Of Lungs and Lungers: The Classified Story of Tuberculosis

Susan Leigh Star and Geoffrey C. Bowker

Graduate Schoolof Library and Information Science

University of Ilinois

501 East Daniel Street

Champaign, IL 61820

s-star1@uiuc.edu

bowker@alexia.lis.uiuc.edu

(217) 244-3280

(217) (217) 244-3302 (Fax)

Illinois Research Group on Classification

University of Illinois, Urbana-Champaign

URL: http://alexia.lis.uiuc.edu/~star/irgchome.html

Final draft: 12 October 1996

Acknowledgments

Robert Dale Parker suggested several helpful references on the literary background of women and disease in the 19th century; Marc Berg, Kirk Johnson, Helen Verran and an anonymous referee gave us very useful comments. Our colleagues in the Illinois Research Group on Classification, University of Illinois at Urbana- Champaign, have been helpful in forming the basis for a social analysis of classification (Niranjan Karnik, Randi Markussen, Laura Neumann, Karen Ruhleder, Stefan Timmermans). We also thank the Advanced Information Technologies Group, University of Illinois, for project support. Conversations with Kari Thoresen about the notion of texture in organizations were very helpful, as were ongoing conversations with Anselm Strauss about trajectory. The work of Mark Casey Condon on the nature of time morality in a men's homeless shelter was helpful in thinking through issues in the concluding section.

The experience of a long-term chronic illness weaves together problems in biography, the body, and the cultural and organizational meanings of the disease. Time meets infrastructure; experience meets classification. We present here a close reading of two studies of tuberculosis, Thomas Mann's novel The Magic Mountain and Julius Roth's empirical work, Timetables. Drawing on research in medical sociology and information science, we show how the trajectories of disease, biography and institution weave together. Mismatches in the trajectories produce distortions (or torques) in time and sense of self, accounting for the phantasmagoric imagery often associated with tuberculosis. We conclude with some general observations about how the concept of trajectory might be used in understand the intersection of biography and organization.

Introduction

"TB is a disease of time; it speeds up life, highlights it, spiritualizes it." (Sontag, 1977, p. 14)

The further one stands away from the disease of tuberculosis, the more it seems like a single, uniform phenomenon. It is associated with one of the great philosophical breakthroughs in medicine - Koch developed his "postulates" for defining disease agency partly with TB in mind. Indeed, he could hardly avoid it since epidemiologists assure us that at the time he wrote them, 1881, one seventh of all reported human deaths, and one third of deaths of "productive middle age" groups were attributable to TB (Brock, 1988, pp. 117, 179-180, emphasis added). Yet this single disease, the holocaust of those in their prime, has historically proved an elusive thing to classify. The work of classification has involved at many levels a complex ecology between localization, standardization and time.

TB also holds an important place in popular culture. For example, the gambler Doc Holliday was a legendary "gunslinger" and partner of Wyatt Earp and his brother, the famed American Wild West law enforcers. He suffered from TB, and a number of movies and books portray him with the disease.

"Lunger," screams the outlaw at Doc Holliday, "come out and fight. Prepare to die." Holliday appeared twice recently on the big screen, in the Hollywood films "Tombstone" and "Wyatt Earp." In both, his pale face, glistening with sweat, is used as a counterpoint to his devil-may-care lifestyle and his gun-happy camaraderie with the Earps. The six years he spent in a sanitarium at the end of his life in Glenwood Springs ,Colorado go unexplored. His tombstone there reads, "He died in bed." As a final irony in the story of romancing disease, localization, and macho myth, another stone reads, "Here lies Doc Holliday whose body is buried somewhere in this cemetery." It seems the body was hidden after he died from potential revenge-seekers; its exact whereabouts has been lost in the records.[1]

Body-snatching and the failure to reach a final localized resting place permeate the history of tuberculosis. As we proceed with this story, the interweaving of myth, biography, science, medicine and bureaucracy becomes thicker and thicker, eluding attempts at standardization and localization from every angle. Just for this reason, though, the story of TB holds some profound insights about how those threads intertwine, tense against each other, and form the texture of a landscape of time.

We bring a number of intellectual tools to the analysis of this landscape. Social studies of science have moved to crisscross nature, culture (including artifacts) and discourse in a seamless web (Latour, 1993); activity theory adds the concept of mediation for tools, cognition and community. We emphasize as well the importance of ongoing hybrid arrangements, in the form of infrastructures of classification and bureaucracy (Bowker, Timmermans & Star, 1995; Bowker & Star, 1994; Star & Ruhleder, 1996; Star, 1995). We draw as well on some concepts from medical sociology: notions of body-biography-trajectory (Corbin & Strauss, 1988; 1991) and the temporal lessons of chronic illness (Charmaz, 1992). We seek to re-center the ways in which time and infrastructure interact, to illuminate the texture of this web so crisscrossed with great divides that in literature and popular myth, the whole borderland has taken on a phantasmagoric shape, unrecognizable to those undergoing the experience.

Part 1 Classification

The question has often been posed in the social and cultural study of medicine: whose body is it anyway that is getting analyzed? (Berg & Bowker, in press) This question is as old as the development of statistics (Hacking, 1990): Quetelet sought for the "ideal type" in a statistical analysis of a regiment of Scottish soldiers. With tuberculosis, the body is constantly in motion, and the disease is constantly in motion. The disease may be localized or spread throughout the body; the state or general condition of the body and of the person's life enters into the treatment regime, which may take months and historically has taken years or a lifetime. Thus, any classification system should include both spatial and temporal dimensions, but standardized classifications tend to emphasize space alone. That is, classifications are rarely developmental, and often presented as spatial demographic distributions. Even where stages of a disease may be categorized, these stages are abstracted away from biographical continuity and more subtle temporal issues. As the problems of time emerge in the lives of patients and the work of classifiers, those spatial compartments break down in interesting ways: a formal hierarchy of mutually exclusive categories becomes a set of overlapping contradictory classes. The formal thesis of this paper is that when the work of classification abstracts away the flow of historical time, then the goal of standardization can only be achieved at the price of leakages in the classification system (cf. Hacking, 1995). Under certain conditions, the shifting terrain between standardized classification and the situated biography of the patient is twisted across an axis of negotiation, scientific work and instruments, suffering, and time.

1.1 The disease is constantly in motion

Tuberculosis is a moving target. It is often presented as the great epidemic disease with a cure - heralding the famous optimism (ironically, just as AIDS was developing and newly-resistant strains of TB on the rise) that epidemic disease could be eradicated from the planet. In September 1994 the World Health Organization made a worldwide press release predicting the eradication of polio from the planet by the year 2000 (http://www.who.ch/programmes/gpv/tEnglish/avail/polio.htm). A year earlier sociologist Fred Davis, who suffered polio in his youth, and one of the most eloquent analysts of uncertainty in illness (Davis, 1963) died of a stroke at the age of 65 -- was this stroke in part the legacy of his earlier illness? Many of those who had polio in the 1940s and 1950s are now beginning to lose their ability to walk as their overburdened spinal cells, designed for backup purposes, are wearing out after years of tough therapy and rehabilitation. Is the disease thus eradicated -- or delayed? In the lives of these patients, the answer is not so clear. And they remain invisible to the original classifications.

As Barbara Bates has pointed out (Bates, 1992, pp. 320-321) many observers "now attribute the decline of tuberculosis chiefly to socioeconomic changes" (pp. 320-321). A historically fully contingent rise in standard of living accompanied by less crowded conditions in the cities possibly worked the real miracle. Others, she points out, offer a more brutal, but still completely historical cause for the cure: they argue that it is a matter of natural selection, and what has happened is simply that those humans most susceptible to the disease are now dead. This sort of convergence is a problem in much scientific and medical research, and increasingly so (Bowker, 1994; Timmermans & Berg, in press; Latour, 1993). Here we have a "global truth" - the cure for TB - that may have been true because the "fitter ones survived" (the susceptible members died out) or because human environments changed (better living conditions) or because there is now an allopathic cure.[2] It is impossible to decide between these three causes, since according to the first two humanity before the "cure" (whichever one plumps for) is not the same thing as humanity after the cure - either the race will have changed biologically or the infrastructure that makes us know what we are will have altered.

The disease has its own history, a broken and contested path. In a series of works put out by the National Tuberculosis Association from 1950 to the present, TB often figures wryly as an actor in the text: much as Roy Porter (1990) has noted that "gout" in the eighteenth century had a character of its own. Thus the 1961 edition of the Diagnostic Standards and Classification of Tuberculosis noted that:

For our present purposes, therefore, tuberculosis is defined as that infectious disease caused by one of several closely related mycobacteria, including M tuberculosis, M. bovis, and M. avium. It usually involves the lungs, but it also involves and sometimes produces gross lesions in other organs and tissues. The clinical and pathologic pictures may range from acute to chronic. (p. vi)

TB taken as an agent traverses history and human bodies, taking hold in some and leaving others in a contingent historical progression, reaffirming the uneven, hierarchical value given to different patient's lives (Glaser & Strauss, 1965). The disease hits at different points in the life cycle (ibid., p. 6); thus becomes a complex mirror composed of nature, culture, discourse and infrastructure.

1. 2 The body is constantly in motion

The development of X-rays was perhaps the most significant breakthrough in the detection and diagnosis of tuberculosis (Pasveer, 1989). Unfortunately, the body itself is constantly in motion and varies by individual: so the ideal measurement is always a projection from a moving picture onto a timeless chart:

The perfect chest roentgenogram . . . is the aim of those who practice roentgenology. The very nature of the problem prevents the realization of this aim. The chest is a moving, dynamic part of the body and cannot be completely still. It varies from person to person. In some it is thin and easy to penetrate. In others it is thick and heavy from fat or muscle and hard to penetrate. Some lungs are stiff and hard to inflate. Others are made full and voluminous without great effort. To register lungs satisfactorily with these variables is at all times difficult. (Diagnostic Standards, 1955, p. 71)

Further, each body subject to TB is going through its own biographical and physiological, historical development - and as it develops TB changes. Thus: "the clinical picture of serious necrotic lesions of primary tuberculosis and widespread disseminations from them is observed more often in infancy than in later life and more frequently in nonwhite than in white persons." (Diagnostic Standards, 1955, p. 17). Thomas Mann describes one of the TB sanitarium. patients in The Magic Mountain: responding to another's new diagnosis of a "moist spot":

You can't tell," Joachim said. "That is just what you never can tell. They said you had already had places, of which nobody took any notice and they healed of themselves, and left nothing but a few trifling dullnesses. It might have been the same way with the moist spot you are supposed to have now, if you hadn't come up here at all. One can never know. (Mann, 1929, p.192)

Not only the disease and the body, but also the patient's experience moves constantly. Thus Bates points out that institutionalization in a sanitarium may well have worked because the relationships with nurses, doctors and other patients - together with removal from bad home conditions - may have done the work. She summarizes: "Psychological factors have long been thought to alter the course of tuberculosis, but their actual impact on outcomes is not known..." (Bates, 1992, p. 320).

1.3 Classification: a still life constantly in motion

With all these historical trajectories being inscribed into the course of tuberculosis at whatever unit of analysis one took (humanity, the disease, the body or the experience of the patient) it will come as no surprise that the work of classifying tuberculosis has generally had very complex temporal ramifications; and these ramifications have often led to problems in classification. It has also led to a sense by those in sanatoria of TB as inhabiting a phantasmagoric landscape, a borderland filled with monstrous experiences and distortions of time and self (Mann, 1929; Roth, 1963). The reasons for this are not simply the physical horrors of the disease, though those are terrible enough, but the ways in which our strands play out against each other in imagination.

One wants to classify tuberculosis first and foremost in order to say yeah or nay, whether a particular patient has the disease. This information can be used in order to suggest a treatment plan for the patient, and a map of action for officials in public health (Should we quarantine, isolate, educate, or give antibiotics?) This classification work is not easy. In the first place, the disease itself is multifaceted: "When faced with a difficult diagnosis, the clinician does well to keep tuberculosis in mind, for its mode of onset and course are protean. This needs to be urged all the more now that tuberculosis is becoming relatively less frequent." (Diagnosis, 1955, p. 23)

Further, it does not have a single cause: most TB in humans, according to official accounts, is caused by mycobacterium tuberculosis, but one should not forget mycobacterium bovis and mycobacterium avium. It does not appear in a single place - generally the lungs are affected, but it could produce lesions in many other organs and tissues. Star (1989) notes that the disease's spread implicated it in all investigations of nervous and brain disease in the nineteenth century; whether a patient had a brain or spinal tumor or TB was often unclear; the disease may cause seizures, paralysis, lameness or dementia. Thomas Mann poses of one of his characters: "the question whether the disease would be arrested by a chalky petrifaction and heal by means of fibrosis, or whether it would extend the area, create still larger cavities, and destroy the organ." (Mann, 1929, p.447)

And indeed even pulmonary TB - its most common form and one of the greatest killers in the history of humanity - cannot be simply classified:

The lesions of tuberculosis are highly diverse in appearance, and their manifestations are numerous. No single system of classification can give information which completely describes the lesions. Certain classifications and descriptions are needed, however, for records and for statistical purposes... the status of a patient's disease at the time of diagnosis, and at any time in the months and years thereafter. These basic classifications should be used for all cases. (Diagnosis, 1961, p. 39)

And they recommend that this basic classification should tell a story, detailing extent of disease, status of clinical activity, bacteriologic status, therapeutic status, exercise status and other "lifestyle" variables as they are called nowadays.

Medical classification work as based on the International Classification of Diseases (ICD), however, does not tell a story - it records a fact (one died of the disease or not). 3 There is a complex narrative written into the death certificate that is the primary outcome of the ICD: doctors and later health workers must sift multiple causes of death to determine proximate, contributing and underlying causes (Fagot-Largeault, 1989). But this is an impoverished story, with only a small range of contributing causes. The ICD cannot contain a protean disease.[4] It is oriented towards a cause-and-effect that resembles a set of slots or bins, or blanks on a form, even where it is multi-valanced and multi-slotted; it is not, like disease and diagnosis, messy, leaky, liquid, and textured with time. Indeed, the problem of tuberculosis has been a long-standing problem for the ICD - leading to the convening of several special committees to produce a standard (Biraben, 1988).

Standard medical classifications, though they may leak at the edges and become configurationally complex, do not reflect the temporal complexity of the disease itself (Clarke & Casper, 1996). They do not represent its composite, amodern nature: culture, nature, discourse and infrastructure. They posit a single answer to the question of whether this person has TB or not. However, as Desrosières (1993, p. 296) has pointed out with respect to all statistical work, this kind of difficulty leads to a contradiction between a logical, top-down approach which accounts for all traits, and a local, pragmatic one which registers the phenomenon as locally encountered. It is one of the purest forms of a deduction vs. induction debate.[5]

For most of the nineteenth century - and into the twentieth - TB was believed to be hereditary. What was classified was a tuberculoid kind of a person, a temperament: romantic, melancholy, given to emotional extremes, hot cheeked and so on. Sontag notes that just before Koch discovered the tubercule bacillus, a standard medical text gave the cause of TB as "hereditary disposition, unfavorable climate, sedentary indoor life, defective ventilation, deficiency of light, and `depressing emotions.' "(1977, p. 54) She also writes of the literary and popular cultural images of TB, noting that many writers have referred to TB as "ethereal," "chaste," and somehow pure and mental, not physical. "TB is celebrated as the disease of born victims, of sensitive, passive people who are not quite life-loving enough to survive." (1977, p. 25) In some circles in the nineteenth century, this became a romantic image, especially for middle-class white women: "The recurrent figure of the tubercular courtesan indicates that TB was also thought to make the sufferer sexy." (1977, p.25). Eventually, this romance bled over into a more diffuse concept of style of life and crafting of self. Sontag even states: "The romanticizing of TB is the first widespread example of that distinctively modern activity, promoting the self as an image. The tubercular look had to be considered attractive once it came to be considered a mark of distinction, of breeding." (p. 29) The mythic person with TB became a romantic exile, and the myth "supplied an important model of bohemian life, lived without or without the vocation of the artist. The TB sufferer was a dropout, a wanderer in endless search of the healthy place." (p. 33)

The work of finding a cure thus involved myriad classificatory activities inserted into a shifting ecology of metrologies and images about temperament and constitution (Star, 1995). Bates (1992, p. 28) notes that members of the Climatological Association in the 1920s compiled measures of altitude, humidity, temperature, sunlight, dampness of the soil, ozone in the air and emanations from pine and balsam forests in order to uncover and classify the ideal placement for sanitarium situation. As she somewhat sardonically notes, though, a skeptic: "might notice that many of the otherwise disparate conclusions shared one characteristic: physicians tended to discover health-giving attributes in their own locales" (p. 28).

It was (and is) also not clear when to stop classifying TB. As the following report on Bergey's manual of determinative bacteriology notes - apparently without irony - there is a need to bring order into the classification "unclassified" when talking about TB. the manual lists four subtypes of unclassified strains ( (Diagnostic Standards, 1961, p. 17) So one can have an "other" or residual category, but at some point even the garbage can will have to be ordered, when it becomes large enough.

Further, the committee on the classification of tuberculosis was forced to recognize in general that: "all classifications are ephemeral" (Diagnosis, 1955, p. 6 - quoting the 1950 edition). They fully recognized the temporary, agreed-upon nature of their classification work. In infrastructural work such as the development of classification systems, there is much greater sensitivity to such factors than appears in the published scientific papers: "Complete agreement . . . is impossible . . . The classification presented represents a well- considered compromise of the views of outstanding clinicians."(Diagnosis, 1955, p. 5 - quoting the 1950 edition).

Indeed, the historiography presented by the texts of diagnostic handbooks was a mixture of pure Whiggish progress tinged with despair ("without roentgenology the fight against TB would be back where is was in the nineteenth century" - Diagnosis, 1961, p.67) and a cyclical view of history that Vico would not have been ashamed to espouse:

Readers will note another of those shifts in emphasis which have characterized expositions of the pathogenesis of tuberculosis for thirty-five years. The concepts presented in the current edition are more closely allied to those of former years than to the views expressed in the last edition. (Diagnostic Standards, 1955, p. 7)

Or again, from the 1961 edition:

The one item of change upon which all of our consultants agreed was the need for a classification to include the increasing number of cases which are neither truly "active" nor "inactive," and, chiefly, cases of the "open negative syndrome." In defining such a new class and seeking a suitable name for it, we have reached back ten years and reinstated the once-retired term, "quiescent," which was previously applied to an intermediate class. (Diagnostic Standards, 1961, p. v)

This example moves us into the terrain of tuberculosis and activity, which we will consider in the next section. For our purposes here, though, it underscores the situation of the classification act in an historical flow, where the pure progress of natural science is transmogrified by time, biography, institutions, and myth.

Part 2 Freeze Frames: Snapshots of a Disease in Progress

A positive skin test does NOT mean that you have tuberculosis; rather, that you may have been exposed to the organisms at some time in the past. In this case, a chest x-ray must be obtained in order to be certain there is no active disease. Additionally, if the reaction is positive, we will want to review your history and talk to you about what you should do in the future. -- Information given to students by the health center at the University of Illinois. (c)University of Illinois Board of Trustees

Throughout the history of tuberculosis classification, one of the key problems has been how to convert a progressive, protean disease to a single mark on a sheet of paper. Many categories have been experimented with. One suggested hallmark was whether or not one tested positive to the tuberculin test. But as with HIV, it was decided that those who tested positive did not have the disease, they were: "considered to have tuberculous infection but not disease" (Diagnosis, 1955, p. 25). Only those who could bring other evidence of disease to the table would be considered worthy of the classification of pulmonary and non pulmonary tuberculosis.

Those who did have the disease, could be lumped into the categories inactive/active/activity undetermined. However, if: a "provisional estimate of the probable clinical status is necessary for public health purposes, the terms (a) "Probably Active" or (b) "Probably Inactive" should be used. Every effort should be made to classify cases and to avoid this category" (Diagnostic Standards, 1955, p. 28). By 1961, it was agreed that a classification somewhere between active and inactive was needed: this would be the "open negative syndrome" and would, as we have just seen, have the word "quiescent" attached to it. "Inactive" would be redefined to include: "constant and definite healing". Ironically, and to underscore the attempt to separate disease from biography, "dead" was also recognized in this classification (p. 41) - presumably to stand as a cross between highly active and completely inactive!

Bubbling out of the freeze frame with these leakages are the struggles with a shifting infrastructure of classification and treatment. Turning now to other presentations and classifications of TB by a novelist and a sociologist, we will see the complex dialectic of irrevocably local biography and of standard classification.

Part 3: Moving Through Tuberculosis and Its Classification

The next sections rely on two detailed readings of classic studies of TB sanatoria and hospitals. The first is Thomas Mann's The Magic Mountain (1929), written in 1912, a 900+ -page tome chronicling a Swiss hospital and the 7-year sojourn of a young German engineer there, Hans Castorp. The account was based on Mann's experience as a visitor to a similar institution during his wife's incarceration for lung disease. The second is Julius Roth's Timetables, a comparative ethnographic analysis of several American TB hospitals in the late 1950s. This volume, too, has a strong base in Roth's own experience as a TB patient while he was writing his doctoral dissertation.

3.1 The texture of time: Lost to the world

When Hans Castorp, the hero of Mann's novel, arrives in the Alps as a visitor to his tubercular cousin, one of his first lessons in local culture is the way that values about time change for those "up here." Everything normal seems to change for him, and the whole place seems macabre and oddly funny. Later in the novel he will explain to another newcomer: "I have no contact with the flat-land, it has fallen away. We have a folk-song that says: "I am lost to the world" -- so it is with me." (Mann, 1929, p. 614) This lostness first takes on the form of time passing very slowly, and in chunks that seem unimaginable to the newcomer: "We up here are not acquainted with such an unit of time as the week - if I may be permitted to instruct you, my dear sir. Our smallest unit is the month. We reckon in the grand style -- that is a privilege we shadows have." (Mann, 1929, p. 59)

Roth compares the commitment to a TB sanitarium with having an "indeterminate sentence" for one's first year of jail -- not knowing how long one will be incarcerated, not having any milestones or turning points that make sense, also makes time seem endless, and distorted with respect to known landscapes, both inner and outer. "Where uniformity rules; and where motion is no more motion, time is no longer time." (Mann, 1929, p.566)

The patients in both Mann's and Roth's hospitals begin to speculate on the meaning of this lost time, this "time out." Is time real, objective, something which can be measured externally -- or subjective, illusory? Hans originally opts for a relativist explanation: "After all, time isn't 'actual.' When it seems long to you, then it is long; when it seems short, why, then it is short. But how long, or how short, it actually is, that nobody knows." (Mann, 1929, p. 66) His cousin Joachim, a rather hard-nosed soldier who wants only to get off of the mountain and back to his regiment, disagrees, and says "We have watches and calendars for the purpose; and when a month is up, why then up it is, for you, and for me, and for all of us." (Mann, 1929, p.66) Hans proceeds to demonstrate how slowly seven minutes can go by while taking one's temperature, and we indeed feel the seconds creep by in Mann's precise language. What is "the same?" he asks. "The schoolmen of the Middle Ages would have it that time is an illusion; that its flow in sequence and causality is only the result of a sensory device, and their real existence of things in an abiding present." (Mann, 1929, p.566)

As time goes on, up on the magic mountain and in each of the hospitals studied by Roth, people inside begin to develop a sense of how to fragment, break up this unbroken monolith. "We are aware that the intercalation of periods of change and novelty is the only means by which we can refresh our sense of time, strengthen, retard, and rejuvenate it, and therewith renew our perception of life itself." (Mann, 1929, p. 107) In one of his many meditations on the nature of time, Mann argues that time and action and space are not separable -- nothing "fills up" time in a platonic-container sense, but that these facets are only knowable with respect to each other:

What is time? A mystery, a figment -- and all powerful. It conditions the exterior world, it is motion married to and mingled with the existence of bodies in space, and with the motion of these. Would there then be no time if there were no motion? No motion if no time? We fondly ask. Is time a function of space? Or space of time? Or are they identical? Echo answers. Time is functional, it can be referred to as action; we say a thing is "brought about" by time. What sort of thing? Change? (Mann, 1929, p. 356)

At the core of this theory of action is the development of what Roth calls "timetables," and which is alluded to in more symbolic terms by Mann. Timetables are breaks in space-time which give meaning to action. They are constructed mediational tools which help order and control mismatches between institution and individual. When will I get out? What will become of me? How will I survive the boredom and the uncertainty of incarceration? Such questions are asked against the specter of unbroken time or eternity, or as Roth's patients and doctors put it for the hopeless cases: "a rather horrifying tubercular Siberia -- a seemingly endless waster (of time) without any signposts along the way." (Roth, 1963, p. 21) or in Mann's words: ""Only in time was there progress; in eternity there was none, nor any politics or eloquence either." (Mann, 1929, p.479) [6]

Gradually a sense that there is in fact no such thing as unbroken time comes about for the patients:

Can one tell -- that is to say, narrate -- time, time itself, as such, for its own sake? That would surely be an absurd undertaking. A story which read: "time passed, it ran on, the time flowed onward" and so forth -- no one in his senses could consider that a narrative....For narration resembles music in this, that it fills up the time. It "fills it in" and "breaks it up," so that "there's something to it," "something going on." (Mann, 1929, p.560)

The patients begin to fill their days with measurement. On the magic mountain, people walk around with thermometers in their mouths, measuring their temperatures several times a day; in both books, patients are conversant with the details of diagnosis and measurement, the myriad of ways in which the monolithic diagnosis may be broken up and measured. As Roth says: "Everyone is frantically trying to find out how long he is in for. The new patient questions the doctors, nurses, and other hospital personnel in an effort to discover how may years, months, and days it will take him to be cured." (1963, p. xvi)

3.2 Metrology

The importance of mediational tools appears in struggles with measurement. One woman has been a patient on the magic mountain for the better part of her life. Eventually, she is cured of the disease and told to go home. But she knows no other life, and panics at the thought of leaving.[7] She sabotages her release: runs out in the snow, jumps in the lake, and sticks her thermometer into her tea to make her appear feverish. When she is discovered by the staff, they make her use a thermometer without any marks on it. The device can only be read by a doctor with a measuring stick. The patient thus cannot calibrate her faking illness. The patients come to call this unmarked instrument "the silent sister," and it becomes the symbol for the ways in which the world of the asylum acquires its own bizarre culture of metrication and control.

Roth notes that patients are quite systematic in creating measurements for the blocks of time they will spend in the asylum. They begin to construct timetables for themselves (I will get out in six months; I will have surgery in two weeks, etc.) "After they have been in the hospital for some time, they find that "mild" and "bad" are not very meaningful categories," and much more detailed matching categories develop. (Roth, 1963, p.19) Patients begin observing how other patients are treated. There is an elaborate system of privileges in TB hospitals, based ostensibly on how healthy one is. If one is making good progress, for example, one is allowed out on brief shopping trips, and so forth. "He divides the patient group into categories, according to his predictions about/the course of their treatment. He can then attach himself to one of these categories and thus have a more precise notion of what is likely to happen to him than he could from simply following the more general norms." (Roth, 1963, pp. 16-17)

Roth goes on to describe an equally elaborate system of observations and comparisons made by all the patients with respect to their own bodies, the length of time "served," the predilections of the individual doctors, and the technical diagnostic material such as x-rays. Not surprisingly, much of the information available is partial or misleading:

Reference points may be more or less clear-cut and stable. If they are prescribed in detail and rigidly adhered to, as in the career of pupils in a school system, one's movement through the timetable is almost completely predictable. As the reference points become less rigid and less clear-cut, they must be discovered and interpreted through observation and through interaction with others of one's career group. The more unclear the reference points are, the harder it is for members of a career group to know where they stand in relation to others and the more likely it is that they will attend to inappropriate clues and thus make grossly inaccurate predictions concerning future progress. The degree of stability is related in part to the changes in timetables through time. (Roth, 1963, pp. 99-100)

Managing this instability increases the intensity of comparison and a sense, often, of bewilderment, unfairness, or even madness. Hans Castorp says to his cousin:

I cannot comprehend why, with a harmless fever -- assuming for the moment, that there is such a thing -- one must keep one's bed, while with one that is not harmless you needn't. And secondly, I tell you the fever has not made me hotter than I was before. My position is that 99.6deg. is 99.6deg.. If you can run about with it, so can I." (Mann, 1929, p.176)

Give me a standard, give me something to hold on to, something clear - in the face of uncertainty, patients become positivists. Mann describes the rebellion of Hans' cousin again the system of metrication in the hospital, the "Gaffky score." This is a composite score for each patient's progress based on a number of measures:

Yes, the good, the patient, the upright Joachim, so affected to discipline and the service, had been attacked by fits of rebellion, he even questioned the authority of the "Gaffky scale." Whether only a few isolated bacilli, or a whole host of them, were found in the sputum analyzed, determined his "Gaffky number," upon which everything depended. It infallibly reflected the chances of recovery with which the patient had to reckon; the number of months or years he must still remain could with ease be deduced from it. (Mann, 1929, p.357)

This questioning of authority appears inevitable in a landscape so filled with uncertainty. One character, a business person, attempts to quantify health care costs and the tradeoffs:

The expense, he whispered, was fixed at a thousand francs, including the anesthesia of the spinal cord; practically the whole thoracic cavity was involved, six or eight ribs, and the question was whether it would pay . . . he was not at all clear that he would not do better just to die in peace, with his ribs intact. (Mann, 1929, p.315)

In the absence of metrics such as this, however, the relationships between doctors and patients come under considerable strain as patients strive to assign themselves to the proper categories, and then to see whether the doctors agree with them (Berg, 1992). In The Magic Mountain, the inmate Settembrini, a slightly satanic character, whispers constantly to Hans about how subjective the reading of the objective measures such as x-rays really is:

You know too that those spots and shadows there are very largely of physiological origin. I have seen a hundred such pictures, looking very like this of yours; the decision as to whether they offered definite proof or not was left more or less to the discretion of the person looking at them. (Mann, 1929, p.250)

Both physicians and patients struggle to find a standard and to localize it, in the face of a constantly shifting interpretive frame. As Roth notes:

The physician finds it difficult to carry out the medical ideal of an individual prescription for each case when at the same time he recognizes the fact that his timing of a given treatment event for a given patient is to a large extent a highly uncertain judgment on his part. If you are going to guess, you might as well make the process more efficient by guessing about the same way each time, especially if you are in a situation where your clients are likely to think that you do not know what you are doing if you change your guess from one time to another. (1963, p.24)

This uncertainty leads to the struggles and negotiations that are at the heart of Roth's analysis. Whose timetable will prevail?

3.3 Classification struggles: Putting in time

"The TB patient conceives of his treatment largely in terms of putting in time rather than in terms of the changes that occur in his lungs." (Roth, 1963, p. xv) The length of time one has been inside, combined with patients' observations about where they belong in the general scheme of things, acquires a moral character over time:

A classification system contains within it a series of restrictions and privileges. When no rigid classification system exists, these privileges themselves become part of the timetable . . . How long is it before he is allowed two hour a day "up time" [out of bed]? . . . these privileges are desired not only in themselves, but for their symbolic value. They are signs that the treatment is progressing, that the patient is getting closer to discharge. (Roth, 1963, p.4)

Timetable norms differ from hospital to hospital and from patient to patient. Trust, often in the form of moral condemnation or approval, may play a big part in structuring the timetable negotiations between doctor and patient. For example, alcoholic patients are often refused outside passes, or sometimes a patient with a "recalcitrant attitude" is refused a pass simply in order to convince them that he or she really is very ill. These moralizing attitudes, well documented within medical sociology, add another texture to the landscape we are examining here, twisting it a little away from a simple formal-situated or realist-relativist axis.

Doctors as well as patients may hold the "deserving" attitude toward someone who has "served their time." Roth notes that in treatment conferences, how long the patient has been confined is always taken into account in deciding the timetable, "this in itself is given considerable weight entirely aside from the bacteriological and X-ray data." (1963, p.27) Even those who seem to be getting better much faster according to these tests are kept in longer because, "TB just isn't cured that fast." (p.27)

Patients know almost to the day when which privileges will arrive: "This relative precision of the timetable results from the emphasis placed upon the classification system by the staff, the consistency in the decisions of the physician in charge, and the physician explicitness in telling the patients what they can expect in the future." (Roth, 1963, p. 7)

There can be a "failure to be promoted" in severe cases, and the reaction to this "varies greatly among TB patients, just as among engineers some of the failures are emotionally disorganized when they do not make the grade while others accept their inferior position with relative equanimity. Some patients regard a few days' delay as a tragedy." (Roth, 1963, p. 15) Bargains are made: "Patients are sometimes given regular and frequent passes to induce them to remain in the hospital." (p. 53)

Uncertainty plays a big role in negotiations about classification in the hospital. When a patient tries to guess their classification, and the physician disagrees, "In effect, the physician tries to get Jones to change his criteria for grouping patients so that his categories will be closer to those of the physician." (Roth, 1963, p. 39) The doctor will provide the patient with examples of others like him or her, and relates details about other similar cases. But the physician too is caught in a double bind: ethically he or she is not allowed to give too many details about others' cases. The doctor is thus reduced to vague generalizations like "No two cases are alike." (ibid., p. 39) for the patient, this contributes to the house-of-mirrors effect:

Most physicians . . . vary their approach from one patient to another according to their own judgment of what the patient can take . . . the physicians do not know with any precision how long it will take the patient to reach a given level of control over his disease. In order to allow themselves a freer hand in deciding what the best time is for the patient to leave the hospital, the doctors try to avoid being pinned down to any precise estimates by the patients. (Roth, 1963, p.45 )

This twisting effect of these silences is especially clear where the norms about timetables are also shifting, either due to changes in medical practice, technology, staff, or organizational change. One patient in these circumstance said: " You never seem to get anywhere because people here don't pay too much attention to the classifications. I've been here now since November and I'm still in Group 1. My husband comes to visit me and looks at this tag and thinks I'm never going to get promoted. He wonders what's going on. Then when you do get promoted to Group 2, you don't know what it means, anyway. You have no idea what additional privileges you have . . . It's like an ungraded school room. " (Roth, 1963, p.10)

The ungraded schoolroom, combined with uncertainties, shifts in bureaucracy, and changes in the person's biography, begin to form the tapestry of a monstrous existence.

Part 4. Borderlands and Monsters: Time's Torquing of Standards and Experience

Greta Garbo as Camille drifts across the screen in a cloud of white organza.[8] She is alternately cruel and flirtatious, vulnerable and powerful. She plays with the affections of her lovers, a baron and a struggling young diplomat, from her position as a farm girl who came to Paris. Early in the movie, we understand that she has been ill; from time to time she discreetly covers her mouth with a handkerchief, or seems to swoon (always artistically). At times she recovers, and in a rhythm complexly played out against her wardrobe, she moves from white to black in dress, from sick to well, from powerful to powerless, from country to city. As the movie progresses she becomes more and more ill, and more and more "pure" -- thinner, whiter, more in love with the worthy poor man and less with the nefarious rich Baron. During the whole course of the movie, no one speaks the name of her illness, any prognosis or diagnosis, nor do we see any blood, sputum, feces, or other despoiling of the purified background. Of course, she has tuberculosis -- and she is the ideal type, the shadow puppet against which both the medical story and the rich cultural criticisms of TB have been played out.

There were those who/wanted to make him "healthy," to make him "go back to nature," when, the truth was, he had never been "natural." (Mann, 1929, p.482)

On the magic mountain, or in any of the hospitals analyzed by Roth, the sense of unreality, of being outside of "normal" time and of making up an idiosyncratic timing is very strong. Furthermore, the very insides and outsides of people become mixed up in an almost monstrous way; Hans carries around his love Clavdia's x-ray in his breast pocket so that he may really know her. External time drops away as does one's biography:

(The inhabitants) accorded to the anniversary of arrival no other attention than that of a profound silence . . . They set store by a proper articulation of the time, they gave heed to the calendar, observed the turning-points of the year, its recurrent limits. But to measure one's own private time, that time which for the individual in these parts was so closely bound up with space -- that was held to be an occupation only fit for new arrivals and short-termers. The settled citizens preferred the unmeasured, the eternal, the day that was for ever the same. (Mann, 1929, p.427)

This sense of time begins to blur important distinctions between life and death, time and space: "But is not this affirmation of the eternal and the infinite the logical -mathematical destruction of every and any limit in time or space, and the reduction of them, more or less, to zero? is it possible, in eternity, to conceive of a sequences of events, or in the infinite of a succession of space-occupying bodies? " (Mann, 1929, p.356) As we approach the zero point in the story, Mann notes in an afterword that time-space relations are shifted so that: "the story practices a hermetical magic, a temporal distortion of perspective reminding one of certain abnormal and transcendental experiences in actual life." (p. 561) In the following section we offer a model for how such a monstrous borderland terrain is constructed.

4.1 Trajectories and twists: The texture of action

No one can ever know for certain just when tuberculosis become active or when it becomes inactive. For that matter, one can never be certain that the disease is inactive, and a patient could logically be kept in the hospital for the rest of his life on the assumption hat some slight undetectable changes might be occurring in his lungs. (Roth, 1963, p.30)

The same train brought them as had Hans Castorp, when years ago, years that had been neither long nor short, but timeless, very eventful yet "the sum of nothing," he had first come to this place. (Mann, 1929, p.520)

The model developed below has three parts (note that these are not stages). The original work of Corbin and Strauss was elaborated by Timmermans, and in this paper, we add to both of them. They are thus parts of an ongoing conversation between all these authors. (For simplicity of representation, time, in the diagrams below appears to be a linear unfolding along the x-axis. Clearly, we are arguing against this simple "unfolding" temporal structure; our own limits on drawing prevent a more complex representation, which would probably resemble a weaving-in-motion in at least 4 dimensions!)

4.1.1 Body-Biography Trajectory: Corbin and Strauss

A model developed by Juliet Corbin and Anselm Strauss describes what happens in the course of a chronic illness (1988, 1991). They posit that bodies and biographies unfold along two intertwined trajectories (the "body-biography chain"), nestled in a matrix of other structural and interactional conditions (Strauss, 1993). For example, a heart attack may temporarily interrupt work, home life, creativity, dragging "down" the trajectory of biography -- of course, this in turn is contingent on a number of other circumstances such as having access to health care, living in a war zone, having another illness which makes recovery longer. The chain can be viewed geometrically, as a topography emerging from the interplays of these factors. Many illnesses do not have such an acute nature; during a long chronic illness there is a back-and-forth "tugging" across the trajectory of the disease/body and of the person's biography, within the conditional matrix. The title of Kathy Charmaz' Good Days, Bad Days: Time and Self in Chronic Illness throws this relation into relief (1991). A long, slow downswing may only very gradually affect biography; a brief acute phase, experienced over and over, may be compensated for by the person's and family's resources, so that the overall trajectory of the biography remains fairly smooth. Many possible shapes are envisioned: a looping shape in the case of a "comeback" after a serious, debilitating illness; a very, very gradual progress of the disease which slowly erodes the biographical trajectory.

This model does not seek a Cartesian "mind-body" dualism, but rather to find a language for the ways in which two (or more) different processes become inextricably intertwined into one thick chain or braid. It makes more complex the sick-well, able-disabled dichotomies, and brings in people's active conversations with and work with their ill bodies as a central concern.

--------------------------------------------------------------

Insert Figure 1 Here

--------------------------------------------------------------

The body's trajectory and the self's are bound together, but not completely tightly coupled. Careers, plans, work, and relationships may continue in spite of, around, and through illness; or, a sudden illness may interrupt plans and biography and reshape the topography. The background "landscape" is a nested set of contingent possibilities and structural features which in turn act upon the shape of the trajectory. Here, the solid line represents the trajectory of the body, as it follows up and down the course of a disease and recovery. The dotted line represents the person's biography, which is "pulled" and "pushed" metaphorically by the body, but is not necessarily wholly determined by it (Corbin & Strauss, 1988; 1991)

4.1.2 Multiple identities along a body-biography trajectory in sudden illness or death

Timmermans (in press; 1995), in a dialogue with the Corbin/Strauss BBC model, suggests emphasizing multiple identities rather than a single biographical identity. He studied more than 100 cases of attempted resuscitations (CPR) with victims of cardiac arrest, in the emergency rooms of hospitals. He attempted to use the trajectory model to explain the sequence and flow of events as people were brought in by the ambulance crews, "worked on" by staff, and either declared dead or saved. (The vast majority of people die.) In this extreme case of the interaction between body and biography, he observed an interested elaboration of the relationship by focusing on the multiple biographies all people do have.

Many have noted that no individual is unitary; we are all multiple selves (Star, 1991). Timmermans notes, along these lines, that each patient who undergoes CPR has multiple intertwining identities outside that of "heart attach victim," each with its own trajectory. A single patient is at once father, farmer, church member, student, and president of the Rotary Club. Each of those selves has its own trajectory, which while intertwined, are also historically independent. During the moments of resuscitation attempt, these multiple selves collapse into a single identity: that of the body/machine (Timmermans, in press). That is, the nurses and doctors and technicians focus down, relating only to the body aspect of the person, and collapsing all of the potential selves and identities of the person into one.

After resuscitation (whether or not this is successful) the multiple identities restart. The body is again accompanied by the other selves, and the father, farmer, etc. returns. On return, each identity will have been differentially altered by the experience -- perhaps the farmer can no longer farm -- yet the person is the same person. In Timmermans' terms, there has been an isomorphic transformation.

--------------------------------------------------------------

Insert Figure 2 Here

--------------------------------------------------------------

Timmermans' modifications to the trajectory model in acute, severe illness (cardiac failure) are pictured above. Here the biographical trajectories, selves, move from complex, multiple activity to a single focus: life-saving. At death, the identities are restored. If the patient survives, they are the same -- but different: an isomorphic transformation (Timmermans, in press).

4.1.3 The twisted landscape: Adding texture to multiplicity and standardization

In this paper, we have added a third trajectory dimension to play off against the interacting trajectories of bodies and multiple identities: the trajectory of classification systems (as part of infrastructure). In looking at an extreme case temporally - where the "time" of the body and of the multiple identities cannot be aligned with the "time" of the classification system, we have suggested that the latter gets twisted by the former. In the case of TB, a variety of monstrous classification schemes bubble through the rift in space/time. There are mismatches between lived experience and the rigidities of institutionalized classifications; there are struggles and negotiations between doctor and patient which are situated and which may seem erratic. For the patients, they have a chronic illness which necessitates withdrawal for a prolonged period from "normal life," sequestering with others with the disease, in an uncertain time frame that is partly dependent on the ways classification schemes are perceived, negotiated, and used by health personnel. As with the models of Corbin, Strauss and Timmermans, our model draws on a matrix of possibilities for the basis for these negotiations. These are culturally and historically specific, and include such factors as how medical knowledge is represented by public health agencies, how classifications are modified during the hospital stay, and images from literature, film and popular science about "what people with tuberculosis are like." The rich topography of body and biography intercalates with a bureaucratic/infrastructural typology (classification scheme).

--------------------------------------------------------------

Insert Figure 3 here

--------------------------------------------------------------

The twisted landscape of multiple identities and classifications is pictured above. In the torqued model, the thick solid line represents the institutionalized classification schemes; these get modified and "broken" more slowly than the timeline of the patients' bodies or biographies. There is still a body trajectory, indicated by the thinner solid line, which classically moves up and down over time, during the long course of the disease. Finally, after Timmermans, we have multiple biographical trajectories -- patients are struck with the disease as they are progressing in studies, family raising, love affairs, careers, and so on. Unlike the Timmermans model, the long years passed "out of time" may fracture these identities and transform them beyond familiarity -- another echo of a "monstrous" or distorted sense of the disease.

The topology created by the body-biography trajectory is pulled against the idealized, standardized typology of the global classification of tuberculosis -- itself a broken and moving target.[9] When standard classifications are added to the scheme, patients try to fit their experiences along both body and biographical trajectories to a standard picture or metric. changing definitions, local arrangements, and complex relations of all three trajectories contribute to "torquing"; this may be amplified by unsuccessful negotiations.Conclusion. Twists and Textures: Classification and Lived Experience

"Time morality is not cut and dried." (Condon & Schweingruber, 1994, p. 63 of MS)

The information infrastructure that deals with tuberculosis, as with other diseases, operationalizes a classification system that does not cope well with the subtleties of biographical time, experience, or negotiations about reality. It often uses spatial coordinates: the disease as localized in this body or not; in this region of the body and no other; it is present among this population but not that one; the cure can be found in this place but not that and so forth. The closest that one gets to the flow of time is the description active or passive, latent or virulent; occasionally, in links with rough life stages such as geriatric or pediatric. But this describes an information infrastructure dealing with medical knowledge - a knowledge that defines itself as being true "for all time" about its subject and so able to abstract contingent historical and biographical flow to uncover the underlying reality.

However, there is not just one kind of classification in the world: classification work is always multiple. As we get further from medical knowledge and closer to the suffering patient, time seeps into the classification systems that get used: how long does Group 1 stay here ? (Roth) ; How may I get reclassified so that I can pass more (or less) time on the magic mountain? (Mann). Camille's morality tale unfolds in time in binary oppositions of good/bad, fit/ill, black/white; promotion or demotion from class to class in a continually downwards career trajectory. Tuberculosis is the archetypal disease of time: chronic, recurrent, progressive.

So what happens when the disease of time meets the classification of space? As we have shown, the formal, spatial classification twists. "Other" categories run rampant, each seeking a way of expressing the elusive, forbidden flow of time (words like "quiescent" and "non-active" abound). A macabre landscape is born. And the historiography of the classification system twists back on itself: in stunning contrast to most medical scientific texts, tuberculosis classifiers speak of a cyclical flow to their own historical time (not linear progress). From the other point of view, that of the patient, orthogonal classifications are developed that do not even interact with medical categories ("I have put in my time here, and I am a good person, so I deserve to be better and to leave.") The individual's disease is given a temporal texture at the price of becoming purely local - abstracted away from the standardized language it becomes once again temporally textured and immediate.

This way of framing the problem introduces the idea of texture as an important one in the representations-work-body-biography literature emerging from both social studies of science and cultural/historical psychology. Lynch's work on topical contextures implies a similar direction, we think: the look and feel of being in a place and using a genre of representations (1995). Kari Thoresen (personal communication), a former geologist, is developing a model and vocabulary for different aspects of texture in organizations and technological networks, examining layers and strata, crystallization processes (a term also used by Strauss and Timmermans), and other metaphors to examine how wires, people and bits are put together by a large organization.

Why should social scientists be interested in such twists? We have suggested that as we can move on from exploring the seamless web of science and society, of nature and knowledge to an analysis of the information infrastructure that acts as matrix for the web. The web itself is textured in interesting ways by the available modes of information storage and transfer. Medical classification work, typified by the ICD, deals in primarily spatial compartments; and these compartments cannot hold when biography and duration is a necessary part of the story. In general, the information infrastructure holds certain kinds of knowledge and supports certain varieties of network; and we believe that it is a task of some urgency for our field to analyze which kinds of knowledge and network. We have also heard echoes throughout this research of the ways it is for those living with and researching AIDS (Epstein, 1995), and hope to add to the rich analysis of AIDS, experience, activism and research currently taking shape in our field. As well, the notion of activity theoretic spaces with development proceeding dialectically between individual, community, and the mediations of material artifacts may be augmented by these notions of texture (cf. Engeström, 1987). The zone of proximal development, and the contradictions posed by mismatches between individual experience and institutional mores, may have some of the textural qualities we discuss here. Under some conditions (of silence, of denial of experiential validity, of delayed answers or justice), the sense of torquing could be quite an important quality. We need ways to understand experience which seems monstrous, or out of place, or "queer" (in the negative sense). We are reminded here of Michel Serres' (1980) invocation of the passage between the natural and humanistic sciences as indeterminate, twisted, and full of ice floes; of the images of cyborg and monster pervading feminist theory about technology (Haraway, 1992; Casper, 1994a, 1994b, 1995).

Much of our previous work (Bowker & Star, 1991; 1994; Star, 1991) has concerned itself with the relationship (which we first conceptualized as a kind of gap) between formal systems of knowledge representation and informal, experiential, empirical, and situated experience. Berg (in press) has shown convincingly that it is never the case of "the map OR the territory," but always, using the example of medical protocols, "the map IN the territory." This is a modest attempt to look at one kind of map in a territory marked by severe biographical interruptions, solitude and aspects of total institutions, and in dialogue with a compelling infrastructure (both informatic and managerial). We see the map-actant in this case as a warping factor, not in the sense of deviating from any putative norm, but in the sense of reshaping and constraining other kinds of experience. Note

Correspondence regarding this article should be sent to Leigh Star or Geoffrey C. Bowker, Graduate School of Library and Information Science, University of Illinois, 501 East Daniel Champaign, IL 61820. Email: s-star1@uiuc.edu; bowker@alexia.lis.uiuc.edu.

References

Bardet, J-P., Bourdelais, P., Guillaume, P., Leburn, F., & Quétel, C. (Eds.) (1988). Peurs et terreurs face la contagion. Paris: Fayard.

Bates, B. (1992). Bargaining for life: a social history of tuberculosis, 1876-1938. Philadelphia: University of Pennsylvania.

Berg, M. (1992). The construction of medical disposals: Medical sociology and medical problem solving in clinical practice. Sociology of health and illness 14, 151-180.

Berg, M. (in press). Rationalizing medical work: Decision support techniques and medical practices. Cambridge, MA: MIT Press.

Berg, M., & Bowker, G. (in press). The multiple bodies of the medical record. The Sociological Quarterly.

Biraben, J-N. (1988). La tuberculose et la dissimulation des causes de decès. In Bardet, J-P., Bourdelais, P., Guillaume, P., Leburn, F., Quétel, C. (Eds.) Peurs et terreurs face à la contagion. (pp. 184-198). Paris: Fayard.

Bowker, G. (1994). Science on the run: Information management and industrial geophysics at Schlumberger, 1920-1940. Cambridge, MA: MIT Press.

Bowker, G., & Star, S.L. (1994). Knowledge and infrastructure in international information management: Problems of classification and coding. In Bud, L. (Ed.) Information acumen: The understanding and use of knowledge in modern business. (pp. 187-213). London: Routledge.

Bowker, G., & S. L. Star. (1991). Situations vs. standards in long-term, wide-scale decision-making: the case of the international classification of diseases. In Proceedings of the 24th Hawaiian International Conference on Systems Sciences. Vol. IV. (pp. 73-81). Washington, DC: IEEE Computer Society.

Brock, T. D. (1988). Robert Koch: A life in medicine and bacteriology. Berlin:

Springer-Verlag.

Casper, M. (1994a). At the margins of humanity: Fetal positions in science and medicine. Science, technology and human values, 19, 307-323.

Casper, M. (1994b). Reframing and grounding nonhuman agency: What makes a fetus an agent. American behavioral scientist , 37, 839-856.

Casper, M. (1995). Fetal cyborgs and technomoms on the reproductive frontier, or which way to the carnival? In Gray, C. H., Figueroa-Sarriera, H. & Mentor, S.(Eds.) The cyborg handbook. New York: Routledge.

Charmaz, K. (1991). Good days, bad days: The self in chronic illness and time. New Brunswick, NJ: Rutgers.

Clarke, A., & Casper, M. (1996). From simple technology to complex arena: Classification of pap smears, 1917- 1990. Medical anthropology quarterly , 10: 450-475.

Condon, M. C., & Schweingruber, D. (1994). The morality of time and the organization of a men's emergency shelter. Unpublished manuscript, Dept. of Sociology, University of Illinois, Urbana-Champaign.

Corbin, J., & Strauss, A. (1988). Unending work and care: Managing chronic illness at home. San Francisco: Jossey-Bass.

Corbin, J., & Strauss, A. (1991). Comeback: The process of overcoming disability. In Albrecht, G. & Levy, J. (Eds.), Advances in medical sociology, 2, 137-158. Greenwich, CT: JAI Press.

Davis, F. (Ed.) (1963). Passage through crisis: Polio victims and their families. Indianapolis: Bobbs-Merrill.

Desrosières, A. (1993). La raison statistique. Paris: Découverte.

Diagnostic standards and classification of tuberculosis, (1961 edition), New York: National Tuberculosis Association.

Diagnostic standards and classification of tuberculosis, (1955 edition), New York: National Tuberculosis Association.

Engeström, Y. (1987). Learning by expanding: An activity-theoretical approach to developmental research. Helsinki: Orienta-Konsultit Oy.

Epstein, S. (1995). The construction of lay expertise: AIDS activism and the forging of credibility in the reform of clinical trials, Science, technology and human values , 20, 408-437.

Fagot-Largeault, A. (1989). Causes de la mort : histoire naturelle et facteurs de risque.

Paris : Librairie philosophique J. Vrin.

Glaser, B., & Strauss, A. (1965). Time for dying. Chicago: Aldine.

Hacking, I.(1990) The taming of chance. Cambridge: Cambridge University Press.

Hacking, I. (1995) Rewriting the soul: Multiple personality and the sciences of memory. Princeton, NJ: Princeton University Press.

Haraway, D. (1992). The promises of monsters: A regenerative politics for inappropriate/d others. In L. Grossberg, C. Nelson, & P. Treichler, (Eds.), Cultural studies now and in the future, pp. 295-337. London: Routledge.

Kingsland, L. C. III, (Ed.) (1989). Proceedings of the thirteenth annual symposium on computer applications in medical care, November 5-8. Washington DC, IEEE Computer Society Press.

Latour, B. (1993) We have never been modern. Trans. Catherine Porter. Cambridge, MA: Harvard University Press.

Lynch, M. (1995) Laboratory space and the technological complex: An investigation of topical contextures. In Star, S. L. (Ed.) (Pp. 226-256). Ecologies of knowledge: work and politics in science and technology. Albany, NY: SUNY Press.

Mann, T. (1992 [1929]). The magic mountain. Trans. Lowe-Porter, H-T. New York: Modern Library.

Pasveer, B. (1989). Knowledge of shadows: the introduction of X-ray images in medicine, Sociology of health and illness, 11, 360-383.

Porter, R. (1990). English society in the eighteenth century. London: Penguin.

Roth, J. A. (1963). Timetables: Structuring the passage of time in hospital treatment and other careers. Indianapolis: Bobbs-Merrill.

Roy, D. (1959). "Banana time": Job satisfaction and informal interaction. Human organization 18:158-68.

Serres, M. (1980). Le Passage du nord-ouest. Paris : Editions de Minuit.

Sontag, S. (1977). Illness as metaphor. New York: Farrar, Straus & Giroux.

Star, S. L. (Ed.) (1995). Ecologies of knowledge: Work and politics in science and technology. Albany, NY: SUNY Press.

Star, S. L. (1991). Power, technologies and the phenomenology of standards: On being allergic to onions. In Law, J. (Ed.) (Pp. 27-57). A sociology of monsters? Power, technology and the modern world, Sociological Review Monograph. No. 38. Oxford: Basil Blackwell.

Star, S. L. (1989). Regions of the mind: Brain research and the quest for scientific certainty. Stanford: Stanford University Press.

Star, S. L., & Ruhleder, K. (1996). Steps toward an ecology of infrastructure: problems of design and access in large-scale information systems. Information systems research. 7: 111-134.

Strauss, A. (1993). Continual permutations of action. New York: Aldine de Gruyter.

Timmermans, S. (in press). Saving lives or identities? The double dynamic of technoscientific scripts. Social Studies of Science.

Timmermans, S. (1995). Saving lives? A historical and ethnographic study of resuscitation techniques. Ph.D. Dissertation, Department of Sociology, University of Illinois, Urbana-Champaign.

Timmermans, S., Bowker, G. & Star, S.L. (1995). Infrastructure and organizational transformation: Classifying nurses' work. In W. Orlikowski, G. Walsham, M. Jones & J. DeGross (Eds.), Information technology and changes in organizational work.. Proceedings IFIP WG8.2 Conference, Cambridge, England, pp. 344-370. London: Chapman and Hall.

Timmermans, S., & Berg, M. (in press). Standardisation in action: Achieving universalism and localisation in medical protocols. Social Studies of Science.