potkov, A world we do not see (Moscow, 1957)). 7.  Fleck recognized the unaccountable character of such changes, using the image of the “vanguard” of an army as his chief metaphor. Genesis and development (ref. 1), 124. 8.  William Bulloch, The history of bacteriology (Oxford, 1930). Terms like ‘dogma’, ‘orthodox’, and ‘heterodox’ arise regularly in the history of bacteriology; see Mendelsohn’s objections: J. Andrew Mendelsohn, “‘Like all that lives’: Biology, medicine and bacteria in the age of Pasteur and Koch”, History and philosophy of the life sciences, xxiv (2002), 3–36, see pp. 5, 19. 9.  Pauline Mazumdar, Species and specificity (Cambridge, 1995). 10.  Robert Kohler, “Innovation in normal science: Bacterial physiology”, Isis, lxxvi (1985), 162–81; idem, “Bacterial physiology: The medical context”, Bulletin of the history of medicine, lix (1985), 54–74; Amsterdamska, “Medical and biological” (ref. 3); idem, “Stabilizing instability: The controversy over cyclogenic theories of bacterial variation during the interwar period”, Journal of the history of biology, xxiv (1991), 191–222; idem, “Achieving disbelief: Thought styles, microbial variation, and American and British epidemiology, 1900–1940”, Studies in the history and philosophy of biology and the biomedical sciences, xxxv (2004), 483–507; Thomas Brock, The emergence of bacterial genetics (Cold Spring Harbor, 1990); William Summers, “From culture as organism to organism as cell: Historical origins of bacterial genetics”, Journal of the history of biology, xxiv (1991), 171–90; Mendelsohn, “‘Like all that lives’” (ref. 8); Ilana Löwy, “‘A river that is cutting its own bed’: The serology of syphilis between laboratory, society, and the law”, Studies in the history and philosophy of biology and the biomedical sciences, xxxv (2004), 509–24; Warwick Anderson, “Natural histories of infectious disease: Ecological vision in twentieth-century biomedical science”, Osiris, 2nd ser., xix (2004), 39–61. 11.  The resurgence of interest in Fleck’s work, following Bradley and Trenn’s 1979 English translation, affected scholarship less than one might have expected. Generally, Fleck has seemed more relevant as a precursor of Kuhn and of SSK than as an historian of his own science. 12.  Imre Lakatos, “Falsification and the methodology of scientific research programmes”, in Lakatos and Alan Musgrave (eds), Criticism and the growth of knowledge (Cambridge, 1970), 91–196. Amsterdamska complained of the constraints of Kuhnianism: “Although many sociologists and historians of science argue that different scientific fields exhibit different patterns of cognitive development, the philosophical vocabulary used to describe cognitive structure has been based almost exclusively on the ideal of fundamental, theory-building sciences. Whether scientific development is considered a result of the hypothetico-deductive process of induction, or whether it is seen in terms of paradigms, progressive research programmes or problem-solving traditions, the focus has been on the construction of theory and the solution of theory-generated problems” (Amsterdamska, “Medical and biological” (ref. 3), 657; see also 659–60; idem, “Achieving disbelief” (ref. 10); Brock, Bacterial genetics (ref. 10)). 13.  David Smith, Donald Martin, Norman Conant, Joseph Beard, Grant Taylor, Henry Kohn, and Mary Poston (eds), Zinsser’s Textbook of bacteriology, 9th edn (New York, 1948; hereafter cited as Zinsser, 9th edn); Fred Tanner, Bacteriology (New York, 1937); Charles Carter, Microbiology and pathology, 3rd edn (St Louis, 1944). I have consulted also the 1939, 8th edn, of the Zinsser text, the last to list Zinsser as author (with Stanhope Bayne-Jones (New York, 1939)). Comparison of the two reinforces my decision to refer to the “Zinsser authors” rather than Zinsser. While the degree of difference varies from topic to topic, the changes are often toward a toning down of Zinsser’s commentaries and speculations. A diachronic review of the evolution of this textbook would be valuable, but my interest here is in a snapshot of a mature science. 14.  Tanner, Bacteriology (ref. 13), pp. vii–viii; Carter, Microbiology (ref. 13), 165. Brock was blunt: “bacteriologists immersed in their immensely technical and highly specialized studies, cared

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

293

little about broader questions in biology (Bacterial genetics (ref. 10), 2). In criticizing this view, Mendelsohn, “‘Like all that lives’” (ref. 8) has argued that biological issues were more central than is usually recognized, but that they existed within a different conceptual regime, that of variation of virulence. “For this most medical and applied science has always seemed separate from the conceptual development of biology … and even epistemically incapable of contributing to it …, following my argument, bacteriology became experimental biology before biology itself is supposed to have become experimental” (p. 6). The acknowledgement of uncertainty and flux in the textbooks reflects the issues to which Mendelsohn calls attention; the great question then, is how that recognition of complexity was kept from undermining a stable basis for practice, the central issue for Amsterdamska (“Medical and biological” (ref. 2)). 15.  Tanner, Bacteriology (ref. 13), 28. 16.  Bulloch, Bacteriology (ref. 8). Cf. Mazumdar, Species (ref. 9), esp. pp. 44–5. Tanner, Bacteriology (ref. 13), 57–8; Amsterdamska, “Medical and biological” (ref. 3), 670; Mendelsohn, “‘Like all that lives’” (ref. 8). 17.  “He [Koch] was aware that botanists did not regard different bacterial forms as distinct and constant species, and he agreed with Brefeld that the formation of distinct species can be justified only ‘when the whole history of development has been traced by cultivation from spore to spore’. Nevertheless, he argued that this ‘theoretically correct demand cannot be made a sine qua non in every investigation of pathogenic bacteria. We should otherwise be compelled to cease our investigation into the etiology of infective diseases till botanists have succeeded in finding out the different species of bacteria by cultivation and development from spore to spore. It might then very easily happen that the endless trouble of pure cultivation would be expended on some form of bacterium which would finally turn out to be scarcely worthy of attention’” (Amsterdamska, “Medical and biological” (ref. 3), 666–7). 18.  “There is nothing medical in this belief in monomorphism; but, in the context of Koch’s research programme, monomorphism, formulated so as to exclude the possibility of morphological or physiological variation in bacteria, was firmly embedded in the germ theory of disease and had clear medical ramifications” (Amsterdamska, “Medical and biological” (ref. 3), 663). 19.  Zinsser, 9th edn (ref. 13), 230. Among those who would criticize this view, Fleck, (Genesis (ref. 1), 59, 117), attributed the origin of belief in microbial monocausal explanation to a pre-scientific age in which disease was seen as invasion by minute devils. 20.  Carter, Microbiology (ref. 13), 165. 21.  Carter, Microbiology (ref. 13), 171–2; Tanner, Bacteriology (ref. 13), 32. 22.  Amsterdamska , “Stabilizing” (ref. 10); Kohler, “Innovation” (ref. 10), 171. 23.  Associated with the work of Paul Ewald, this issue is sometimes presented as new. It has received great popular attention in recent decades in connection with the origin of HIV and the threat of new forms of flu: see Ilana Löwy, “Epidemics and populations”, Studies in the history and philosophy of biology and the biomedical sciences, xxxiii (2002), 187–94. 24.  Zinsser, 9th edn (ref. 13), 112, 114. The authors recognize that bacteria may even “have been active agents in the natural selection of man and animals and plants”. 25.  “… [D]isregard for variation was motivated not only by various methodological objections (which could always be invoked against work which did not seem to fit with the assumption of monomorphism) but also by the belief that such variations were practically irrelevant” (Amsterdamska, “Medical and biological” (ref. 3), 670). 26.  Mendelsohn, “Like all that lives” (ref. 8), 4–5, points to the emergence of “virulence” as an effectively irreducible organizing entity that did not require translation into pathology: “… at the center of this reorganization of bacteriological knowledge lay neither a theory nor a phenomenon clearly characterized according to one or more branches of science, but ‘virulence’. Virulence was an intellectually empty, almost purely operational concept: virulent cultures killed, attenuated ones

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

294 ·

CHRISTOPHER HAMLIN

did not.” 27.  Zinsser, 9th edn (ref. 13), 120. Mazumdar notes: “In Koch’s experiments, the body of the animal was itself an apparatus for producing a pure culture of the organism, and at the same time and indicator of the properties of the species cultured: the disease was one of the differentia of the species, as constant as its length and breadth” (Species (ref. 9), 66). 28.  Kohler, through his explorations of the emergence of biochemistry, has recognized the conflicts: “Chemists came to the study of microbes with a rather simplistic view of living organisms. Many medical chemists believed that microbes were essentially chemical catalysts and that the dramatic effects of infections were caused by simple chemical toxins produced by bacterial reactions.” Kohler notes the complaint of the maverick biochemist and pioneer in bacterial physiology, Marjory Stephenson: “chemists conceived of bacteria as convenient bags of enzymes to be extracted, separated, and examined in vitro” (Kohler, “Innovation” (ref. 10), 164, 166–7). 29.  Careers too, most conspicuously that of the organic chemist-bacteriologist Paul Ehrlich, exhibit the close ties. Ironically, it is the chemist Pasteur who liberates bacteriology from chemistry; the botanist-physician Koch who most closely models bacteriology on chemistry. As Mazumdar, in Species (ref. 9), 61, points out, the central role of staining bacteria with the new organic dyes arose out of the close personal and familial relations between Ehrlich, his cousin the organic chemist Carl Weigert, the pathologist Julius Cohnheim, the botanist Cohn, and the young Koch during years when all were working in Breslau. 30.  Sinclair Lewis, Arrowsmith (New York, 1925), chap. 4, section IV. 31.  Vernon, “Pus” (ref. 5); Russell Maulitz, “Pathology”, in The education of American physicians, ed. by Ronald Numbers (Berkeley, 1980), 138–9. 32.  Kohler, “Bacterial physiology” (ref. 10); Amsterdamska, “Medical and biological” (ref. 3), 671. 33.  Amsterdamska, “Medical and biological” (ref. 3), 667. 34.  Peter Baldwin, Contagion and the State in Europe (Cambridge, 1999); Löwy, “River” (ref. 10). 35.  Zinsser, 9th edn (ref. 13), 230, italics mine. 36.  Summers, “Culture as organism” (ref. 10), notes that the concentration on colonies as units hampered recognition of statistics — recognition that within a colony there would be a normal distribution of variability. 37.  Carter, Microbiology (ref. 13), 165, italics mine; Tanner, Bacteriology (ref. 13), 83–4. See also Schneider, Hybrid nature (ref. 5). 38.  Lewis, Arrowsmith (ref. 30), chap. 4, section IV. 39.  In the 1930 film of the book, the ethics are highlighted beyond their status in Lewis’s narrative: while the tone is ambivalent, much is made of the colonial governor’s view that subjecting the population to an experiment of this source is a moral outrage. This long predates both Tuskegee and the reaction to it. 40.  Fleck, Genesis (ref. 1), 55–65, 115. 41.  Mazumdar, in Species (ref. 9), 52, 57, notes: “It is very difficult, on reading Cohn’s paper, to disentangle from the web of relationships anything resembling either a dichotomous or a common ancestor classification.” Cf. Bulloch, Bacteriology (ref. 8), 193–4, 203: “It is true that some of the questions which he [Cohn?] raised are still in the melting-pot, and nowhere is this more apparent than with regard to the classification of bacteria.…” 42.  Tanner, Bacteriology (ref. 13), 83; Zinsser, 9th edn (ref. 13), 135–6. 43.  Zinsser, 9th edn (ref. 13), 59. See also Brock, Bacterial genetics (ref. 10); Summers, “Culture as organism” (ref. 10). 44.  Effectively, the bacteriologists were practising a competitive exclusion principle long before it was articulated by the ecologist Gause (working with microorganisms) in the early 1930s. That is, they were defining species in terms of niche and niche in terms of species. Since no two

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

295

species may occupy the same niche, one must either complicate (subdivide) niche or merge the apparently different species when this appears to be the case. Gause’s apparent independence from institutional bacteriology is another example of the isolation of bacteriological ways of working and knowing from bacteriology at large. 45.  “Chemists use periodic tables, and group closely related elements. They make such groups as halogens, alkali metals, heavy metals, rare earths, etc.” (Tanner, Bacteriology (ref. 13), 84). 46.  “In making cultures the medium most likely to grow the suspected organism or organisms and, in addition, the medium most likely to grow any organism that might be present should be chosen” (Carter, Microbiology (ref. 13), 124; cf. Fleck, Genesis (ref. 1), 113–14). 47.  See Fleck’s ironic invocation of this principle (Genesis (ref. 1), 123). 48.  Bruno Latour, The Pasteurization of France, transl. by Alan Sheridan and John Law (Cambridge, MA, 1988). 49.  Lester King, The road to medical enlightenment (London, 1970). 50.  Edward Frankland, “On chemical changes in their relation to microorganisms”, Chemical news, ii (1885), 78–80; Kohler, “Innovation” (ref. 10), 189–1. 51.  As Mendelsohn (“‘Like all things that live’” (ref. 8), 13) points out, Pasteur himself did not uniformly maintain this position: “the specific fermentative or pathogenic effects of a given microbial species were relative to the milieu. Mycoderma aceti in wine, for instance, broke down alcohol into acetic acid and carbon dioxide, thus making vinegar. In vinegar, however, it turned the acetic acid into carbon dioxide and water.” 52.  Zinsser, 9th edn (ref. 13), 59. 53.  As its title indicates, the arch pleomorphist, Carl von Nägeli, in his book Die niederen Pilze in ihren Beziehungen zu den Infektionskrankheiten und der Gesundheitspflege, recognized a role for microbes in health and disease, but environment mediated microbial action (Amsterdamska, “Medical and biological” (ref. 3), 663). 54.  More than a half-century later, the audacity of that equating would be revealed in the first two of Evans’s precepts on acute respiratory diseases: “1) The same clinical syndrome may be produced by a variety of agents; 2) The same etiological agent may produce a variety of clinical syndromes” (A. S. Evans, Causation and disease: A chronological journey (New York, 1993), 46). Mazumdar (Species (ref. 9), 66) points to Koch’s 1878 paper on wound infections, and notes that Koch had added “a new dimension to his definition of the bacterial species, that of the species of disease with which it is associated: the disease and the bacteria define each other.…” 55.  K. Codell Carter, “Koch’s Postulates in relation to the work of Jacob Henle and Edwin Klebs”, Medical history, xxix (1985), 353–74. 56.  Mendelsohn (“‘Like all things that live’” (ref. 8)) considers such recognition of variability in the disease production by a supposed single species of microbe as evidence of the dominance of the concept of variable virulence by followers of Koch as well as of Pasteur. Virulence, he argues, effectively blended pleomorphist and monomorphist approaches, and served as a proxy for the exploration of many biological questions within bacteriology. Amsterdamska, “Medical and biological” (ref. 3) takes a more dichotomous view. She emphasizes the heuristic value of monomorphism for public health practice; while Mendelsohn (p. 5) is interested in “how the most applied and least theory-driven sciences may have intellectual shape”. To some degree, their differences may be explained by chronology and rhetorical situation. Mendelsohn focuses primarily on an earlier period; my own sense of Koch’s views on cholera suggests a hardening of his position between 1884 and 1885, but also some variation with regard to rhetorical occasion (Christopher Hamlin, Cholera: The biography (Oxford 2009)). More important, however, is the coexistence within bacteriological orthodoxy of what are at any given time, if not contradictions, statements which effectively take the form: “bacterium X is the necessary and sufficient cause of disease Y except when it is not.” Further research, into both variable virulence and immunity, may better clarify

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

296 ·

CHRISTOPHER HAMLIN

the conditions under which cause accounts for effect. In the meantime, the practical advantages of monomorphism were far preferable to the anarchic implications of pleomorphism. 57.  Zinsser, 9th edn (ref. 13), 112–13; Tanner, Bacteriology (ref. 13), 398. 58.  Kohler, “Bacterial physiology” and “Innovation” (ref. 10); Summers, “Culture as organism” (ref. 10). 59.  Tanner, Bacteriology (ref. 13), 78. 60.  Bulloch, History of bacteriology (ref. 8), 217. 61.  Tanner, Bacteriology (ref. 13), 89–91; Zinsser, 9th edn (ref. 13), 135. 62.  Summers, “Culture as organism” (ref. 10); Brock, Bacterial genetics (ref. 10). 63.  Tanner, Bacteriology (ref. 13), 84. 64.  Tanner, Bacteriology (ref. 13), 90. 65.  Zinsser, 9th edn (ref. 13), 152. 66.  Zinsser, 9th edn (ref. 13), 125–6. Amsterdamska (“Medical and biological” (ref. 3), 675–6) recognizes that several of the domains in which bacteria varied — enzymatic operations, serological capacities, and even mutation itself, were, ironically, translated into components of stability, and therefore, acquired taxonomic significance. “Initially reports of changes in fermentation characteristics of enteric bacteria were seen as a hindrance; by casting doubt on the validity of diagnostic tests they appeared to create new practical difficulties for bacteriological work.” “In investigations of antigenic structure, variation was no longer a subject but rather a useful method for investigating new problems … what was initially seen only as a hindrance to reliable clinical procedures was quickly transformed into a tool.…” 67.  In fact, such unions are hardly rare, Pickstone (Ways of knowing (ref. 4)) reminds us: most notably in medicine, crafts depend on a natural history of relevant active ingredients. 68.  Zinsser, 9th edn (ref. 13), 77. 69.  Philip Hadley, “Microbic dissociation: The instability of bacterial species with special reference to active dissociation and transmissible autolysis”, Journal of infectious diseases, xl (1927), 1–312, p. 2. Hadley’s magisterial review has intrigued both contemporaries and historians. Textbook authors recognize these varieties of instability, though they usually spread their discussion across multiple chapters (but see Carter, Microbiology (ref. 13), 76). 70.  Zinsser, 9th edn (ref. 13), 59, 130. 71.  Zinsser, 9th edn (ref. 13), 64; Latour, Pasteurization (ref. 48). As Kohler (“Innovation” (ref. 10), 179) notes, the potential range of measures of productivity was enormous: “Gale systematically measured the relative proportions of some twenty enzymes in E. coli as he varied the condition of growth: acidity, nutrients, age of culture, aeration, and so on. A few patterns did emerge: some enzymes seemed to increase in amount as if to compensate for adverse changes in the milieu; others seemed to proliferate only under conditions optimal for their activity.” But, as Kohler (“Bacterial physiology” (ref. 10)) points out, such explorations took place in the unorganized territory of biochemistry rather than the friendly confines of bacteriology. 72.  Mendelsohn, “‘Like all things that live’” (ref. 8). As Amsterdamska (“Medical and biological” (ref. 3)) points out, “Koch’s adherence to monomorphism was not simply a precondition of the germ theory of disease, but a guarantee that the aetiological connections he could establish by laboratory investigation would have relevance for the practical medical management of infectious diseases” (p. 665). 73.  Anderson, “Natural histories” (ref. 10). 74.  Hadley (“Microbic dissociation” (ref. 70), 2), is rare in entertaining the possibility that behind “chaos” there may be only chaos. 75.  Carter, Microbiology (ref. 13), 74–5 (italics mine). 76.  Zinsser 9th edn (ref. 13), 126–9. In retrospect, it would be suggested that the bacteriologists themselves were getting habituated to a Lamarckian handling of phenomena which seemed a

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

BACTERIOLOGY AS A CULTURAL SYSTEM    ·

297

good deal more profound and challenging to those outside bacteriology or who came after. E.g. Tanner, Bacteriology (ref. 13), 45: “these small single-celled microorganisms are quite likely to assume new characteristics when grown in the laboratory away from their natural habitat. New generations follow one another with great rapidity and new characteristics may be picked up quickly.” See Summers, “Culture as organism” (ref. 10); Kohler, “Innovation” (ref. 10), 178; Brock, Bacterial genetics (ref. 10). 77.  Zinsser, 9th edn (ref. 13), 130. As Amsterdamska (“Stabilizing” (ref. 10)) points out, monomorphists found such cyclogenetic models deeply problematic, even though they were much closer to a monomorphic ordering of the world than was the earlier pleomorphism. See also Summers, “Culture as organism” (ref. 10). 78.  Tanner, Bacteriology (ref. 13), 62–3. 79.  Tanner, Bacteriology (ref. 13), 58; Carter, Microbiology (ref. 13), 74–5. See also Zinsser, 9th edn (ref. 13), 125. 80.  While Nägeli had coined the term in contradistinction to evolution, one may well wonder whether its resonances drew also from “inversion”, the contemporary psychiatric term for homosexuality — one was concerned in both cases with deviance. 81.  Tanner, Bacteriology (ref. 13), 58–9. Both are giving voice to Hadley’s protest, quoted by Brock, Bacterial genetics (ref. 10), 26: “Under its [monomorphism’s] influence … there were set up strict notions of ‘normal’ bacterial cell types, ‘normal’ colony forms, and ‘normal’ cultures. Whatever departed from the expected normality was at once relegated to the field of contaminations; or to the weird category of ‘involution forms’, ‘degeneration forms’, or pathological elements possessing neither viability, interest, nor significance….” Hadley (“Microbic dissociation” (ref. 70), 2) complained that that efforts to explain away variation in pursuit of taxonomic stability prevented important studies of its causes. 82.  Zinsser, 9th edn (ref. 13), 125–6. 83.  Zinsser, 9th edn (ref. 13), quoting Chick, 80. “Although geneticists would later realize that a pure culture is a clone and can therefore be thought of as a population of dividing and occasionally mutating cells, bacteriologists were generally incapable of thinking in terms of populations of cells” (Brock, Bacterial genetics (ref. 10), 2; see also Summers, “Culture as organism (ref. 10)). 84.  Zinsser, 9th edn (ref. 13), 125. 85.  Brock, Bacterial genetics (ref. 10), 26. 86.  This scratches the surface. I have stressed identity here, but a broader focus on the relation of bacteriology to the germ theory of disease would highlight transformation in practices of causal attribution. I have hinted at, but not pursued, the remarkable moral status that bacteriology, and certain bacteriological practices acquired. 87.  Bruno Latour, Science as action (Cambridge, MA, 1987). 88.  Cf. the contrasting treatments of the work of Margery Stephenson by Brock, Bacterial genetics (ref. 10) and by Kohler, “Innovation” (ref. 10). “Another difficulty in the study of bacterial variability arises from the fact that bacteria, more than higher organisms, regulate enzyme synthesis in dramatic ways. The phenomena of enzyme induction and repression … are dramatically developed in bacteria, and lead to extensive physiological change when culture conditions are changed. Although these changes are phenomena of whole populations rather than of individuals, they confused many workers” (Brock, p. 46). Kohler (p. 179) complains: “After the discovery of bacterial genetics in the late 1940s, inducers of enzyme adaptation were seen as chemical messengers, not as participants in the chemical reactions of enzyme synthesis: induction by substrates was a specific process unrelated to other varieties of adaptive behavior. In this new view, the complex phenomena that had seemed so promising to Stephenson were merely red herrings.” 89.  Citing Lederberg’s 1947 work, the Zinsser authors (Zinsser, 9th edn (ref. 13), 131) are intrigued by one unexpected version of gene exchange. “By growing deficient strains in symbiosis, new organisms

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016

298 ·

CHRISTOPHER HAMLIN

were isolated which had the synthetic properties of both parent strains, suggesting that a fusion of two bacilli had occurred with an exchange of genes analogous to sexual fusion in higher forms.” 90.  Other means contributed too. A survey of later textbooks suggests that the widespread availability of electron microscopy by the mid-1960s made it possible to characterize much more of the fine structure of bacteria, and put less stress on differentiation by cultural or chemical means. 91.  Donna Haraway, Primate visions, gender, race, and nature in the world of modern science (New York, 1989); cf. Fleck, Genesis (ref. 1), 112. 92.  Hamlin, Cholera (ref. 56), 258–78. 93.  J. Reidl and K. Klose, “Vibrio cholerae and cholera: Out of the water and into the host”, FEMS microbiology reviews, xxvi (2002), 125–39.

Downloaded from hos.sagepub.com at UNIV OF NOTRE DAME on June 8, 2016