Repetition strain injury in Australia: medical knowledge, social movement, and de facto partisanship

Gabriele Bammer and Brian Martin 

Social Problems, Vol. 39, No. 3, August 1992, pp. 219-237. The version here was extensively sub-edited by Social Problems prior to publication. Any quotations should be checked with the Social Problems text.


Go to

Brian Martin's publications on scientific controversies

Brian Martin's publications

Brian Martin's website


 

Abstract

One of the most vehement debates over medical knowledge in Australia in the 1980s concerned "repetition strain injury" or RSI. We analyze the Australian RSI experience using two contrasting approaches: the sociology of medical knowledge and social problems as social movements. Each approach tends to delegitimate the position of those who claim that RSI is work-related and has an organic basis. A key factor leading to the de facto partisanship associated with each approach is the choice to analyze the Australian RSI experience in the first place. The de facto partisanship associated with the choice of a framework of analysis and the choice of an issue to study is an important aspect of understanding social problems which has been largely ignored until now.

 

In recent debates about the appropriate approach to the study of social problems, it is possible to identify three positions. First is the "objectivist" approach, in which the existence of social problems is a consequence of real, knowable social conditions. Objectivist analysts do not attempt to provide a social explanation of social reality, since this reality is thought to be known as objective fact.

A second approach is that of social constructionism (Spector and Kitsuse, 1977). In the strict version of constructionism, no assumptions are made about objective realities. Rather, the analyst studies the social activities of actors, especially the activities by which the actors define certain things as "social problems" - a process called claims-making.

A third, intermediate approach can be called "contextual constructionism" (Best, 1989b). It acknowledges making some claims about social reality, and uses these to provide a framework in which to analyze the social processes of claims-making. This approach has both objectivist and constructionist elements.

This categorization of approaches is of course a simplification of the diversity of approaches found in the social problems field, but it will suffice for the purpose here of informing our own study. The important point is that proponents of each approach claim for it a methodological superiority linked to its assumptions about the reality of what is being studied.

Whatever approach they adopt, social problems analysts typically present themselves as social scientists, not as partisans for a particular viewpoint. Our concern here is the process by which analysts may prejudge their conclusions by their choice of analytic framework. According to constructionists, objectivists prejudge their conclusions through their assumptions about social reality, because they do not attempt a social explanation of this reality. Woolgar and Pawluch (1985a) extended this critique to many ostensibly constructionist analyses, pointing out that while analysts subjected some claims to scrutiny, others were unexamined and hence essentially treated as objective, a process they called "ontological gerrymandering". In both the objectivist and contextual constructionist analyses, the analysis imputes greater credibility to the views treated as objective, and thus often provides de facto - and sometimes open - support for those views.

This sort of partisanship is not overcome by a strict constructionist or relativist analysis. Such an analysis, by treating all claims as subject to social explanation, undermines the views backed by greater scientific authority much more than views critical of the orthodoxy and thus is likely to serve the interests of the latter (Scott, Richards, and Martin, 1990).

By the expression "de facto partisanship", we refer to the way in which choice of a framework for analysis tends to lead to conclusions favoring a particular point of view. De facto partisanship does not depend on conscious intent and can, indeed, be contrary to the intent of the analyst. An assessment of de facto partisanship must be made sociologically, by examining the use made of social problems accounts.

An assumption in all the approaches to studying social problems is that the analysis is carried out independently of the struggles involved with the social problem. But this assumption is untenable, at least for controversial, contemporary issues. Whether it is intended or not, analysts and their work may be taken up by partisans in the debate being studied. Indeed, a range of claims-makers may try to "capture" analysts to serve their own purposes. This process of attempted capture cannot be avoided by the analyst and undermines any claims to neutrality (Hess, forthcoming; Scott, Richards, and Martin, 1990).

Our aim in this paper is to explore the de facto partisanship associated with the choice of a framework for social analysis using a case study. From the above discussion, it would be unsurprising to find that a single approach exhibited de facto partisanship. Therefore, to probe more deeply into this process, we deploy two frameworks of analysis: the sociology of medical knowledge and social problems as social movements.

Our chosen case study is "repetition strain injury" (RSI), which became a controversial social issue in Australia in the 1980s. There was a dramatic increase in the number of reported cases [1], which became responsible for a large fraction of payments for workers' compensation. The symptoms reported generally affect the neck, shoulders, and/or upper limbs, and include pain, tenderness, loss of strength, fatigue, and lack of coordination.

A major struggle occurred between those supporting and those opposing the recognition of the cases as real work-related injuries. While there has been some debate about the reported symptoms themselves, the major focus of contention has been the meaning given to the symptoms and their causes.

As will be described later, the phenomenon of RSI has been explained at the level of individuals in terms of organic injury, malingering, compensation neurosis, conversion hysteria, normal fatigue, and social iatrogenesis. Each of these explanations of RSI cases has been linked to a more general explanation of the rise of RSI as a social problem in Australia in the 1980s: proponents of the organic injury explanation prefer an explanation in terms of a hidden pattern of injuries, exacerbated by economic, social, and work changes (especially technological innovation), finally achieving social recognition; critics prefer an explanation in terms of a social epidemic of cases triggered by reporting, availability of compensation, and faulty diagnosis which encouraged lying (malingering) or psychosomatic manifestations.

Building on these explanations made by RSI partisans, social scientists have added another layer of explanations. It is readily noticeable that the explanations offered by social scientists have reflected their particular expertise and interests. For example, Helen Meekosha (1986; Meekosha and Jakubowicz, 1986), a feminist, has focused on the role of patriarchy in RSI; Wayne Hall and Louise Morrow (1988), working in psychiatry and psychology, have drawn on causal attribution; Evan Willis (1986; see also 1983, 1989) and Andrew Hopkins (1989), sociologists, have used the sociology of medical knowledge; Merrelyn Emery (1988) and Trevor Williams (1985), supporters of industrial democracy, have focused on the role of work organization in RSI. There is no simple and automatic relation here, but a tendency is obvious for analysts to use frameworks compatible with their professional field and personal commitments. (Of course, we do not suggest that the views of these writers are encompassed by our short descriptive phrases.)

There is nothing very surprising in finding a link between analysts' expertise and interests and their choice of conceptual frameworks. A closer inspection, though, reveals some intriguing discrepancies. Some analysts who have chosen a constructionist approach - such as Willis (1986) - are quite clearly in sympathy with workers with RSI, yet their work has been cited by those who dispute that RSI is an organic, work-related injury. This raises the question of whether it is possible to undertake a constructionist analysis of RSI and yet not provide de facto support for the critics of RSI. Is the use of the analyst's work by RSI critics a result of the choice of analytic framework, or is it somehow linked to the study of RSI in the Australian context, or both?

In the following sections, we begin with an analysis of RSI using the framework of the sociology of medical knowledge. We then analyze the de facto partisanship associated with the sociology of RSI knowledge. After this, we repeat the process with an analysis of RSI using the framework of social problems as social movements. The final section brings together some ideas about de facto partisanship.

Our presentation thus has three levels. The most obvious is an account of RSI as a social problem [2]. The second level to our presentation is applications, to the same issue, of the frameworks of the sociology of medical knowledge and of social problems as social movements. The third level is an investigation into these frameworks to see where the de facto partisanship associated with their use arises. The latter investigation suggests insights into some characteristics of de facto partisanship.

THE SOCIOLOGY OF MEDICAL KNOWLEDGE

The sociology of scientific knowledge is an attempt to analyze scientific knowledge in a manner similar to other social phenomena (Barnes, 1974; Bloor, 1976; Mulkay, 1979). The special status of scientific knowledge - namely an alleged correspondence with the realities of "nature" - is rejected. Instead, scientific knowledge is treated like other belief systems, such as religion. The influence of social structures, funding, professional vested interests, and the micropolitics of the laboratory on the form and content of what is accepted as scientific knowledge are all topics for investigation.

In RSI, the focus is on a condition disputed within medicine. The sociology of scientific knowledge becomes the sociology of medical knowledge, which looks at the role of social factors in the creation and negotiation of knowledge claims, including beliefs about health and disease, the social organization of medical care, and the distribution of power in society (Figlio, 1978, 1982; Gubrium, 1987; Richards, 1988, 1991; Wright, 1980; for a critique see Bury, 1986). This is a challenge to the conventional belief that the physical realities of health and disease are revealed unproblematically by clinical examinations, supplemented by biochemical and other scientific methods for assessment of evidence. From the point of view of the sociology of medical knowledge, the orthodox view is seen as a convenient gloss on actual medical practice, serving to provide a set of meanings which unify the medical profession and give it status.

Following the prescription of the sociology of scientific knowledge outlined by Collins (1981), a program for the sociology of medical knowledge can involve three stages: (1) demonstrating the "interpretive flexibility" of medical findings, namely that they are open to more than one interpretation; (2) describing the social processes, known as closure mechanisms, that allow medical controversies to be terminated, namely the processes that limit interpretive flexibility; (3) showing links between closure mechanisms and the wider social setting.

In the following, we will describe the claims and counterclaims about RSI, thereby demonstrating the interpretive flexibility involved. In fact, the critics of the standard medical view have seen it as their task to demonstrate this flexibility. The second stage of describing closure mechanisms must be more tentative, since the debate in Australia is not yet closed. The final stage, relating closure mechanisms to social structure, has only rarely been carried through in case studies. We offer some observations on the final stage, in so far as it relates to our analysis.

THE SOCIOLOGY OF RSI KNOWLEDGE

Until about 1988, the dominant medical position in the Australian debate was that the term RSI covers a group of organic injuries caused by either rapid repetitive movements, less frequent but more forceful movements, static load, or a combination of these. Some of the cases are diagnosed as relatively well-defined entities such as tenosynovitis and epicondylitis, whereas others are more diffuse syndromes. The underlying pathology for the injuries is a matter of some contention, but is thought to involve one or more of muscles, nerves, and tendons.

RSI is usually graded into several stages according to its severity. The earliest stages are characterized by no physical signs, with pain and tenderness going away at night or on days off work. At the most severe stages, physical signs are present and symptoms persist even during rest. It is commonly thought that individuals can progress through the stages if they persist in the causative activities. At the early stages, RSI is thought to be reversible through modification of work activities, rest breaks and exercises. At the severe stages, no treatment seems to offer any real solution, except perhaps complete avoidance of any activity that causes pain (Brown, Nolan, and Faithfull, 1984; Champion et al., 1986; Ferguson, 1984; Fry, 1986; McPhee, 1982; Quintner, 1989; Stone, 1983; for a survey see Bammer and Martin, 1988).

This view is very much tied to orthodox medical approaches to disease and injury. RSI is interpreted as an organic problem caused by activities that induce injury. It can be diagnosed through the regularity of symptoms and, in severe cases, through clinical signs. Although some practitioners suggest a role for psychosocial factors, these are interpreted in the context of the main focus, which is on the body and injury to it. With this focus, the problem is brought into line with standard diagnoses of related injuries such as muscle tears, as well as the wider array of diseases.

We refer to an approach to disease or injury using orthodox medical concepts and techniques as following a medical model. The view that RSI is a work-related organic injury we refer to as the "standard medical view"; it is one possible medical model of RSI. The critics of the standard medical view are also critics of the recognition of a unified entity called RSI; for convenience we refer to them simply as "the critics". The critics have adopted a variety of interrelated positions, which we consider briefly: that people with RSI are malingerers; that they have a form of compensation neurosis; that they have a form of conversion hysteria; that they are suffering "normal" aches and pains due to reversible fatigue; and that they are encouraged by doctors and others to become patients with pain, a process that has been called social iatrogenesis [3]. Some of these are alternative medical (psychiatric) models whereas others are nonmedical models.

Some people, including employers, workers and doctors, think that many or most people claiming to suffer from RSI are faking their symptoms in order to obtain time off work or compensation benefits. This highly derogatory interpretation seldom makes its way into print (but see Bloch, 1984; Ireland, 1986; Scarf and Wilcox, 1984).

Unlike malingerers, those said to be suffering from compensation neurosis are considered to genuinely experience pain and other symptoms. Their problems may begin with a real injury, but psychological mechanisms generate disproportionate disability and delayed recovery. The main mechanism cited is an unconscious desire for secondary gain, such as financial rewards, invalid status, or escape from work (Bloch, 1984; Rush, 1984).

A similar explanation holds that RSI is a form of conversion disorder, in which there is no initial injury at all. Pain and disability, according to this explanation, result from emotional disturbance or unresolved psychological conflict converted into perceived symptoms, allowing an escape from the psychological problem (Black, 1987; Cleland, 1987; Ireland, 1986; Lucire, 1986a, 1986b).

Another explanation is that the reported symptoms are normal aches and pains, usually due to simple fatigue, with no underlying injury. The RSI problem is considered to be a rash of reporting of what is always present anyway. The solution is rest and ergonomic changes at work (Brooks, 1986; Hadler, 1986).

Closely related to earlier explanations, especially that of simple fatigue, is the "pain-patient" explanation, in which people, experiencing normal pain at work, are encouraged by doctors and others (trade unionists, co-workers, etc.) to become patients with pain. Psychosocial or economic incentives help people to define themselves as patients, or to be so defined by others. This pattern can be described as "social iatrogenesis" (Bell, 1989; Cleland, 1987; Spillane and Deves, 1987).

The critics of RSI have made a range of claims against the standard medical view, notably that there are no clinical signs or identifiable underlying pathology, that there is no reliable pattern of symptoms, that this is an "Australian disease", that there is no effective treatment, that a causal link to work is not clear, that preventive strategies are not effective and that epidemics are usually caused by viruses or psychogenic factors (Brooks, 1986; Cleland, 1987; Hadler, 1986; Ireland, 1986). To illustrate the debate, we briefly discuss the first three of these criticisms; for others see Bammer and Martin (1988).

It is important here to recognize the distinction, central to the medical model, between "signs", which are organic changes in the body that can be observed, and "symptoms", which are the sensations reported by the patient. Medical science normally expects signs indicative of underlying pathology to be present if there is a "real" disease or injury. Symptoms are considered valuable for diagnosis and are supposed to be verified by assessment of signs.

Supporters of the standard medical view accept that there are usually no signs for RSI in early stages; there is no unified position on the signs in the later stages. The critics have repeatedly homed in on this weakness to suggest that without objective signs, the case for organic injury is insufficient. A few critics go further and say that some of those signs which do occur may be generated by psychological mechanisms (Lucire, 1986a, 1986b).

The crucial role of signs and associated pathology in the debate over RSI depends on acceptance of a one-dimensional medical model. Since advances in medical science may in the future result in detection of organic changes that are currently invisible or unrecognized, the absence of signs is not a definitive argument. In addition, people claiming to suffer from other conditions such as migraine headaches are usually accepted as genuine even though there are seldom clinical signs linked to the symptom of pain. The acceptance of symptoms without signs seems to depend on a range of social factors, such as the status of those reporting the symptoms.

The critics have also argued that the symptoms of RSI do not make clinical sense, in that they are diffuse, vary from patient to patient, do not relate sensibly to conceivable sites of injury, do not fit with those objective signs that do exist, and do not fit any pattern recognizable as due to a pathology (Brooks, 1986; Cleland, 1987; Hadler, 1986; Ireland, 1986). For example, some people with RSI, originally reporting symptoms in a single hand or arm due to repetitive motions at work, later report symptoms developing in the other limb during time spent away from the job. (Whether or not this results from compensatory activity by the other limb used for housework and other tasks is a point of dispute.) Again, the critics appeal to an image of a standard injury or disease in which an underlying pathology results in symptoms in a regular and predictable fashion. This picture omits the many strange and irregular patterns of symptoms of well-recognized diseases.

Another criticism of RSI is that it is an "Australian disease", namely that it is unknown or rare outside Australia, or at least that there has been no "epidemic" of it elsewhere. Mentioned by only a few of the medical critics (Awerbuch, 1985; Bell, 1986; Brooks, 1986; Morgan, 1986; Sharrod, 1985), this view is commonly raised among the media and general public. The assumption is that an organic disorder would develop in a similar fashion in all countries with similar technologies and work organization, and that reporting of the symptoms would follow a similar pattern. Therefore the rash of reports in Australia is better explained by any one of the alternative views. This view again follows the medical model, with the extension that organic injuries are reported and assessed more or less independently of social arrangements.

Proponents of the standard medical view can cite many studies over many decades and in many countries of pain and disability in workers' hands, arms, necks, and shoulders as a result of physical stresses on the job (Bammer, 1987a, 1990c; McDermott, 1986; Quintner, 1989; Task Force, 1985; Wallace and Buckle, 1987). Critics seldom mention these studies. The few who do mention them either dismiss them or interpret them as supporting their own case (Bell, 1986).

This short survey illustrates the methods of the critics of the standard medical view. The critics have shown what would otherwise be ignored (or left to sociologists), namely how the diverse symptoms and the social phenomenon of RSI have been interpreted and organized into a traditional medical injury model, with the inadequacies and loose ends being dropped along the way. The supporters of the standard medical view use what has been called the "constitutive repertoire" (Collins and Pinch, 1979; Gilbert and Mulkay, 1982; Mulkay and Gilbert, 1982), namely the language of medical science, to describe RSI, whereas the critics are more likely to use the "contingent repertoire", which highlights "non-scientific" factors, such as reference to malingerers and to an "Australian disease".

The arguments of the critics are quite vulnerable if their demands for objective signs and predictability of treatment are applied to their own explanations (Bammer and Martin, 1988; Foster and Fry, 1988; Mullaly and Grigg, 1988). For example, what are the clinical signs for compensation neurosis? Yet, although the critics have written many articles and letters to journals, for the most part these have not been answered and refuted by the supporters of the standard view, who have largely rested on their medical credentials and their studies of RSI which, therefore, can be considered to be their principal means for attempting to close the debate.

By contrast, the critics' initial efforts were directed towards opening the debate by demonstrating interpretive flexibility. Part of their argument was that all discussion of the problem had to cease in order to stop recognition of what they saw as a "nonproblem". This strategy succeeded when the Medical Journal of Australia - the main forum for technical discussion and debate - in 1988 terminated publication of articles and letters on RSI for over a year.

Thus far we have analyzed the disputes over the nature of RSI by implicitly treating evidence and arguments as resources used by the various actors to promote their favored explanations. From this basis, there are various ways to proceed to the third stage of dealing with the wider social dynamics associated with RSI.

One way is by using the concept of interest. The connection of arguments to social interests is straightforward at one level. Those promoting the standard medical view have legitimated the claims of people with symptoms of pain and disability as organic, work-related and hence compensable. Their arguments have been taken up by people with RSI, trade unions, and various support groups to press for improved workers' compensation and ergonomic changes, including improving furniture, equipment, and the organization of work.

The arguments of those criticizing the standard medical view have been taken up by employers and insurance companies seeking to deny compensation claims. Some proponents of the alternative explanations, notably Mark Awerbuch, Peter Brooks, Damien Ireland, and Yolande Lucire, have been prominent in testifying for employers and insurance companies against RSI claims by employees (Campbell, 1988).

While the concept of interest has been widely used in the sociology of knowledge, the method for attributing interests is not inherent in the analysis of knowledge claims, but rather relies on some theoretical assessment, explicit or implicit, of the dynamics of society (Barnes, 1981; MacKenzie, 1981, 1984; Woolgar, 1981; Yearley, 1982).

A different way to proceed is to attempt to build up an understanding of society from observations of behavior at the micro level, as proposed by Bruno Latour (1987) and others. Yet another approach is to draw on already-existing concepts, such as social class, patriarchy, and professions, from various bodies of theory (Russell, 1986).

This diversity of options for undertaking Collins' (1981) third stage of linking closure mechanisms with the wider social setting suggests that, while the sociology of medical knowledge, as a theoretical framework, is well equipped to deal with micro-struggles over knowledge, it is poorly defined when it comes to making links with the wider dynamics of society. This is the reason why we have kept our analysis using the sociology of medical knowledge mainly at the level of knowledge claims, a restriction that does not hinder an assessment of de facto partisanship.

DE FACTO PARTISANSHIP (I): SOCIOLOGY OF RSI KNOWLEDGE

The sociology of medical knowledge is founded on the deconstruction of medical knowledge, thereby opening the possibility for social explanations of the origins, development, and deployment of medical knowledge. This program of analysis involves no overt assessment or moral judgement of patients, physicians, or others. Yet, applying the sociology of medical knowledge leads to a de facto intervention into medical debates, which is very apparent in the case of RSI. The critics of RSI have as their first task the deconstruction of claims that RSI is an organic condition. Appropriately, RSI critics occasionally refer to the literature on the social construction of reality. For example, Lucire (1986a) cites the classic social constructionist book by Berger and Luckmann (1966); Bell (1989) cites Willis (1986), who is sympathetic to the cause of workers, and also cites two leading critics of conventional medicine, Illich and Zola, to support his case.

In principle, the process of deconstruction can just as readily be applied to the RSI critics, since they too ground much of what they claim in traditional medical models. But, in practice, the sociology of medical knowledge has been selectively useful to those questioning medical orthodoxy.

What exactly is it that makes the sociology of medical knowledge provide de facto support to the critics of the standard medical view? A social analysis with a objectivist view of disease and medical authority will attempt to give social explanations only for dissenters from orthodoxy, and provides de facto (and sometimes overt) support for the medical orthodoxy. By contrast, the relativist sociology of medical knowledge - when deployed in a social environment dominated by objectivist assumptions - serves to undermine medical orthodoxy because most observers consider that a social explanation for a phenomenon gives it less credibility than one grounded in alleged physical or biological reality.

The de facto partisanship of the sociology of medical knowledge thus rests on the relation of its symmetrical method to the asymmetry of credibility of knowledge claims (and power) in the medical community and the wider society. If the medical orthodoxy were toppled and replaced by one of its challengers, the sociology of medical knowledge would become threatening to the new orthodoxy. The de facto support provided by the sociology of medical knowledge therefore depends on the issue to which it is applied, and the time and place. Applying it to RSI in Australia in the 1980s means undermining the standard medical view that RSI is organic and work-related.

Consider, by contrast, applying the sociology of medical knowledge to RSI in the United States, where RSI has mostly been invisible until recently. The issue has gained increasing attention in the late 1980s, but there has been no debate in US medical journals like the one in the Medical Journal of Australia. Because there has been relatively little debate in medical circles and the general public in the US, social scientists have not been drawn to analyze the issue as they have in Australia. The sociology of medical knowledge is much less likely to be applied when there is no publicly visible debate. This points to an application bias: when there is a dominant position backed by medical authorities, the deconstructionist sociology of medical knowledge will provide de facto support for challengers. But when there is no debate at all, such a relativist social analysis is less likely to be made in the first place.

SOCIAL PROBLEMS AS SOCIAL MOVEMENTS

Of the approaches to the study of social problems, the one closest to the sociology of scientific knowledge is the definitional or social constructionist approach whose program is set out by Spector and Kitsuse (1977; see also Best, 1989a; Gusfield, 1981; Hilgartner and Bosk, 1988; Schneider, 1985; Schneider and Kitsuse, 1984). When relevant actors define something as a social problem, then it can be transformed into one. This process can involve a range of activities, including categorizations by doctors, reports by journalists, policy statements by governments, public meetings by citizen groups, and studies by social scientists.

Both the sociology of scientific knowledge and the definitional approach to social problems deal with the processes by which categorizations of reality are made. The former usually concentrates on the social construction of knowledge claims among specialists, whereas the latter typically emphasizes the wider processes by which something is defined as a social problem.

Much of the work done under the aegis of the definitional approach retains an overt or covert commitment to the reality of social conditions. Hence, the work is not totally constructionist, but rather applies constructionist analysis selectively, Woolgar and Pawluch's (1985a) "ontological gerrymandering". Best (1989b) defends this approach, which he terms "contextual constructionism". By contrast, Ibarra and Kitsuse (forthcoming) argue for a "strict" constructionism. (See also Hazelrigg, 1986; Pfohl, 1985; Woolgar and Pawluch, 1985b.)

Because the strict constructionist approach shares with the sociology of scientific knowledge a commitment to analyzing all sides to the struggle using the same tools, the comments about de facto partisanship made earlier about the sociology of medical knowledge apply also to the strict constructionist approach. Therefore, in order to explore a contrasting theoretical perspective, we have chosen the perspective of social problems as social movements. Mauss (1975, 1989), who champions this approach, sees social problems as inseparable from social movements, because the characteristics of social problems are typically those of social movements. This includes subjective definitions of reality, the formation of interest groups and their respective definitions of reality, and efforts to mobilize public opinion. In a social movement, there are several levels of participants, including the sympathetic public, the active membership, and the principal leaders and organizations. The movement undergoes a natural history dependent on its interaction with the surrounding society, typically including incipiency, coalescence, institutionalization, fragmentation, and demise. Its legacy may include residues and redefinitions at the levels of popular culture, norms, and laws.

Troyer (1989), in a comparison between constructionist and social movements approaches, points out a number of similarities and differences and argues that it is not useful to say that social movements and social problems are the same thing. However, our main purpose here is not this debate but rather an examination of de facto partisanship on RSI. Hence, setting aside the theoretical limitations of the social-problems-as-social-movements perspective, we apply it to the RSI issue. To extend the contrast with the sociology of medical knowledge, it is useful to adopt for this exposition the objectivist assumption that RSI is an organic, work-related injury. Afterwards, we use this application to probe the sources of de facto partisanship in the approach.

RSI AS A SOCIAL MOVEMENT [4]

In industrialized countries, pain and disability associated with work have long existed in many occupational groups but received little public attention. A large fraction of workers who carry out monotonous physical motions with little respite are manual workers. The subjugation of manual workers to imposed conditions, often unpleasant, reflects their weak occupational position vis-a-vis employers. This same weakness is associated with their difficulty in effectively presenting claims about occupational injuries. Relatively few manual workers - especially those from ethnic minorities - have the skills, confidence, and personal connections required to enter the middle-class arenas in order to contest medical evaluations, undertake legal actions, lobby politicians, or organize media coverage.

Ethnic and class differences are compounded by gender differences in the work force. Occupations which put male manual workers at high risk of developing RSI are welding, meat processing, and car assembly; for female manual workers, key areas are assembly-line electronics, food processing and packaging, and the garment industry. The gender division of labour is central to the social geography of occupational health, but the special health problems of manual workers of either gender have received relatively little attention.

Several conditions provided the basis for an Australian RSI movement in the 1980s. Until the 1970s, Australia experienced low unemployment and stable economic growth. Low unemployment meant that workers could easily change jobs. Many of those experiencing pain or injury, or suspecting incipient physical problems, found it easier to change jobs than to apply for workers' compensation. Increases in unemployment and inflation in the 1970s led many workers to stay in jobs they would have left in previous years and to put up with conditions they might once have rejected.

In the rapid industrial restructuring of the 1970s, new technologies were introduced ostensibly in order to maintain international competitiveness, while unemployment weakened the capacity of trade unions to defend traditional work practices. For many workers, the changes meant intensification of work rates, thus appearing to provide the conditions for greater physical stresses on the body. This led to an increased number of people experiencing pain and disability.

While the process of mechanization has proceeded for many decades (Giedion, 1948), its latest dramatic impact has been on clerical workers in offices with the advent of visual display units (VDUs). The introduction of VDUs increased two of the risk factors for RSI, namely rapid repetitive movements and static load (often combined with awkward postures). Further, VDUs often brought about changes in work organization, including increased workload and reductions in task diversity, autonomy, control, and peer cohesion, all of which are associated with RSI, plus a real threat of unemployment (Bammer, 1987a).

While clerical workers are nominally "white collar", their low wages and limited control over work conditions provide great similarities with manual workers. In the lower echelons of clerical work, white English-speaking women have predominated.

The health problems of women commonly receive less concern than those of men (Lewin and Olesen, 1985; Scully, 1980). In addition, the low status of clerical work makes it difficult to generate concern about any health problems associated with it. But several factors counterbalanced these weaknesses. As well as being white and English-speaking, many Australian clerical workers have moderate to high family incomes. In addition, the "second wave" of the feminist movement has given much added confidence to women, individually and collectively, to pursue their interests. A major focus for the feminist movement has been women's health issues (Boston Women's Health Book Collective, 1976). Both an organizational network, including contacts with health practitioners, and a willingness to take the concerns of women about their health seriously, helped lay the basis for an RSI movement.

These factors of intensification of work rates, introduction of VDUs, and the strength of the women's movement are hardly unique to Australia, although their exact manifestations have their own local flavor. More likely, the key difference in the Australian situation was the role of the trade union movement. Australian trade unions have long played a prominent role in the society; many have a history of activism on social issues, of which the green bans (bans on construction or other work in environmentally sensitive areas) are the most well known (Roddewig, 1978). Specific Australian institutions reflecting the strength of trade unions were crucial in the rise of an RSI movement.

One important factor here is Australian provision for compensation in the event of injury or death of workers which is widely said, in Australia, to be relatively generous [5]. Access to workers' compensation - a legacy of previous workers' struggles - laid the basis for an expansion of claims around a new pattern of injury, or increased reporting of a pre-existing pattern of injury.

Another Australian factor behind the rise of an RSI movement was workers' health centers, established by trade unions in the 1970s to deal with health problems of special concern to workers. In the late 1970s, doctors at some of these centers began to publicize pain and disability among manual workers due to repetitive work and static load. Taking this up at the beginning of the 1980s, a small number of Australian medical practitioners began to write about similar problems among white collar workers which they attributed to repetition strain and static load especially associated with VDU work (Walker, 1979; see also Browne et al., 1984; Stone, 1983; Taylor, Gow, and Gorbett, 1982; Taylor and Pitcher, 1984). By adopting the common term RSI to include a broad range of symptoms and conditions, recognition of similar problems by different practitioners and by others was facilitated.

The medical interpretation of RSI was crucial for those seeking to legitimate it as a social problem. The stamp of medical approval certifies complaints as having a real basis, backed by a highly credible profession. A number of doctors subscribing to the standard medical view, especially those who have done research and written about RSI, have been prominent in testifying in court on behalf of RSI complainants (Campbell, 1988).

The interpretation by these doctors of the potential for serious injury progressing through identifiable stages was taken up by several groups: trade unions, the media, women's health groups, and workers themselves. The link with VDUs was important because the various concerns associated with the rapid spread of this technology made the issue very topical.

In recent years, some of the white collar unions, most of whose members are government employees, have become increasingly militant. Trade unions developed increased interest in occupational health and safety in the 1980s. They publicized the problem of RSI through their internal journals, by organizing industrial action, and by backing some workers in legal test cases (Bammer, 1990b).

Reporters picked up the story, which expanded over a period of years until RSI became a household word [6]. Not only did the media pick up the story, but they maintained a sustained interest in it.

Women's health groups played a prominent role by providing support and encouragement and interacting with trade unions, media, and those with RSI. In addition, as workers saw their workmates acknowledge and report RSI, and learned first hand about the symptoms and contributing factors, they were much more likely to take note of the same problems in their own bodies.

The net result of this symbiotic process was a rapid increase in reported cases of what was dubbed RSI. The syndrome was legitimated as organic and serious by ever more doctors. Those affected were, in many cases, supported by trade unions, which demanded responses from employers such as rest breaks, "ergonomic" furniture and work reorganization. Media reports alerted many others to the problem. Support groups, often entirely women, were set up for people with RSI.

Throughout the RSI expansion, there were countervailing forces. Employers, including the government, perceived that they had much to lose from RSI. Privately, many of them considered many "victims" to be malingerers who were abusing the system to illegitimately claim workers' compensation. In this view they were supported by some doctors and co-workers. Another interpretation was that the symptoms were either exaggerated or hysterical or nothing really to worry about. It has never become entirely safe to report RSI: few people consider it to be as valid as a broken arm [7].

As the expansion in the number of RSI claims continued in the mid 1980s, a variety of responses emerged. "Ergonomic" furniture, rest breaks and exercise routines were instituted in many work places (for example, Australian Apparel Manufacturer, 1985; Dunstone, 1985; Kemp, 1984; Rowe, 1987; Tasker and Westerly, 1985). On a different front, employers (often through their insurers) attempted to minimize payments, and in several instances went to court to challenge employee claims. In this they were able to rely on a number of doctors who questioned the validity of claims of occupational injury.

The rise of RSI in Australia, rather than in some other country, arguably depended on the generation of a movement drawing on sympathetic doctors and trade unions, on the resource of the availability of workers' compensation payments, on concerted activism by committed feminists, trade unionists, health care workers, and people with RSI, and support from members of the media and the general public.

The movement included core activists (such as advocates at workers' health centers and organizers of support groups), active supporters (trade union officials, journalists), and passive supporters. It used a variety of resources to mobilize concern, including personal communication, various organizations, and media coverage. It included an interpretation of the world, in this case of occupational injury, which justified social action. The initial rapid expansion, the plateau, and the counteraction by opponents are typical of social movement dynamics.

If the recognition of RSI as a social problem depended on the existence of an RSI movement, then it could have been expected that the "problem" of RSI would seem to recede as the RSI movement fragmented and declined. There are a number of reasons which help explain why social movements decline. Internal to the movement, key activists become burnt out after years of campaigning. Unless the movement is institutionalized in the form of jobs, laws, clients, and income, many participants will drop out or move on to other issues. Failures can lead to disillusionment, whereas success can lead to a perception that the problem is being handled.

Exterior to the movement, the role of the media is important. Media interest in issues is often short-lived, and a movement must provide increasingly dramatic stories to maintain media coverage. Finally, opposition to and accommodation of the movement are crucial. Overt opposition can thwart movement initiatives, reduce morale, and block success; accommodation or cooption involves addressing the problem, often in a limited fashion, and removing its urgency or saliency.

Each of these factors played a role in the decline of the RSI movement in Australia. As the number of RSI claims stabilized and declined, the opponents of RSI mobilized. As discussed earlier they were effective in capturing and silencing the debate. Statistics on the number of cases are no longer routinely published. Although new cases continue to be reported, there is little media attention to the problem and hence a perception in many quarters that it has gone away. In a major court case, the federal government argued that RSI was not an organic injury (Campbell, 1988). On the other hand, as outlined earlier, a variety of measures, including "ergonomic" furniture and routine breaks were introduced to mitigate the problem. Thus, the movement has left some residues, including a popular awareness of RSI and likely risk factors, a network of sympathetic doctors, researchers and support groups, and some changed work practices (Bammer, 1990a, 1990b).

DE FACTO PARTISANSHIP (II): RSI AS A SOCIAL MOVEMENT

In principle, this description of RSI as a social movement does not necessarily legitimate or delegitimate the reality or importance of RSI. In practice, the social movement characteristics of RSI have been used as a basis for attacking its legitimacy. Most notably, the "pain-patient" or "social iatrogenesis" explanation of RSI focuses on many activities characteristic of social movements. Bell (1989) refers to actions by trade unions, media, sympathetic doctors, and governments, all of which he cites as a contrast with, rather than as a response to, real clinical signs of injury. Spillane and Deves (1987:48) say simply that "RSI is a social movement and not a medical epidemic." Others also cite activities characteristic of social movements to delegitimate the standard medical view. For example, Bloch (1984:685) suggests that RSI is "a figment of vested interests and politics," citing the role of trade union literature, media presentations, and traveling theater groups.

A social movement explanation tends to delegitimate RSI because it is commonly assumed - except by analysts of social problems - that a real, organic condition will be recognized as a social problem without the entrepreneurial activities of a social movement. Of course, it is precisely this assumption that is challenged by the social movements interpretation. It is the contrast between the standard assumption and the perspective of social movements as necessary to create a social problem that builds a de facto partisanship into the latter.

What is it exactly that makes the social movements perspective give de facto support to the critics of RSI? When there was no social movement, there was no debate over RSI and the issue remained invisible, both socially and sociologically. In other words, the status quo was silence about the issue. This social context of nonrecognition of RSI problems was not examined sociologically. The de facto partisanship of studying a social movement thus arises from examining the social activities of one side in the RSI struggle.

One way to change this emphasis is to focus on countermovements. In the case of RSI, a number of insurance companies and employers (including the government) have mobilized against the standard medical interpretation of RSI. Not only have they contested the awarding of workers' compensation, but they have promoted a climate of skepticism about the validity of complaints and may have helped to bring about changes in the laws in several Australian states and territories limiting opportunities to seek compensation through common law (CCH, 1987; Journal of Occupational Health and Safety - Australia and New Zealand, 1987).

The countermovement to RSI has different characteristics from the RSI movement itself, in that it is trying to deny the presence of a "real" social problem, and has strong links to powerful economic groups. Nevertheless, the countermovement is a form of social mobilization; focus on this mobilization in Australia can yield insight into activities of employers in other countries who have an interest in preventing the emergence of a movement over RSI.

The examination of countermovement activities would benefit from comparative studies. The existence of the struggle over RSI in Australia can sensitize analysts to the social conditions and arrangements of social forces in other countries which have precluded the emergence of RSI as a social problem there. Unfortunately there are relatively few comparative studies even of risks which are recognized in different countries, much less of those which are unrecognized in the dominant societies [8].

The de facto partisanship of the social movements explanation of RSI thus stems primarily from its application to Australia in the 1980s, where the issue has been most prominent, rather than other countries or other times, where a real - or potential - anti-RSI "countermovement" has been sufficiently dominant to keep the issue off most public agendas. To speak of a "potential" countermovement is to highlight the de facto partisanship associated with studying social movements.

CONCLUSION

The Australian RSI phenomenon is an excellent case study for assessing perspectives for studying social problems. The standard medical view has been central to RSI, but its deficiencies have been exposed by critics who have also presented alternative explanations. The prominence of the dispute in the Medical Journal of Australia has meant that negotiations and disagreements over medical knowledge claims, which in most cases are not easily examined, have been made public. The open involvement in the vociferous public discussion of a wide range of groups, including doctors, trade unions, employers, women's groups, and journalists, has meant that a wealth of material for examining the social dynamics of the rise and fall of RSI as a prominent social problem in Australia is openly available.

An initial motivation for this study was the question of why diverse approaches to the social examination of RSI all seem to be more useful to the critics than the proponents of RSI. The sociology of medical knowledge, with its symmetrical analysis of negotiations over knowledge claims, selectively aids the critics of RSI because, in the circumstance of the debate, a deconstruction of knowledge claims undermines to a greater extent the position of those who say RSI is a real organic condition. The approach of social problems as social movements, with its concentration on social movement activities, also selectively aids the critics of RSI because, outside sociological circles, it is commonly thought that a social problem does not need a social movement to be recognized. Therefore, pointing out movement activities, even in the context of objectivist assumptions about the existence of RSI, tends to undermine the position of RSI proponents.

Consequently, each approach is associated with a de facto partisanship supporting the critics of RSI. This is shown by the fact that partisans - only a few of whom are social scientists - who have attacked the standard medical view have undertaken a social deconstruction of medical knowledge characteristic of the sociology of medical knowledge, and have cited social-movement-like activities. This practical use of these approaches is congruent with theoretical expectations that any social explanation of a problem will undermine the viewpoint that more successfully claims to be founded on physical reality. Accordingly, proponents of the standard medical view have not referred to social explanations for RSI, but rested their case on medical evidence.

More fundamentally, the de facto partisanship in each case is linked to a contrast between sociological explanation and prevalent views about medical knowledge and reality. Hence, a plausible generalization is that any sociological explanation of RSI in Australia is likely to serve the cause of the critics. Even an objectivist analysis provides little help for the standard medical view, because the issue of the medical status of RSI has been opened up by the critics: any discussion of the debate tends to give added attention to the critics, as long as they have less credibility in the medical profession.

This assessment of de facto partisanship depends on a recognition of the artificiality of the distinction between analyst and actor. Approaches to studying RSI may appear nonpartisan in the abstract, but cannot remain so in practice because sociological concepts and studies have been taken up in the debate. Realistically, many of the social analysts have seen themselves as making a contribution to the debate rather than just commenting on it to a hypothetically separate social science community [9]. The problem from this viewpoint is that it seems so difficult to analyze the Australian RSI issue without providing de facto support for the critics of RSI.

The identification of de facto partisanship in the 1980s Australian RSI debate is aided by the existence of a well-established orthodoxy and by the case study approach used in sociological studies of the issue. Both these conditions apply to many social problems, but not all. In some cases the debate is sufficiently balanced or fluid that there is no clear orthodoxy. In such situations, de facto partisanship becomes more complex. The case study approach, which is standard in the study of social problems and which we have used in this paper, provides a ready resource to actors who want to use sociological findings for partisan purposes. De facto partisanship may be harder to identify and assess in studies that do not focus on particular issues but deal instead, for example, with the general dynamics of certain types of social problems.

The case of RSI highlights a common feature in standard approaches to the study of social problems: their lack of any way to select for analysis either social conditions which are not social problems or scientific evaluations which are not subject to challenge. Social scientists began studying RSI only after the issue became prominent, and they have only looked at the problem in Australia where there has been a vehement debate. In many other countries, where pain and disability associated with repetitive work have been widely documented but no suitable scientific definitional activities or social movements have arisen, the issue has been seemingly invisible to social scientists [10]. Thus, a key factor in creating de facto partisanship on the RSI issue is the choice to examine the Australian debate in the 1980s, when the standard medical view that RSI is an organic, work-related injury came under attack.

This sort of de facto partisanship could be reversed by dealing with issues when there has been little or no public debate. This would mean switching from studying what has come to be defined as a social problem (in a particular time and place) to studying what the analyst believes should be a social problem (perhaps in another time and place). Such an approach would not eliminate partisanship, and indeed would make it much more obvious. Considering that such partisanship is inescapable, this is a preferable approach.

It is certainly possible for social scientists to analyze the social structures which inhibit the emergence of a social problem as well as the activities which signal the recognition of one. One way to do this is to use the concept of "nondecisionmaking" (Bachrach and Baratz, 1962; Crenson, 1971; Lukes, 1974) to examine the nonexistence (or "unpolitics") of particular social problems.

If a problem exists in one country but not in another, that can be a reason to study either its dynamics in the first country or its failure to emerge in the second. Similarly, a debate restricted to one medium (scientific journals, mass media, "fringe" publications, particular organisations, personal accounts) can provide a trigger to study either the debate that exists or the lack of discussion elsewhere. It is obvious here that the analyst must make a choice of what to study; this choice entails partisanship, whether open or de facto.

This conclusion has implications for the sociological justification, or warrant, for studying a social issue. For objectivists, the warrant is real social conditions. For constructionists, denied the methodological resource of objective reality, the warrant is social debate and struggle. The inevitability of de facto partisanship suggests that the range of acceptable warrants should be expanded. In order to encourage study of topics which are not (yet) defined as social problems, it should be sufficient either that there is scientific evidence (not necessarily widely accepted), that there is debate, or that there are critics (even if no public debate). Since a sociologist can be a critic, in principle no further warrant is needed than the analyst's own judgment.

Of course, this prescription no more than describes the actual practices of some sociologists, who on occasion have been willing to analyze what they alone consider to be issues of significance. The concept of de facto partisanship provides a theoretical context for such initiatives. Given that partisanship is built into the choice of theoretical framework and the choice of where and when to apply it, it is futile to try to eliminate partisanship, de facto or otherwise. Instead, a plurality of partisanships should be encouraged, in the spirit of the maxim that "there is no single road to truth".

In summary, we propose the following generalizations on the basis of our examination of accounts of RSI in Australia. To begin, there is a de facto partisanship associated with the choice of a framework for analysis. For example, symmetrical constructionist analyses tend to more severely undermine those views with greater cognitive authority. But de facto partisanship at this level is only a tendency, which can be either accentuated or counteracted by the choice of what issue to study. Sociological accounts undertaken where and when a social problem is receiving greatest attention sociologize and hence tend to delegitimate the dominant view of what is a social problem; similarly, sociological accounts of why some issue is not considered a social problem tend to legitimate that issue as a social problem. But this second facet of de facto partisanship is also only a tendency. In order to assess de facto partisanship in practice, a sociological examination of the uses of sociological accounts and sociological perspectives in ongoing social problems activities is essential. Finally, these generalizations require sociological study, namely more examination of de facto partisanship and, especially, the uses of sociological accounts in the activities being studied.

ACKNOWLEDGMENTS

Useful comments on earlier drafts of this paper were received from Ilse Blignault, Dorothy Broom, Phyll Dance, Merrelyn Emery, Andrew Hopkins, David Legge, John McCallum, Armand Mauss, Jan Reid, Pam Scott, Janis Shaw, Terry Stokes, Roslyn Woodward and several anonymous reviwers. Stefanie Pearce provided valuable assistance with production of the paper.

FOOTNOTES

1. There have been detailed examinations of the epidemic among government employees (Task Force Report, 1985), employees of the national telephone company (Hocking, 1987) and employees at one university (Bammer, 1987b).

2. The account is not a single unified view - which would be the usual way - but is composed from the two separate perspectives of the sociology of medical knowledge and of social problems as social movements which are partly competing and partly complementary.

3. These views are considered in more depth in Bammer and Martin (1988). Somewhat different categorizations have been produced by Meekosha and Jakubowicz (1986), Spillane and Deves (1987), and others.

4. Quite a number of factors and groups have been nominated as important or possibly important in the rise of RSI in Australia. These include changing work organization (especially increased productivity), inability to change jobs, migrant workers, white middle-class female workers, Australian trade unions, laws on workers' compensation, the medical profession, and the media. Here we draw on these factors and a wide range of sources, including Bartlett (1984), Davis and Lansbury (1986), Meekosha and Jakubowicz (1986), Reid and Reynolds (1990), Stone (1984), and Willis (1986).

5. Hopkins (1990) documents some evidence for this in a comparison of federal government employees in the United States and Australia.

6. In our literature search, we encountered a wide range of business and occupational group newsletters and journals containing articles about RSI. Usually there was only a single article in the mid 1980s, subscribing to the standard medical view. See also Bammer (1990b).

7. Ironically, Reid and Reynolds (1990) suggest that this denial that RSI is a real problem has been a major factor in making symptoms chronic, because doctors who do not believe that the disorders are real fail to provide appropriate care and constantly put their patients on the defensive by needing to prove they are in pain.

8. Among the few comparative studies of risk are Gillespie, Eva, and Johnston (1979), Irwin (1985), and Jasanoff (1986). Bammer (1988) has carried out a pilot comparative study across seven countries showing that although the diagnoses given by doctors varied greatly from country to country, RSI is common among office workers. There is some evidence that it is more common in Australia.

9. This comment is based on our interaction with most of the analysts cited.

10. Keisler and Finholt (1988) is a US article which looks at the Australian problem with little comment on the United States except as a baseline comparison of "facts".

REFERENCES

Australian Apparel Manufacturer
1985 "Workers' compensation - who hurts the most?" June/July:12,15,18.

Awerbuch, Mark
1985 "RSI, or 'kangaroo paw'." Medical Journal of Australia 142:237-38.

Bachrach, Peter, and Morton S. Baratz
1963 "Two faces of power." Americal Political Science Review 56:947-52.

Bammer, Gabriele
1987a "How technologic change can increase the risk of repetitive motions injuries." Seminars in Occupational Medicine 2:25-30.

Bammer, Gabriele
1987b "VDUs and musculo-skeletal problems at the Australian National University - a case study." In Work with Display Units 86 (Selected Papers from the International Scientific Conference on Work with Display Units, Stockholm, Sweden, 12-15 May 1986), eds. Bengt Knave and Per-Gunnar Widebäck, 279-287. Amsterdam: North-Holland.

Bammer, Gabriele
1988 "The prevalence of work-related neck and upper limb disorders among office workers in 7 countries - a pilot study." Preprints of the International Conference on Ergonomics, Occupation Safety and Health and the Enviroment, Volume 1, Beijing, October, 297-305.

Bammer, Gabriele
1990a "The epidemic is over ... or is it?" Australian Society 9 (April):22-24.

Bammer, Gabriele
1990b "Occupational disease and social struggle: the case of work-related neck and upper limb disorders." National Centre for Epidemiology and Population Health Working Paper Number 20.

Bammer, Gabriele
1990c "Review of current knowledge - musculoskeletal problems." In Work with Display Units 89 (Selected Papers from the Second International Scientific Conference on Work with Display Units, Montreal, 11-14 September 1989), eds. Louis Berlinguet and Diane Berthelette, 113-120. Amsterdam: North-Holland.

Bammer, Gabriele, and Brian Martin
1988 "The arguments about RSI: an examination." Community Health Studies 12:348-58.

Barnes, Barry
1974 Scientific Knowledge and Sociological Theory. London: Routledge and Kegan Paul.

Barnes, Barry
1981 "On the 'hows' and 'whys' of cultural change (response to Woolgar)." Social Studies of Science 11:481-98.

Bartlett, Ben
1984 "History of the Sydney workers health movement." Australian Left Review 88:40-45.

Bell, David S.
1986 "RSI." Medical Journal of Australia 145:117-18.

Bell, David S.
1989 "'Repetition strain injury': an iatrogenic epidemic of simulated injury." Medical Journal of Australia 151:280-84.

Berger, Peter L., and Thomas Luckmann
1966 The Social Construction of Reality: a Treatise in the Sociology of Knowledge. Garden City, New York: Doubleday.

Best, Joel, ed.
1989a Images of Issues: Typifying Contemporary Social Problems. New York: Aldine de Gruyter.

Best, Joel
1989b "Afterword". In Images of Issues: Typifying Contemporary Social Problems, ed. Joel Best, 243-53. New York: Aldine de Gruyter.

Black, Peter
1987 "Psychiatric aspects of regional pain syndrome." Medical Journal of Australia 147:257.

Bloch, Bernard
1984 "Repetition strain injuries." Medical Journal of Australia 140:684-85.

Bloor, David
1976 Knowledge and Social Imagery. London: Routledge and Kegan Paul.

Boston Women's Health Book Collective
1976 Our Bodies, Ourselves. New York: Simon and Schuster.

Brooks, Peter M.
1986 "Regional pain syndrome - the disease of the '80s'." Bulletin of the Postgraduate Committee in Medicine, University of Sydney 42:55-59.

Browne, Christopher D., Bernard M. Nolan, and Donald K. Faithfull
1984 "Occupational repetition strain injuries: guidelines for diagnosis and management." Medical Journal of Australia 140:329-32.

Bury, M. R.
1986 "Social constructionism and the development of medical sociology." Sociology of Health and Illness 8:137-69.

Campbell, Stephen A.
1988 "Taxing work: Susan Cooper's RSI case." Paper presented at the National Conference of Social Science Research on Occupational Health and Safety: Past, Present and Future, Adelaide, August.

CCH Occupational Health and Safety Editors
1987 "Workers compensation - end and means." Journal of Occupational Health and Safety - Australia and New Zealand 3:159-61.

Champion, G. David, Judith Cornell, Christopher D. Browne, Raymond Garrick, and Timothy J. Herbert
1986 "Clinical observations in patients with the syndrome 'repetition strain injury.'" Journal of Occupational Health and Safety - Australia and New Zealand 2:107-13.

Cleland, Leslie G.
1987 "'RSI': a model of social iatrogenesis." Medical Journal of Australia 147:237-9.

Collins, H. M.
1981 "Stages in the empirical programme of relativism." Social Studies of Science 11:3-10.

Collins, H. M., and T. J. Pinch
1979 "The construction of the paranormal: nothing unscientific is happening." In On the Margins of Science: The Social Construction of Rejected Knowledge, ed. Roy Wallis, 237-69. Keele: University of Keele.

Crenson, Matthew A.
1971 The Un-politics of Air Pollution: A Study of Non-decisionmaking in the Cities. Baltimore: Johns Hopkins University Press.

Davis, Ed, and Russell Lansbury
1986 "Democracy and control in the workplace: an introduction." In Democracy and Control in the Workplace, ed. Ed Davis and Russell Lansbury, 1-29. Melbourne: Longman Cheshire.

Dunstone, William G.
1985 "How one company tackles problem." Safety in Australia 8 (August):11-13.

Emery, Merrelyn
1988 "Learning about the unpredicted: the case of repetition strain injury (RSI)." Studies in Continuing Education 10:30-45.

Ferguson, David
1984 "The "new" industrial epidemic." Medical Journal of Australia 140:318-19.

Figlio, Karl
1978 "Chlorosis and chronic disease in 19th-century Britain: the social constitution of somatic illness in a capitalist society." International Journal of Health Services 8:589-617.

Figlio, Karl
1982 "How does illness mediate social relations? Workmen's compensation and medico-legal practices, 1890-1940." In The Problem of Medical Knowledge: Examining the Social Construction of Medicine, eds. Peter Wright and Andrew Treacher, 174-224. Edinburgh: Edinburgh University Press.

Foster, L., and H. Fry
1988 "Conflicts in medical knowledge - the case of 'RSI'." Search 19:176-81.

Fry, Hunter J. H.
1986 "Overuse syndrome of the upper limb in musicians." Medical Journal of Australia 144:182-85.

Giedion, Sigfried
1948 Mechanization Takes Command: A Contribution to Anonymous History. New York: Oxford University Press.

Gilbert, G. Nigel, and Michael Mulkay
1982 "Warranting scientific belief." Social Studies of Science 12:383-408.

Gillespie, Brendan, Dave Eva, and Ron Johnston
1979 "Carcinogenic risk assesment in the United States and Great Britain: the case of aldrin/dieldrin." Social Studies of Science 9:265-301.

Gubrium, Jaber F.
1987 "Structuring and destructuring the course of illness: the Alzheimer's disease experience." Sociology of Health and Illness 9:1-24.

Gusfield, Joseph R.
1981 The Culture of Public Problems: Drinking-driving and the Symbolic Order. Chicago: University of Chicago Press.

Hadler, Nortin M.
1986 "Industrial rheumatology: the Australian and New Zealand experiences with arm pain and backache in the workplace." Medical Journal of Australia 144:191-95.

Hall, Wayne, and Louise Morrow
1988 "'Repetition strain injury': an Australian epidemic of upper limb pain." Social Science and Medicine 27:645-49.

Hazelrigg, Lawrence E.
1986 "Is there a choice between 'constructionism' and 'objectivism'?" Social Problems 33:S1-S13.

Hess, David
Forthcoming Science in the New Age: A Cultural Critique of Skepticism and the Paranormal in the United States. Madison: University of Wisconsin Press.

Hilgartner, Stephen, and Charles L. Bosk
1988 "The rise and fall of social problems: a public arenas model." American Journal of Sociology 94:53-78.

Hocking, Bruce
1987 "Epidemiological aspects of 'repetition strain injury' in Telecom Australia." Medical Journal of Australia 147:218-222.

Hopkins, Andrew
1989 "The social construction of repetition strain injury." Australian and New Zealand Journal of Sociology 25:239-59.

Hopkins, Andrew
1990 "The social recognition of repetition strain injuries: an Australian/American camparison." Social Science and Medicine 30:365-72.

Ibarra, Peter R., and John I. Kitsuse
Forthcoming "Vernacular constituents of moral discourse: an interactionist proposal for the study of social problems." In Perspectives on Social Problems, Volume 4B, eds. James A. Holstein and Gale Miller. Greenwich, Conn.: JAI Press.

Ireland, Damian C. R.
1986 "Repetitive strain injury." Australian Family Physician 15:415-18.

Irwin, Alan
1985 Risk and the Control of Technology: Public Policies for Road Traffic Safety in Britain and the United States. Manchester: Manchester University Press.

Jasanoff, Sheila
1986 Risk Management and Political Culture: A Comparative Study of Science in the Policy Context. New York: Russell Sage Foundation.

Journal of Occupational Health and Safety - Australia and New Zealand
1987 "Workers compensation - common law reforms." 3:524.

Keisler, Sara, and Tom Finholt
1988 "The mystery of RSI." American Psychologist 43:1004-15.

Kemp, Patricia
1984 "R.S.I. control - a team effort." Proceedings of the Australian Physiotherapy Association ACT Branch Symposium: Repetition strain injury - A pain in the "?" Canberra: Australian Physiotherapy Association.

Latour, Bruno
1987 Science in Action: How to Follow Scientists and Engineers through Society. Milton Keynes: Open University Press.

Lewin, Ellen, and Virginia Olesen, eds.
1985 Women, Health and Healing: Toward a New Perspective. New York: Tavistock.

Lucire, Yolande
1986a "Neurosis in the workplace." Medical Journal of Australia 145:323-27.

Lucire, Yolande
1986b "RSI: when emotions are converted." Safety Australia 9 (February):8-12.

Lukes, Steven
1974 Power: A Radical View. London: Macmillan.

MacKenzie, Donald
1981 "Interests, positivism and history." Social Studies of Science 11:498-504.

MacKenzie, Donald
1984 "Reply to Yearley." Studies in History and Philosophy of Science 15:251-59.

Mauss, Armand L.
1975 Social Problems as Social Movements. Philadelphia: J. B. Lippincott.

Mauss, Armand L.
1989 "Beyond the illusion of social problems theory." In Perspectives on Social Problems, Volume 1, eds. James A. Holstein and Gale Miller, 19-39. Greenwich, Conn.: JAI Press.

McDermott, Francis T.
1986 "Repetition strain injury: a review of current understanding." Medical Journal of Australia 144:196-200.

McPhee, Barbara
1982 "Deficiencies in the ergonomic design of keyboard work and upper limb and neck disorders in operators." Journal of Human Ergology 11:31-6.

Meekosha, Helen
1986 "Eggshell personalities strike back - a response to the bosses' doctors on RSI." Refractory Girl 29 (May):2-6.

Meekosha, Helen, and Andrew Jakubowicz
1986 "Women suffering RSI: the hidden relations of gender, the labour process and medicine." Journal of Occupational Health and Safety - Australia and New Zealand 2:390-401.

Morgan, R. G.
1986 "Reply to Farrelly." Medical Journal of Australia 144:501-2.

Mulkay, Michael
1979 Science and the Sociology of Knowledge. London: Allen and Unwin.

Mulkay, Michael, and G. Nigel Gilbert
1982 "Accounting for error: how scientists construct their social world when they account for correct and incorrect belief." Sociology 16:165-83.

Mullaly, Janelle, and Lyn Grigg
1988 "RSI: integrating the major theories." Australian Journal of Psychology 40:19-33.

Pfohl, Stephen
1985 "Toward a sociological deconstruction of social problems." Social Problems 32:228-32.

Quintner, John L.
1989 "The pain of RSI: the central issue." Australian Family Physician 18 (December):1542-47.

Reid, Janice, and Lyn Reynolds
1990 "Requiem for RSI: the explanation and control of an occupational epidemic." Medical Anthropology Quarterly 4:162-90.

Richards, Evelleen
1988 "The politics of therapeutic evaluation: the vitamin C and cancer controversy." Social Studies of Science 18:653-701.

Richards, Evelleen
1991 Vitamin C and Cancer: Medicine or Politics? London: Macmillan.

Roddewig, Richard J.
1978 Green Bans: The Birth of Australian Environmental Politics. Montclair, New Jersey: Allanheld, Osmun.

Rowe, Susan A.
1987 "Management involvement - a key element in preventing musculoskeletal problems in visual display unit users in Australia." Ergonomics 30:367-72.

Rush, Jonathan
1984 "The overuse of tenosynovitis." Medical Journal of Australia 145:614-15.

Russell, Stewart
1986 "The social construction of artefacts: a response to Pinch and Bijker." Social Studies of Science 16:331-46.

Scarf, Garry E., and David Wilcox
1984 "Alleged work-related injuries." Medical Journal of Australia 141:765.

Schneider, Joseph W.
1985 "Social problems theory: the constructionist view." Annual Review of Sociology 11:209-29.

Schneider, Joseph W., and John I. Kitsuse, eds.
1984 Studies in the Sociology of Social Problems. Norwood, New Jersey: Ablex.

Scott, Pam, Evelleen Richards, and Brian Martin
1990 "Captives of controversy: the myth of the neutral social researcher in contemporary scientific controversies." Science, Technology, & Human Values 15:474-94.

Scully, Diana
1980 Men Who Control Women's Health: The Miseducation of Obstetrician-Gynecologists. Boston: Houghton Mifflin.

Sharrod, Fred
1985 "RSI, 'kangaroo paw', or what?" Medical Journal of Australia 142:376.

Spector, Malcolm, and John I. Kitsuse
1977 Constructing Social Problems. Menlo Park, Calif.: Cummings.

Spillane, Robert, and Leigh Deves
1987 "RSI: pain, pretence or patienthood?" Journal of Industrial Relations 29 (March):41-48.

Stone, William E.
1983 "Repetitive strain injuries." Medical Journal of Australia 2:616-18.

Stone, William E.
1984 "Occupational repetitive strain injuries." Australian Family Physician 13:681-83.

Task Force Report
1985 Repetition Strain Injury in the Australian Public Service. Canberra: Australian Government Publishing Service.

Tasker, David, and Margaret Westerly
1985 "The introduction and practice of ergonomics at Avon." Proceedings of Pain at Work Symposium, Sydney: Ergonomics Society of Australia and New Zealand (NSW Branch) and Safety Institute of Australia (NSW Division), November, 67-72.

Taylor, Richard, Chris Gow, and Steven Gorbett
1982 "Repetition injury in process workers." Community Health Studies 6:7-13.

Taylor, Richard, and Michael Pitcher
1984 "Medical and ergonomic aspects of an industrial dispute concerning occupation-related conditions in data process operators." Community Health Studies 8:172-80.

Troyer, Ronald J.
1989 "Are social problems and social movements the same thing?" In Perspectives on Social Problems, Volume 1, eds. James A. Holstein and Gale Miller, 41-58. Greenwich, Conn.: JAI Press.

Walker, Jim
1979 "Tenosynovitis: a crippling epidemic in industry." New Doctor 13:19-21.

Wallace, Meredith, and Peter Buckle
1987 "Ergonomic aspects of neck and upper limb disorders." International Review of Ergonomics 1:173-200.

Williams, Trevor A.
1985 "Visual display technology, worker disablement, and work organization." Human Relations 38:1065-84.

Willis, Evan
1983 Medical Dominance. Sydney: Allen and Unwin.

Willis, Evan
1986 "Commentary: RSI as a social process." Community Health Studies 10:210-19.

Willis, Evan
1989 "The industrial relations of occupational health and safety: a labour process approach." Labour and Industry 2:317-33.

Woolgar, Steve
1981 "Interests and explanation in the social study of science." Social Studies of Science 11:365-94.

Woolgar, Steve, and Dorothy Pawluch
1985a "Ontological gerrymandering: the anatomy of social problems explanations." Social Problems 32:214-27.

Woolgar, Steve, and Dorothy Pawluch
1985b "How shall we move beyond constructionism?" Social Problems 33:159-62.

Wright, Peter W. G.
1980 "The radical sociology of medicine." Social Studies of Science 10:103-20.

Yearley, Steven
1982 "The relationship between epistemological and sociological cognitive interests: some ambiguities underlying the use of interest theory in the study of scientific knowledge." Studies in the History and Philosophy of Science 13:353-88.