Lived reality and the multiplicity of norms: a critical tribute to George Canguilhemкод для вставкиСкачать
Economy and Society Lived reality and the multiplicity of norms: a critical tribute to George Canguilhem Canguilhem considered the lived reality of a disease that makes a person visit a doctor with clinical complaints as more important than the de
This article was downloaded by: [University of Leeds] On: 27 April 2014, At: 08:29 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Economy and Society Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/reso20 Lived reality and the multiplicity of norms: a critical tribute to George Canguilhem Annemarie Mol Published online: 28 Jul 2006. To cite this article: Annemarie Mol (1998) Lived reality and the multiplicity of norms: a critical tribute to George Canguilhem, Economy and Society, 27:2-3, 274-284, DOI: 10.1080/03085149800000020 To link to this article: http://dx.doi.org/10.1080/03085149800000020 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the вЂњContentвЂќ) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Lived reality and the multiplicity of norms: a critical tribute to George Canguilhem Annemarie Mol Downloaded by [University of Leeds] at 08:29 27 April 2014 Abstract Canguilhem considered the li\ed realitj of a disease that makes a person visit a doctor with clinical complaints as more important than the de\iance that may be detected in the laboratory. H e also insisted that doing medicine is a technique mobilized to improve life rather than an assemblage of neutral scientific facts. But these two ways of insisting on lived reality have different consequences. In line with the second, I present various ways in which clinical normality and laboratory normality are handled in current day medical practice. I consider where that leaves the first approach of setting standards. T h e multiplicity of normalities detected raises the question of how the various medical normnli~iesrelate. For if they hang together coherentll, medicine, by normalizing, might actively help to order the society of which it forms a part. But what if the various medical normalities contradict each other and inform different orders? Keywords: Canguilhem; norm; normal; clinic; laboratory; diabetes. In m y contribution to this collective unravelling o f t h e dense tissue o f G e o r g e Canguilhem's written work, m y a i m is n o t explanation b u t exploration. T h e questions I ask a n d t h e stories I tell are a p r o d u c t o f reading Canguilhem while investigating c u r r e n t medical practice.' The laboratory and the clinic I n The Normal and the Pathologi~alG e o r g e Canguilhem defended clinical normalit? against t h e n o r m s o f t h e laboratory (Canguilhem 1966). Deviance, h e argued, is n o t a positive fact. T h e difference between a-normal a n d ab-normal Economy and Soc~etyU,lz~?ne27 Numbers 2G3 Map 1998. 271-284 0Routledge 1998 Downloaded by [University of Leeds] at 08:29 27 April 2014 Lzzed renlzty (in(/ the multzplzcztjt of'nosms 275 cannot be established by careful observation of bodily tissues under the microscope. And measuring variables of an isolated organism may yield a lot of knowTledge, but it does not say anything about the question of where health turns into disease. Instead, patholog)~is U liued reality. It is a matter of the relation between the organism and its environment. In this relation one a-normality may turn out to be a good thing for the organism and another may bring along the suffering that marks it off as ab-normality Only when the organism can no longer react creatively to new elements of its surroundings, when it loses its potential to set new norms, does it falter. At that point it loses its order and turns into chaos which, for the organism, implies death. This was overtly framed as an argument against the idea that the natural sciences would soon understand the living world. Canguilhem warned that there is a qualitative difference between the physical and chemical ICLUIS that govern particles and the norms that mark the difference between viable biological order and the chaos that comes with death. Less overtly, but with apt timing, Canguilhem's 1943 thesis was also an argument against those who claimed that they could separate out healthy races from deviant ones scientifically. Canguilhem's analysis is undermining for the scientism which infused most of the eugenics of those days, since it shows that the question of which genes are 'good genes' cannot be answered as if it were a positive fact. For the object of pathology eludes neutral terms and it requires qualification. It is not simply a fact, says Canguilhem, but also a value. It is not so much a subjective value, set by an thinking mind, as a vital value that comes out of the relation between an organism and its surroundings. Canguilhem's argument for the primacy of the clinic over the lab comes in two steps. The first is about the specificity of organisms. Their ordering is marked by their active ability to set norms. This is their lived reality rather than something predictable from the laws of dead nature. Rut there is a second step as well. Canguilhem also inserted the very attempt to know; to ask questions about reality, inside life. He took knowing in order to intervene to be one of the activities of the living. Canguilhem's insistence on liaed realitjt therefore not only made him create thresholds in the Comtean pyramid (between physical and chemical laws and biological norms). He also pointed out loops in the order of the sciences. A physicist who knows everything about the laws of nature, said Canguilhem, is prone to get ill at some point, and to die. Or, another way to frame this, a laboratory may now claim to be in a better position than a doctor in a consulting room to know whether a patient is healthy or ill. But even in those instances where this claim is justified, laboratory knowledge is not basic. For laboratories only ever came into being because at some point in history they were built. And they were built in an attempt to meet the needs of those who came to see their doctor with complaints. These two steps in Canguilhem's argument for the appreciation of lived reality over positive facts each have a different consequence. Canguilhem's insistence on the specificity of organisms is normative: the complaints patients come to see their doctor with are more important than laboratory numbers, and Downloaded by [University of Leeds] at 08:29 27 April 2014 276 Annemarie M o l complaints, therefore, are what medicine should be about. In this same normative vein the lab should be accorded its proper place, that is a place which allows it to support a well-directed clinical practice. However, treating knowledge as a part of lived reality is different. It does not state a norm, but suggests a question. This is the question: what is the place accorded to the laboratory in the lived reality of present-day medicine? How do the numbers of technologically supported diagnostic devices and the complaints uttered in the clinical encounter between patient and physician currently relate? In what follows I will not judge present-day medical practice against Canguilhem's clinical standard, but take the second route of asking the question how clinic and lab relate in medical practice. Not in order to give an exhaustive overview, but to indicate the direction in which we might look for the answer. Or, rather, the answers, in the plural. For this is what I would like us to take seriously: in present-day medicine disease is not diagnosed in a single way. There is, instead, a whole array of well-established relations between 'the clinic' and 'the lab'. The current health-care organization of most Western countries is one that privileges clinical diagnosis. For most diseases, entrance to the health-care system depends on the initiative of patients. People who have no complaints are screened for deviance in only a small number of well-delineated cases. Formerly the most important of these was tuberculosis, now it is cervical carcinoma. There are also screening programmes for new-born babies. But for most of the hundreds of diseases that may plague a person medical detection and diagnosis depend on the initiative of lay people in seeking out medical help. Once people are in a consulting room, however, the number of diseases diagnosed in a clinical way is a lot smaller. A variety of patterns emerges. The clinical findings point to a single diagnosis, but the lab is called in for a confirmation. (It is likely this patient has diabetes, please check blood sugar.) Or: the clinical findings allow for several possible diagnoses and the lab is called in to decide between them. (The patient has not had a period so she may be pregnant, but it is also possible she is just stressed, please do a pregnancy test.) Or: the clinical findings are not conclusive and the lab must help to increase or exclude possibilities. (Patient is dizzy and endlessly tired: please check haemoglobin level, for this might be anaemia, but also do a sedimentation test, for there may be an infection.) Or: the clinical findings give no clue about what the lab discovers when it is used for a routine test. (Some general practitioners routinely measure the blood pressure of all elderly patients whatever the complaints they present. A high blood pressure may thus be found in someone who came in with a sore back or a sore throat.) T h e next important step in a patient's itinerary, then, is the decision about treatment. This need not parallel the diagnosis. For example, someone may have bad arteries, but when operating on this person involves a high risk of increasing the problems he or she faces in daily life, while there is only a small chance of actually improving this life, no operation follows. T h e reason for operating to treat atherosclerosis reflects the complaints patients present, not the small lumen Downloaded by [University of Leeds] at 08:29 27 April 2014 Lived reality and the multiplicity o f norms 277 of an atherosclerotic artery visible on angiographic X-ray images. Or, it is even more complicated. While complaints are the 'reason for treatment' in the leg vessels, for the carotid arteries that feed the brain this may be different. Since a thrombosis here may cause brain damage, the 'reason for treatment' may shift from current complaints to the risk of a cerebro-vascular accident that comes in the absence of treatment.2 The question about how 'the clinic' and 'the laboratory' relate in resent-day medicine thus cannot be answered in general terms. Any attempt to offer a serious answer immediately brings to light an array of different relations. And we are not there yet. For in the examples discussed so far the clinical and the laboratory diagnosis of the diseases were presented as if these were independent of each other. In the intricacies of medical practice, however, they may be intertwined. This happens when it comes to setting standards of normality. For how does one establish laboratory norms? This tends to be done by investigating 'normal' patients. But normal by which standard? Since the laboratory norm is not yet known - it has to be established -the clinical norm is drawn upon. For instance, if one wants to find out whether or not elderly patients have a lower normal H b level than young adults, it is bad practice to measure the H b levels of patients in a hospital and draw a normal curve through the outcomes. Instead, people living at home and reporting no complaints are taken to set a proper standard. It also works the other way around. In order to establish whether or not a given complaint belongs to the clinical pattern of a specific disease or syndrome, it may be necessary to make use of the laboratory. Thus a group of general practitioners' researchers wanted to know whether the isolated complaint of 'tiredness' is enough for a clinical suspicion of anaemia. They measured the H b of patients complaining of tiredness and compared these to a control group. Finding no difference between the Hb-levels of the two groups, they discarded 'tiredness' as a clinical reason serious enough to indicate anaemia.-l The intertwining of clinic and laboratory is, finally, sometimes a matter of a patient's lived reality. This most clearly happens when patients carry their own miniaturized laboratories along with them. One of the oldest of the portable, home-used diagnostic devices is the thermometer. Many people who have had periods of illness with fever have the experience of ending up no longer vaguely feeling 'ill' but instead feeling that they 'have a temperature'. Sometimes they are even able to guess quite accurately what this temperature is.4 Since the 1980s there have also been miniaturized laboratories for patients with diabetes.These allow people to measure their own blood sugar. Blood-sugar levels of clinically normal people fluctuate between 3 and 8 mmol/l. People with diabetes have to try to keep their own blood-sugar levels within this range. If they do not inject any insulin they end up with far higher blood sugar levels some of the time. As it is, they have to find a balance between their food intake, their physical exercise and the insulin they inject themselves with. How to do so? Some people live their bodies in a predominantly clinical way: they say they never use their measurement device because they are able to feel if their blood Downloaded by [University of Leeds] at 08:29 27 April 2014 278 Annemarze Mol sugar is getting too high or too low. Others live in a laboratory mode: they say that they 'don't trust their feelings' and act upon what they find when they measure themselves. And then there are mixtures of these strategies. After measuring blood-sugar levels regularly for a number of months and keeping a diary of these measurements, some people say that they acquire the ability to feel what they are going to measure. Others tell stories about persistent mzsmatches between feeling and measurement: they measure their blood-sugar levels when they do not feel well, but then may have a blood sugar of 7 mmol/l, which is within the normal range. The story is that they may have felt bad, not because they had an 'abnormal' blood-sugar level there and then, but because they had just dropped from, say, 15 mmol/l. T h e measuring thus stops them from eating until they feel well again, but are 'way too high in their sugars'. And then there are people who say that they feel better after they have taken the measurements since the numbers reassure them. In The Normal and the Pathological Canguilhem defended the lived reality of clinical normality over the norms set in the laboratory. That was the first, normative line of his argument. He also argued that knowledge is not only about the world, but in the world as well. Engaging in medicine is a human activity, it is part of our lived reality. In line with that latter argument, I have presented some examples of the ways in which the relation between clinical normality and laboratory normality is lived in present-day medicine. What we end up with, then, is an array of lived medical realities but also a new question. This is the question: given that they relate in many different ways, how should we think about the normative plea that clinical normality should be taken as more important than laboratory normality? I do not know the answer to this question. The primary importance of the clinic might be mobilized as a counterfactual norm by which to judge the various configurations just described. But maybe the thickness of actual practice, in all its intricacies and complexity, resists the gestures of setting normative standards from the outside because such gestures are simply too large. The specificities I have talked about might indicate that norms are necessarily local. But then again, another possibility, maybe adhering to clinical normality is not a norm from outside, and we should not treat it in that way. Perhaps, instead, it is already there, in the lived reality of medical practice, informing the very relations between clinic and lab I have just described. I do not know the answer but wish to stress here that the question is generated by the juxtaposition of the two different ways in which Canguilhem stressed the importance of 'lived reality' - by favouring the clinic over the lab and medicine's technicity over its fact finding. Somehow Canguilhem was able to avoid this juxtaposition. He never explored what his judgements might mean in practice. But today this seems an urgent question. Medical ethicists keep on articulating and legitimating norms. Anthropologists and sociologists give more and more detailed descriptions of hospital life, the surgeries of general practitioners, But where does it all go? How might norms be applied and even of lab~ratories.~ Lzued realzty and the m u l ~ ~ p l z cof z t ~norms 279 to a reality in which they already figure? What becomes of norms if we are serious in saying that they are part of lived reality already? And what should we make of the dazzling plurality of configurations, of the mixtures and interferences of multiple normalities? Downloaded by [University of Leeds] at 08:29 27 April 2014 The social and t h e vital One of the most intriguing essays in The Normal and the Pathological is called 'From the social to the vital'. It was added to the second edition of 1966, and in this essay Canguilhem compares the norms of life with those of society. In both instances 'norm' is a term used to mark the difference between order and chaos. Chaos is normlessness, while there are as many orders as there are norms. Like an organism, Canguilhem explains, a society may be ordered by norms. There is, however, a difference. While for an organism the norms are given with its life and the surroundings in which it lives, for society this is different. Social norms are actively set. They merely rnzmzc organic norms. They are created and they may be altered. This is the point of the analysis: to show that social norms may be altered. Or, there is a second, to show the difference between the way norms order a society and the way in which this is done by state laws. If laws are not obeyed, punishment follows. But with norms this is different: an individual person or entity who does not meet the norms is not punished. Instead, those who meet the norms have a series of advantages. As an illustration Canguilhem mobilizes the norms of grammar. A speaker who is capable of speaking and writing 'correctly' is advantaged in a society of norms. He also mentions the screw thread. There is no inherent reason, built into the screw, why its thread should turn one way rather than the other. But, once there is a norm, a screw thread that lives up to it is a far more useful object than one that turns its own, alternative way. Playing with a Marxist vocabulary Canguilhem goes so far as to state that in the time between 1759 (first use in French of the word 'normal') and 1834 (first use of the word 'normalized', again in French), a normntzve class established itself (in France). This class claimed to lay bare 'the norms of social life' as if these existed out there waiting to be discovered. However, says Canguilhem, it did not discover these norms. Instead it invented them. The normative class actzve!)~set the norms that achieved an ordering force in their society. And it set them up as a system, as a coherent whole. For if a society regulated according to norms mimics the order of the organism, it mimics the organism's coherence, too. This is organization: to have an order that resembles the order of an organism and that hangs together like an organism - through a set of interdependent norms. Canguilhem presents no medical example of normalization, for, in his analysis, biological norms are contrasted with social norms. Social norms are constructs, they are actively set and thus make society mimic life: the very life whose norms are given with the organism-in-its-surroundings; the life medicine deals Downloaded by [University of Leeds] at 08:29 27 April 2014 280 Annemarie Mol with. Implied in this is the argument that the norms of medicine are, or should be, those given with the life of an organism in its surroundings. Not those set for society. But here appears, again, the tension that I noted earlier. Canguilhem insists on the importance of the lived reality of ab-normality but also, simultaneously, on understanding medicine as a part of life. But where does the latter argument lead now? It suggests that we should stop wondering about the grounds of medicine's definitions of normality and the question as to whether these are justified and instead focus on their effects. In doing so it becomes apparent that, even if medical norms were those of the organism, given along with its vitality, actively trying to uphold them through medical care would, from the point of view of the social, be a way of setting norms. A society with a medicine striving after the normality of its people differs from one without such a medical effort - which is what Michel Foucault has shown us in so many ways6 And thus it is his work that turned the term normalization into a word for the way in which modern medicine helps to govern the society of which it forms a crucial part: by ordering; by holding up normality as a norm, a standard, an ideal for each and everybody (every body) to attain. Once the lived reality of acquiring and handling knowledge is taken seriously, the social consequences of operative medical norms become more important than whether these norms are given with the organism or actively set by one social class or another. Doing medicine may well be a specific way to mediate between vital norms and their social consequences. But what are these social consequences? Does striving after normality indeed imply that the society comes to mimic the organism? This, again, does not so much depend on where the norms comefrom as on how they relate. Do the various normalities that inform medical interventions cohere, do they hang together as a system, do they form a tightly knitted whole? Earlier we saw that there is not just a single way of establishing normality in medicine, but there are many. So far, however, I have merely described this variety without considering how the various normalities relate. In order to deal with this question I will now look at a single example a little more closely. It is the example of diabetes. Diabetes is marked by the body's inability to produce its own insulin. This problem cannot be solved. There is no therapy available to correct this condition. Thus the 'normalizing' norms in the life of a patient with diabetes are not those that mark the distinction between the normal and the pathological. Someone with diabetes does not strive after a normal insulin regulation but has another aim: that of maintaining a normal blood-sugar level. As I noted earlier, according to the textbook the normal human blood sugar level lies between 3 mmol/l and 8 mmol/l.' Once diabetes has been diagnosed, a good physician begins by trying to establish the complex, somewhat paradoxical, severe but non-neurotic relation of her patients to this norm. Patients have to try to normalize their blood sugar: it must stay within this normal range. If they let their blood sugar drop too low, they may fall into a coma. If nobody gets them out of that, the ultimate risk is death. If one often suffers Downloaded by [University of Leeds] at 08:29 27 April 2014 Lived reality and the multiplicity o f norms 281 from such states of hypoglycaemia brain damage may occur. But if patients allow their blood sugar to rise too high, they are prone to lots of complications in the long run, including blindness, neuropathy (lack of sensation) and an early onset of atherosclerosis. So patients have to try hard. But they also have to accept that sometimes their blood sugar will drop too low or rise too high. If they do not accept this they are bound to suffer from trying too hard and experiencing repeated disappointments. A normal blood-sugar level is thus a level that does not get either too low or too high. It is maintained by adjusting one's diet, exercise and insulin injections. It is maintained by monitoring one's own blood-sugar level with regular measurements and acting accordingly. The proper way to execute all this, and thus to relate to one's own normality, is by trying hard but not too hard. As a normal person. Thus: what we have here is not a single norm, but at least two: one for blood sugar, one for a person's attitude towards her own treatment. How do these norms relate? They happen to be interdependent: a normal person is in the best position to maintain normal blood-sugar levels. It is through such an interdependence that norms may hang together and form coherent wholes. However, this is not all there is to say about diabetes because the norms do not always fit so nicely together. Take the situation of M r Hanssen, a 35-year-old owner of a small enterprise. His special task in the enterprise is to do the marketing and the contacts with the clients. This means that he regularly has to go and see these clients and meet potential new clients. The enterprise is far too small for it to be able to pay for a driver. This is no problem for M r Hanssen, for he likes the driving. Since he lives in the Netherlands he is allowed to do so, even though he was diagnosed as diabetic four years ago, for, after a few months with adjustment problems, M r Hanssen has been able to regulate his blood-sugar level. He takes measurements of his own blood-sugar levels five times a day one day a week, and then whenever he feels odd. He shows these numbers to his physician every few months. His physician praises him for the way he manages to keep his blood sugar within the normal range, despite the irregularity of his responsible work. Now on a sad day in a busy period in March, Mr. Hanssen is stopped by a policeman, who suspects him of being drunk. He is not drunk, his blood sugar is too low. Since the latter is at least as risky as the former for both himself and other people on the road, Mr Hanssen is temporarily forbidden to drive. First, he has to be medically tested again. What to do? Here is the dilemma that M r Hanssen faces. H e may set the blood-sugar level he is aiming for slightly higher. So far, he has tried to stick to a good 6 mmol/l as a mean value. But, since this has once caused him to drop to the 2 mmol/l that caused his driving problems, he may decide to set his target somewhat higher - at 8 mmol/l, for instance. Doing so would allow him to drive again. He could then keep his enterprise going the way it does. But with 8 as a target, his values would be above 10 a lot more often than they have been while 6 was his target. And that, he knows, is a longterm risk. He would increase his chances, of, say, going blind at 50 or having a fatal heart attack at 53. What to do? Downloaded by [University of Leeds] at 08:29 27 April 2014 282 Annemarie Mol T h e coherence has vanished. While, earlier, the normal person and her normal blood-sugar range fitted so beautifully, now there are two possible normal values to strive after: 6 or 8. One of these is linked up with a higher risk of hypoglycaemia and the other with a higher risk of long-term complications. They have a relation to Mr Hanssen as a person, too, but not to how normal he is, but to what will become of him. For sticking to one norm would force M r Hanssen to stop driving, which would mean he would have to change completely a working life he both enjoys and which pays him (and those he works with) well; while sticking to the other would allow him to continue to do his work but might have major consequences for the kind of life he is likely to lead in twenty years' time - if indeed by then he is still living. M r Hanssen faces the question of what to do. Which complications to 'accept': a small risk of brain damage or a larger one of blindness, neuropathy and/or atherosclerosis? To prevent death due to a car accident he will be forbidden to drive if he tends to have hypoglycaemias, but, if he avoids these so as to be able to drive, he risks dying of an early heart attack before he is 50. So turning himself into a normal person is not enough for M r Hanssen. Instead, he faces the question of what kind of life he will lead: one without a job, or (if he can find it) one with a calm indoor job, or one as the travelling marketing man of his own enterprise? Such questions do not have a single answer. There is not one normality at play here, rather there are two. The effects of going with one norm, a bloodsugar level of 6 mmol/l, differ from those of going with the other norm, a bloodsugar level of 8 mmol/l. Instead of hanging together these possibilities exclude each other. They clash. M r Hanssen cannot live up to both norms, for there is a divergence between them. What happens with such divergences? In current medicine, or so it seems to me, they are handled simultaneously in two very different ways. On the one hand, there are instances where dilemmas such as these are treated as problems to be solved, questions that require a proper answer. Clinical trials that must help to judge the relative merits of various treatments are designed in order to forge a choice between them. Thus they order the effects of such treatments on a single scale. They put them in a hierarchy derived from a few physical parameters and some variant of a quality of lzfe scale. They try to achieve singularity and to set the best treatment as standard. On the other hand, however, there is also a rather different move. This is to give up trying to solve practical dilemmas with the techniques available to the multi-disciplinary research team. Instead of being ranged in a vertical hierarchy, the dilemmas are laid out side by side on a horizontal line. In this way the various treatment possibilities are turned into options. In the situation of M r Hanssen this is what happened. His physician sat down to talk with him. He told him that medicine is able to inform him about the likely effects of setting 6 or rather 8 mmol/l as a norm for his blood-sugar level, but cannot tell him what to do. It cannot decide what is best, cannot make the choice for him. 'I don't know what you should do, M r Hanssen, we can of course talk about it, but when it comes to it, it is up to you.' Downloaded by [University of Leeds] at 08:29 27 April 2014 Liaed reality and the multiplzcit~)~ o f norms 283 If there is no single norm to orient normalization, if instead there are several plausible but different norms which do not cohere but coexist in tension, soczety does not stay the same. For, if the normalities that mark its order d o not hang together as a system, society does not mimic the organism. There is not one order. More likely there are various rnodes o f ordering that clash here, diffract each other there and interfere with one another somewhere else.8 I n medicine normalization as a mode of ordering comes to coexist with something that rather resembles the market. This is not only because of the fact that M r Hanssen's choices have economic aspects but also because he is presented two goods by the doctor and h e has -how to call it - the right, the obligation, the task - t o choose between them. And this is true even if M r Hanssen may feel he has no choice, for what will drive him one way or the other are the intricate specificities of his life history rather than a singular medical norm. If multiple normalities come to figure as competing goods, normalization is no longer capable of providing a coherent order. T h u s terms like social system and social order lose their former power. A term like social slruggle is not appropriate either since the tensions between modes of ordering follow erratic lines and r u n right through the constituent parts of the non-whole. This non-singular society is complex: a term that indicates that accounts cannot be closed without leaving remainders, and that what is added u p forms no sum total; that adding u p may not, indeed, be a good metaphor at all. But to mobilize a term like 'complexity' is not to reach a satisfying conclusion. I t only marks a new beginning. I t requires us to investigate and reflect upon societ11all over again. O r does it? Is this what it does? For, who knows, it may also be that 'society' is over. N o t because there are only individuals and families left, but rather because the fleshy, financial, metal and fluid matters that constitute it, form tense, multivocal, non-organic patterns for which the words are still to be invented. Utrecht Notes 1 They are also the product of the work of many. I would like especially to thank here Peter van Lieshout with whom I wrote (in Dutch) about Canguilhem and social theory more than ten years ago, Dick Willems for our discussions about medical technology, several anonymous patients with diabetes who were willing to talk about their lives and Claar Parlevliet for conducting interviews with them, Harold deValk and Willem Erkelens for their hospitality and information, and John Law for his encouragement and for correcting my English. 2 The intricacies of the relation between the object of diagnosis and the object of treatment form a topic for discussion in medical anthropology and sociolog>.For this and the example of atherosclerosis of the leg vessels, see Mol and Elsman (1996). 3 This co-consitution in tension of norms for anaemia is more extensi~elydescribed in Mol and Berg (1994). 4 The intertwining between a patient's body and the technology used to diagnose and treat is analysed for the case of asthma by nick Willems, see, for example, U'illems (1992). 284 Annemarie Mol 5 Situating him in French intellectual history, I would argue that Bruno Latour, however much he has attacked the normative first line of Canguilhem's work, has simultanuously drawn upon the second anthropological line of it by studying science as a human activity; see especially Latour (1984). 6 Canguilhem was Foucault's thesis supervisor during the time he must have been working on 'From the social to the vital'. This makes it is hard to tell which of them invented the term normalization. Their shared dislike of the obsession of historians of ideas with inventors and matters of authorship makes it easy to care more about what is written than about who disowned whom. For Foucault's use of 'normalization', see, for example, Foucault (1975). 7 That is, the precise numbers that indicate the 'normal range' differ slightly from one source to another. These numbers come from a widely used British textbook (Souhami and Moxam 1990). 8 For the term modes of ordering and a far more extensive exploration of what it may mean to talk about 'society', see Law (1994). Downloaded by [University of Leeds] at 08:29 27 April 2014 References Canguilhem, G. (1966) Le Normal et le Pathologique, Paris: PUE Foucault, M. (1975) Surveiller etpunir: Naissance de la prison, Paris: Editions Gallimard. Latour, B. (1984) Les Micmbes: Guerre et paix/Irrt!ductions, Paris: Mktaille. Law, J. (1994) Organizing Modernity, Oxford: Blackwell. Mol, A. and Berg, M. (1994) 'Principles and practices of medicine: the coexistence of various anemias', Culturt, Medicine and Psychiatry 18: 247-65. Mol, A. and Elsman, B. (1996) 'Detecting disease and designing diagnosis: Duplex and the diagnosis of diseased leg vessels', Sociology of Health and Illness 18: 609-3 1. Souhami, R. L. and Moxham, J. (1990) Textbook of Medicine, Edinburgh: Churchill Livingstone. Willems, D. (1992) 'Susan's breathlessness - the construction of professionals and laypersons', in Jens Lachmund and Gunnar Stollbert (eds) The Social Construction of Illness, Stuttgart: Franz Steiner Verlag Stuttgart.