Losing the "eyes in the back of our heads": social service skills, lean caring, and violence.

Violence in the social services work place in general, and the developmental services in particular, has increased in the last several years. Findings from an ethnographic study suggests that new, lean forms of work organization remove opportunities to use or learn many of the tacit or practice skills workers previously used to keep themselves and their clients safer in the work place. This article describes many of these skills and the new management schemes that remove the possibility to develop or transmit these praxis skills. The article concludes by analyzing the convergence between the new labour processes and the competency approach to work place skills. Noting the loss of praxis skills that kept workers and clients safer, the conclusion highlights the hidden costs to developmental sector clients and workers.

Key words: violence, social services, skills, workplace, labour, safety, workers

**********

While non-fatal injury and occupational illness rates have been falling in workplaces across Canada and the United States, rates in health care and the social services have been soaring (Boyd, 1995; National Institute for Occupational Safety and Health, 2000). Evidence from one Canadian province indicates a tenfold increase in claims from social service workers (Boyd, 1995). Many of these claims seem to be related to the high level of client perpetrated violence against staff. In Canada, the rate of injury in the developmental services sector is higher than that in other parts of the social services. The developmental services sector is a subsector of the social services which provides community services to people with intellectual disabilities. Until de-institutionalization in the 1970s, most people with intellectual disabilities spent their lives in long-term care facilities or asylums (Charlton, 1996; Michalko, 2002; Oliver, 1996). While some argue that the impetus behind de-institutionalization was a recognition of the rights of people with disabilities to reach their full potential, others assert that inadequate funding has resulted in the continued warehousing and stigmatization of people with intellectual disabilities (albeit in smaller settings), rather than their integration into a caring community (Braddock and Hemp, 1997; Taylor and Bogdan, 1989; Traustadottir, 2000).

Like other parts of the social services sector, the developmental services sector has, in recent years, been hit with funding cuts, and has integrated new mandates and models of management that lean out service delivery, and reflect a move away from social obligation and mutual responsibility (Baines, 2004a, 2004b; Abramovitz and Withorn, 1999; Clarke and Newman, 1992; Fabricant and Burghardt, 1997). Part of this new lean model and logic of service delivery is a reformulation of workers' skills. Under restructuring, skills are no longer recognized as a complex synergy of individual and collective knowledge that develop incrementally and over time. Rather, skills are recast as "competencies" or "human capital" that workers should own or obtain (Jackson, 1997). Competencies or skills are measured against "value-added, which means basically contributions to profits." (124) Dominelli and Hoovgelt (1996) note that the notion of "competencies" was introduced as "a way of defining the core skills professionals needed to do the jobs which employers wanted them to do" (55), rather than defining core skills within the broader context of what social service workers should be doing. Within the competency framing of skills, if activities and interactions in the social services do not contribute to cost-saving, they are likely to be eliminated from the prescribed tasks and routines of workers.

This restructured managerial and skill regime appears to be one of the major shifts that has increased the risk and likelihood of violence in the developmental sector. A recent ethnographic study of three developmental service agencies in Ontario, Canada showed that, while violence is endemic in the developmental services workplace, many workers, particularly those hired prior to social service restructuring, displayed skills that appear to significantly reduce client-perpetrated violence (Baines, Hadley, Pollack, Slade, Brooker, Fay, Lewchuk, Preston, and Dimatrova, 2002). These skills were not generally found in the academic text books, compentencies discourse, or performance management literature that guides most contemporary approaches to human service management. Known in the skills literature as tacit-knowledge (Darrah,1997), these skills were learned on the job by workers who had sufficient time and support to develop a sense of working as an informal team, and to learn from each other, their clients, and their own lived-in experience. While managerial approaches such as Total Quality Management (TQM) purport to capture and make use of tacit-skills, in most work places, the empowerment aspects of TQM are never realized and the approach to skills continues to operate under a top-down, human resources framing which denies the existence of tacit-skills, the social nature of work and learning, and the importance of interpersonal networks in the workplace (Darrah, 1997; Manwearing and Wood, 1984). Indeed, various aspects of the contemporary organization of developmental services work means that opportunities to use, learn and transmit these tacit-or praxis-level skills are being eliminated, along with the skills themselves, and the greater safety they provided to workers and clients.

The Research Project

This article analyses a subset of data from a larger qualitative study aimed at developing multi-level portraits of three developmental service agencies in order to identify and theorize about factors that contribute to stress, violence and workload issues in the restructured social services sector. A multi-method ethnographic approach was used to gather interview, participant observation and document data in three unionized, developmental services agencies in Ontario, Canada's most populous and wealthy province. The study's eleven participant observation sites were selected based on recommendations by key informants (executive director, union president, and key workers) which were cross-referenced. Field notes taken during the participant observations were subsequently transcribed and analyzed in concert with interview data.

Forty-one in-depth, semi-structured, audio tape-recorded interviews took place in which research participants were posed broad questions relating to changes they had seen or experienced in the workplace in the last five years in relation to work organization, pace, workload, stress, health concerns, violence, and other workplace issues that might concern them. A purposive (Lincoln and Guba, 1985) sample was built for key informants (executive director, union president, key managers and workers) while a snowball sample of front-line staff was built in two of the three sites. At the request of the executive director (E.D.) in one site, starting at a random point in a staff list supplied by the E.D., every ninth worker was contacted and asked to participate in the study until a full sample was achieved.

The average age of those interviewed across the three sites was 37.1 years, with an average of 9.9 years employment in the agency. The sample was approximately 35 per cent male. As workers of colour and youth have been identified in the literature as especially vulnerable to stress and injury (National Occupational Research Agenda, 1999; National Institute for Occupational Safety and Health/National Occupational Research Agenda, 1995), a small purposive sample of each of these groups was also interviewed and will be written about separately.

The study also included a review of agency documents related to health and safety such as incident reports, first aid report, liability claim forms, and Health and Safety Committee minutes. Document data were compiled and compared across reporting forms, and statistical profiles were developed where possible. Data were analyzed using a quasi-grounded theory (Strauss and Corbin, 1990) method involving constant comparison of the data until themes emerged and connections could be made across and among data. Data analysis diverged from grounded theory in that emerging themes were linked to larger economic, political and managerial trends in the sector in order to develop themes beyond those found in individual narratives and observations.

Defining the Problem: Worker Behavior or Work Organization

Studies in Canada place rates of violence among social service Workers at 61 per cent of respondent (Canadian Union of Public Employees, 1994). These findings are similar to those conducted in the U.S. where the rate of assault ranged from 57 per cent (Newhill, 1997) to 26 per cent (Tully, Kropf and Price, 1993). Shultz (1987) and Brockman and MacLean (2000) found that the type and severity of violence varied according to the type of social service agency, with certain kinds of services experiencing significantly higher levels of violence. For example, Canadian studies indicate that child welfare and developmental service workers experience higher levels of violence than do workers in welfare or community services (Canadian Union of Public Employees, 1999; Ontario Public Service Employees Union, 1999; Pizzino, 1994).

While the research described above substantiates high levels of violence in the social services, little is known about the negative impacts new forms of work organization have had on stress and workplace abuse or how older forms of work organization may have kept developmental sector workers and clients safer. It will be argued that it is not just that older forms of work organization permitted workers to interact more positively and in a less stressed way with clients, thus reducing frictions that may lead to violent outbursts. It is also that within the older work organization environments, workers could and did develop tacit or practice skills that kept them and their clients safer. The new forms of work organization fail to recognize the existence of these skills, and create conditions under which it is very difficult to develop or transmit them. Eventually, the widespread reliance of new forms of work organization could result in the elimination of tacit skills from the everyday praxis-level knowledge of developmental workers and an exacerbation of workplace violence.

The current increase in workplace violence (Pizzino, 2002, 1994; Rey, 1996; Tully, Kropf and Price, 1993), has been studied from three major positions--as an outcome of lean work organization (Boyd, 2002; Taylor and Tyler, 2000), as a lack of skills and behavior strategies on the part of the workers (Donovan, 1987; Rey, 1996), or as both (Chapell and Di Martino, 1998; Pizzino, 2002; Stanley and Goddard, 2002). Behavior or human resource strategies focus on initiating changes in the workers' skills, activities and conduct with the aim of minimizing or diffusing workplace violence. Combining human capital (competencies) and human resource development, behavior strategies usually focus on worker training, violence policy development, and supervisory reinforcement of preferred worker behaviors. (Arway, 2002; Kinney, 1995; Rey, 1996; Stanley and Goddard, 2002). However, Taylor and Tyler (2002) assert that these strategies accept that violence is an occupational hazard to which workers must adjust and place the onus on the workers to reduce or accommodate violence, rather than on management to remove the root causes of workplace assault. In contrast, the work organization literature shifts the exclusive focus on employee responsibility to the following factors: reductions in staff levels; decreased time to perform quality work; unmet client expectations; and client frustration with inferior services (Boyd, 2002; Pizzino, 2002; Taylor and Tyler, 2002).

It is well established that workplace violence is linked to the way work is organized, suggesting that certain kinds of conditions, such as handling cash, or working alone, generate a higher risk of violence (Boyd, 2002; Chapell and Di Martino, 1998; Pizzino, 2002; National Institute for Occupational Safety and Health, 1996). Even minor changes in work organization can lessen or increase the potential for violence in the workplace. Moreover, other kinds of workplace problems, such as stress and workload issues, are also linked to work organization (Baines et al, 2002; CUPE, 1999; Gardell, 1982; Karasek and Thorell, 1990; Karasek, 1979; Kohn, 1976; Messing and Reveret, 1983). Scholars from occupational health, labour studies, sociology, and social work have pointed to the way that new lean and flexible forms of work organization increase workplace stress and negative health impacts. For example, Landsbergis, Cahill and Schnall (1999) argue the introduction of lean production management systems in health care, and in industry increase stress and associated health problems. Research further substantiates that violence occurs more frequently among highly stressed workers than those experiencing lower stress (Northwestern National Life, 1993). The National Institute for Occupational Safety and Health (NIOSH) (1996) notes that while stress can be a cause and an effect of workplace violence, occurrences of workplace violence inevitably lead to increased stress and even to long-term health disorders.

Praxis Skills That Keep Workers and Clients Safer

Workers involved in this study reported that they relied on close relationships with their clients and their co-workers to keep the environment safe and on a positive footing. Long-term workers talked about being able to discern how clients are likely to behave by how they walk through the door. A capacity to instantly assess clients, inter-client dynamics, and likely trigger points is required throughout a shift. Multi-level observation skills were coupled with a detailed and nuanced knowledge of each client and their likely triggers. Daily activities were planned and organized in such a way as to minimize factors that could escalate stressed clients or create an environment in which tempers flare. Workers reported detailed knowledge of how each client got along with other clients and each staff member, which types of activities produced anxiety for each client, as well as which activities were likely to de-escalate hurt or tense feelings for each individual. Seating plans and staff assignments have special salience in violence prevention strategies, as do calming activities and interventions at tricky transition points such as activity changes, shift changes, the evening meal and getting on and off transportation. Workers called this "up-front work" in which they learn to "read" clients' emotional states, and remain tightly attuned to the smallest change in moods, activity levels, or new stressors in the environment. Up-front work also involved the development of individualized prevention strategies for each client aimed at the avoidance of triggers that may threaten the safety of the individual and group. Both verbal and non-verbal communication skills were used in up-front work as many clients had communication disorders or are completely or largely nonverbal.

Diffusion of heightened tensions is another important in situ set of skills as the introduction of new staff or clients, new furniture or equipment, and changes in daily programming or meal times could disrupt otherwise unflurried routines and result in stressed shifts and problematic inter-personal dynamics. Observation and timely, client-specific interventions were key to preventing violence. As one worker put it, a detailed knowledge of "what sets the guys (clients) off", and what works best to calm them down, means that, "we can nip things in the bud before they actually get out of hand."

Proud of the capacity for professional, ongoing, multi-factor monitoring, workers likened it to a mother's capacity to supervise children while completing numerous other tasks within the home. As one day program worker observed of her team's ability to assess potential trouble spots and intervene quickly, "we all have eyes in the back of our heads and we know what is going on all around us all the time." One of the remaining intact team of mostly full-time, permanent workers laughed about their senior worker's capacity to call out admonitions to fractious clients during meal times without even turning her head to see who was causing an uproar. This capacity gave her an almost mythical standing among clients and staff, although it is not unlike practices of mothers and others who provide care to the young.

Despite the measures described above, violence sometimes occurred. Pre-arranged intervention strategies and team work were seen as key to quickly quelling violent situations and reducing disruption of clients and workers. One worker reported that "you need to know that the other staff on the shift will back you and that they know what they are doing." Team work requires time to plan, a sense of common goals, and as one worker put it, "a kind of shorthand form of communication" in which single words or short phrases elicit the back-up strategies needed to isolate or restrain assaultive individuals, assist the victim, remove other clients and staff from the vicinity, de-escalate the crisis, and call for medical or supervisory assistance. Debriefing with clients and co-workers is the final step needed in a good violence intervention plan. This step re-establishes that everyone is safe, and assesses how well strategies worked as well as what changes may be needed in the future.

The skills described in this section include instant assessment of individual clients, ongoing monitoring and assessment of individual clients, inter-client dynamics, and group dynamics, planning, coordinating, strategy and intervention development, team communication, violence diffusion, violence intervention, client and co-worker debriefing and program re-evaluation and fine-tuning. As a number of workers observed, and the data produced by the study substantiates, the most important skills on the job are "communication" and "interpersonal skills". Communication and interpersonal relating between staff and clients and among staff, never stopped and when impeded in some way, difficulties almost always emerged. While some of these skills are recognized by management and parts of many are taught in formal education settings, their operationalization within a developmental services setting requires the kind of knowledge and skills that come from actual practice with real people in their lived-in actualities.

The work force in the developmental services is predominantly female and, as noted earlier, some of the skills described above resemble those of mothers. Such skills are often thought to be natural for women, rather than learned, and generally not understood to involve high level skills, knowledge and creativity (Baines, Evans and Neysmith, 1998; Glazer, 1993; Herd and Meyer, 2002). Managerial, governmental and academic evaluations of skills tend to devalue the kinds relational skills many women use in their work, and elevate the technical skills required in men's work (Duane-Richard, 2000; Jenson, 1989). Jane Gaskell (1990) argues that "skills designations are nothing more than power relations expressed in the language of skill.'(143) These same power relations are seen in evaluations of competencies which tend to emphasize formal teaching from managers or institutions of higher learning, and downplay or ignore informal skill sharing among workers, the ways that groups of workers accomplish tasks using the diversity of skills and strengths present in a group and the synergy of strengths that develop within a team (Darrah, 1997). As Duane Richard (2000) notes, skill evaluations are tools by which wage inequality is perpetuated and occupation segregation is justified.

Strategically using the performance management notion of "best practices", in a technical report to managers our research team noted that "best practices" were found in environments in which staff had sufficient time and resources to build supportive relationships with clients, and positive relationships with each other (Baines et al, 2002). Best practices also included the work of a supervisor who demonstrated a supportive and active interest in clients and workers, acted as a problem solver and hands-on trouble-shooter, while providing latitude for staff decision making and support for their initiatives (Baines et al, 2002). Unfortunately, as the next section will show, situations in which best practices were found were regrettably rare and are likely to become increasingly so as restructuring and funding cutbacks continue in the social services sector.

Lean Caring and the Erasure of Praxis Skills

In this study, several factors were identified that affect the development of praxis skills among social service workers. Central among these was workload. Cuts in staffing have resulted in an increase in case loads for most workers. "When I started here I had a case load of eighteen, and now it's thirty with no increase in resources," said one worker. Another observed that, "we've lost over fifteen positions since 1989 and none of the work; in fact, the work has increased ... including massive amounts of work downloaded from supervisors." Funding cuts have impacted negatively on staff and supervisors, resulting in reductions in the numbers of both. Many of the tasks previously completed by supervisors are now the responsibility of front-line workers who find there is less time and fewer staff to complete the work. Like other parts of the human services, the discourse of accountability has raced through the social services sector, and has been operationalized as increased documentation and statistics keeping (Baines, 2004b; Foster and Hoggett, 1999; Richmond and Shields, 2003) further expanding the work load of the front-line developmental services employee.

In addition to increased paperwork, the sector has seen an increase in both the absolute number of clients and the level of care required. This is due, in large part, to the general aging of the North American population which increases workload in the developmental services sector in two ways. Many of the clients who were part of the first wave of de-institutionalization and community care are now geriatric or near-geriatric and require increasing levels of care. In addition, many of the new clients entering the system have been living with their parents who are now too elderly to continue to provide care. All these clients require additional resources as they adjust to new environments. An increasingly geriatric client population increases lifting, feeding, and bathing responsibilities and necessitate the development of age-appropriate activities. As one worker noted, "it is a bigger work load partly because you have to spend more time doing everything. You can't expect an elderly man to put on his clothes as fast as a younger one. They just need more time and help." This expanded workload coupled with staff reductions is another factor contributing to staff overload and burn-out.

Funding cuts have exacerbated this situation resulting in a general deficit of programming and activity resources, as well as staffing and time. As one worker noted, "even if we had money to take these guys out on field trips and activities, who has time to plan anything?" Another added, "very little of the day is spent preparing or planning because there is no time for it. Most of the day is spent flying by the seat of your pants." Over-stretched workers and inadequate programming monies means that clients are often bored and inactive, while workers struggle to remain on top of their workloads creating an environment that is full of anxiety and the potential for frustration and outbursts. One worker commented that in the past they could provide services in a "proactive way"; now due to under staffing and lack of programming and planning funds, "instead of being proactive, you're always being reactive."

Increased workload impacts on workplace safety and skill loss in two ways. First, as noted earlier, workers depend on their detailed knowledge of each client, their mood changes, their likely triggers, interventions that deflect or diffuse conflict, what form an outburst might take, how long it will last, how often it will change form, and how best to assist the client in stopping the process, debriefing from it and moving on. This is not the kind of knowledge that can be developed when the pace and intensity of work is such that workers find they can do little but keep clients fed and relatively calm. Secondly, workers need to develop a sense of their co-workers strengths and abilities and ways of communicating among themselves, so that a clear sense of team work emerges, back-up plans can be confirmed, and violence prevention strategies can be enacted by a short phrase or quick word. Again, dependable relationships between co-workers cannot be developed when the pace and intensity of work means that communication is minimal and must be restricted to the reactive strategies noted above.

The near-mythical veteran worker, described earlier, who could admonish or encourage clients without turning her head to see what was going on, was part of one of the few remaining teams of workers who had worked together for a long period of time and were mostly full-time, permanent workers. The way this team worked together was evocative of a ballet in which each participant had her part fully memorized, and skillfully and almost intuitively, synchronized her movements to those of her co-workers and clients, as well as to the harmony of community care and client development. The work site was, by far, the most positive and creative program we observed in the course of this study. That is not to stay that anxiety and violence did not occur during our observations on this site, only that these moments, like the other events in this environment, were responded to in a well-choreographed way, in which disruption was minimized, other clients and workers were kept as safe as possible, the feelings of all present were attended to (including feelings about the outburst itself and its impact), and the work of the program continued. In short, this group of employees exceeded the expectations one might have of workers in the developmental sector by pooling their skills, using the best of each other's strengths, and trusting that, as a team, they could handle difficulties and promote growth among their clients, and themselves. It is quite possible that as the mostly full-time, permanent status of team members, and long history of working together was not replicated in other work sites, neither was the high level of skill, synergy of team strengths, and increased safety.

Another factor that exacerbates the loss of skills and violence prevention strategies in the leaned-out developmental service sector is the increased reliance on temporary and various categories of part-time workers including casual, crisis, on-call, contract, and regular part-time. Whereas full-time workers could expect benefits such as pensions, drug plans, and long-term disability (medical insurance is provided by the Canadian government as a right of citizenship), even among the better employers in the developmental sector, the various categories of part-time and temporary employees experience lower wages and few or no benefits. Economically, employees experienced increased job insecurity as well as decreased income and benefits, while the agencies save costs in these same areas. Thin shifting such as solo shifts, and access to fellow staffers only at other work sites (for example, other group homes) have become increasingly the norm, particularly for night shifts, which as the literature shows, increases the risk of violent attack (Chapell and DiMartino, 1998; NIOSH, 1996). Other forms of flexible work such as split shifts, wherein workers put in an hour or two at the beginning of the day and return in the evening for a few more hours, are also becoming prevalent and prove to be very disruptive to the family life of the predominantly female work force. One split shifter reported "barely seeing my kids", and another questioned how her family would "survive".

The inconsistent staffing, characteristic of flexible work forms, increases the likelihood of violence in the developmental work place in three ways--it disrupts clients, prevents the development of detailed knowledge of each clients, and curtails communication and strength pooling between and among staff, as well as the synergy of team work. Particularly in agencies where clients have communication disorders, even tiny changes in the everyday environment can precipitate high levels of anxiety which may result in lashing out and other violent behavior, much of which is likely unintentional. Intentional or not, the impact of assaultive behavior on workers and other clients can be serious and definitely increases the overall level of stress within a work place (Northwestern National Life, 1993). High levels of stress may spark new outbursts and even higher levels of anxiety, dropping the work/care site into a cycle of tension, flare-ups, increased tension and further escalations of frustration, inappropriate behavior, and increased potential for worker or co-client assault. As one front-line worker noted, "it's like a hamster wheel, the stress and violence go round and round feeding off each other and no one can get off it."

According to managers and staff alike, minimizing changes in the everyday environment is key to preventing the disruption of clients. Flexible staffing in which workers are employed for irregular shifts lengths (ranging from one and a half to twenty four hours or more), irregular days (for example, one at the start of the month and seven consecutively at the end), and a largely part-time work force that is shifted from program to program depending on demand, is antithetical to the aforementioned minimization of changes and client anxiety. In reality, flexible staffing looks like a formula to escalate client anxiety, and diminish client and staff safety. As one worker noted, "in the larger programs you used to have a number of staff to rely on ... to bring various skills to use to de-escalate a situation without resorting to physical restraint ... the ability to handle aggressive clients has been made harder, no doubt about it." Flexible staffing also means the staff are less likely to know or get to know the clients. They are also less likely to work with other staffers with whom they are familiar and who know they can depend on for consistent back-up or diffusion of tense moments. Thus, team building and the pooling of strengths was seen to be almost impossible.

Finally, lean work organization ensures there are few routes for skill transmission between and among workers, even if workers found a way to develop the high level praxis skills that helped to keep the work environment safer prior to the introduction of heavier workloads and flexible staffing. Few opportunities to observe or communicate with other staff, very little in the way of agency training, and a diminished sense of team culture cuts off the avenues previously used to share and build anti-violence prevention and intervention skills and strategies. Tense work environments also reduce morale and the likelihood that workers will communicate effectively with each other, or be open to new ideas and approaches.

Discussion and Conclusion

As noted earlier, the current discourse regarding professional skills is couched in terms of competencies. While this article has examined ways that new forms of work organization have inhibited and extinguished the praxis-skills that kept workers and clients safer in the developmental services, the current preoccupation with skill as competencies is another force that narrows social service practice, making it easier to increase work load and pace, employ a less skilled and temporary work force, and remove the spaces in which tacit-knowledge developed and flourished. As Dominelli and Hoogvelt (1996) argue, competencies "fragment complex processual tasks into discrete components" (p. 56) and remake a caring profession which had a fair degree of discretionary decision making power into a Taylorized, routinized form of labour (p. 56). Fragmentation is a form of deskilling which means that the work is easier to complete, hence professional credentials required for the job, and wages costs can be lowered. It is also makes it easier to supervise work hence management components can also be reduced resulting in further cost saving. Moreover, the remaining work can be monitored through a series of bureaucratic standards and technical controls, such as competency guidelines, rather than on-the-job, contextual interpretations, independent judgement, reflexivity, and a breadth of social, political, and professional skills. As Jackson (1997) argues, competency-based approaches use skill definition and skill training to "increase the subordination of the worker to the wishes of the employer" (p. 125). Although professional social work associations in the U.S.A. and Canada emphasize the pursuit of social justice as a pivotal facet of social work practice (National Association of Social Workers, 1999; Canadian Association of Social Workers, 1994), the process of defining core skills or competencies occurred within the context of employers who stressed cost reductions and value-added skills, and downplayed the social and political nature of skills, skill transmission, and how these skill sets may fit into social service workers' pursuit of social justice. As such, numerous skills, such as those used by the workers in this study, were made invisible and space has not been made for them in professional core-skill definitions or agency-level work organization plans.

Marie Campbell (2000) notes that restructuring in health care has removed the capacity of front-line nurses to participate in the informal planning, task coordination, relationship building, and mentoring that have been the bedrock of quality nursing care and practice-skill transmission. She warns that this knowledge may be permanently lost as control of the work process transfers increasingly to management, and away from the front-lines. With graver warnings of how the restructured care environment can result in skill loss, Geraldine Lee-Treweek's (1997) shows that many workers in long-ter care survive by viewing their increasingly alienating work as "just a job" rather than a contribution to caring for vulnerable populations (96-97). Within this defensive framing, workers survive their jobs by treating clients like objects and cutting themselves off entirely from their clients' pain and humanity (97). As the development of community-based care was an attempt to move away from inhuman warehousing approaches to people living with intellectual disabilities, it would be a large step backwards if workers were to lose their caring skills and capacities. Within the context of an increasing violent work place, the removal of the ability to keep oneself, one's client and co-workers safer is a sure way to increase worker alienation and frustration. It is difficult to imagine that this will not have a ripple effect on worker morale, empathy, patience, creativity, communication and other skills associated with caring labour.

While the loss of the tacit-skills described above certainly qualifies as one of the hidden costs of social service restructuring, the cost to teams of workers and individual workers in terms of increased stress, frustration, and injury is also high. As the findings from this study show, and workers and managers confirmed, clients are the victims of violence from other clients far more often than are workers. Thus, the loss of violence reduction skills and strategies extracts a high price from the most vulnerable people in the developmental services sector, namely clients who have few opportunities to seek other services, or means to resolve or redress assaults on themselves, their workers, and their friends. As financial savings in the short term seems to be the only possible benefit of the new forms of work organization, it appears that the costs--both social and individual--are far too high in the long term and have been extracted from those least able to resist or enact meaningful changes in their worlds.

References

Abramovitz, M. and A. Withorn. 1999. Playing by the rules: Welfare reform and the new authoritarian state. In Without justice for all, edited by Adolph Reed, Jr. Boulder, CO: Westview Press. 18(2): 9-24.

Arway, A. G. (2002). Causal factors of violence in the workplace: A human resource professional's perspective. In Violence at work. Causes, patterns and prevention, edited by Martin Gil, Bonnie Fisher and Vaughan Bowie. Portland, OR: Willan Publishing.

Baines. (2004a). 'Pro-market, non-market: The dual nature of organizational change in social services delivery'. Critical Social Policy. 24 (1): 5-29.

--. (2004b). Caring for Nothing: Work Organisation and Unwaged Labour in Social Services. Work, Employment and Society. 18(2):267-295.

--, K. Hadley, S. Pollack, B. Slade, A.-S. Brooker, K. Fay, W. Lewchuk, S. Preston, and D. Dimatrova. (2002). Improving work organization to reduce injury and illness: Social services, stress, violence and workload--final report. Hamilton: McMaster University Institute for Work in a Global Society. Baines, C., P. M. Evans, and S.M. Neysmith. 1998. Women's Caring. Feminist Perspectives on Social Welfare. Toronto: Oxford Press.

Boyd, C. 2002. Customer violence and employee health and safety. Work, Employment and Society 16:151-169.

Boyd, N. (1995) Violence in the workplace in British Columbia: A preliminary investigation. Canadian Journal of Criminology. 19(1):491-519.

Braddock, D. and R. Hemp. 1997. Towards family and community: Mental retardation services in Massachusetts, New England, and the United States. Mental Retardation. 34:241-256.

Brockman, M. and J. McLean. 2000. Review paper for the national task force: Violence against social care staff. London: National Institute for Social Work Research Unit.

Campbell, M. (2000) "Knowledge, Gendered Subjectivity, and the Restructuring of Health Care: The Case of the Disappearing Nurse." In Restructuring caring labour. Discourse, state practice and everyday life, ed. by Neysmith, S.M. Toronto: Oxford Press, pp. 186-208.

Canadian Auto Workers. (1994) Work reorganization: Responding to lean production. Toronto: CAW.

Canadian Union of Public Employees. (1999). Overloaded and underfire. Report of Ontario social services work environment survey. Ottawa: CUPE Health and Safety Branch.

Chappell, D. and V. Di Martino. (1998). Violence at work. Geneva: International Labour Organization.

Charlton, J. I. (1998). Nothing about us with us: Disability, oppression and empowerment. Berkeley: University of California Press.

Clarke, J. and J. Newman. (1992) The managerial state. London: Sage.

Darrah, C. (1997). Complicating the concept of skill requirements: Scenes from a workplace. In G. Hull (ed.) Changing work. Changing workers: Critical Perspectives on language, literacy and skillls. Boulder, CO: Westview Press.

Dominelli, L. and A. Hoogvelt. (1996) Globalization and the technocratization of social work". Critical Social Policy. Volume 16:45-62.

Donovan, R. (1987). Stress in the workplace: A framework for research and practice. Social Casework: The Journal of Contemporary Social Work. 68:259-266.

Duane-Richard, A.M. 2000. The social construction of skill. In Jenson, J., J. Laufer, and M. Maruani. The gendering of inequalities: women, men and work. Burlington USA: Ashgate.

Fabricant, M. B. and S. Burghardt. (1997) The Welfare State Crisis and the Transformation of Social Service Work. New York: M. E. Sharpe

Foster, D. and P. Hoggett (1999) "Change in the Benefits Agency: Empowering the Exhausted Worker?" Work, Employment and Society. Volume 13, Number 1:19-39.

Gardell, Bertil. 1982. Scandinavian research on stress in working life. International Journal of Health Services. 12:31--41.

Gaskell, J. 1990. What counts as skill? Reflections on pay equity. In Fudge, J. and P. McDermott (eds.) Just wages: A feminist assessment of pay equity. Toronto: University of Toronto Press.

Glazer, Nancy Y. 1993. Women's Paid and Unpaid Labour. The Work Transfer in Health Care and Retailing. Philadelphia: Temple University Press.

Herd, Pamela, and Madonna Harrington Meyer. 2002. Care work. invisible civic engagement. Gender and Society. 16:665-688.

Howe, Holly L. 1995. Homicide while at work. Differences in workplace homicides by sex, 1993. Bureau of Justice Statistics (http://www.bis.gove/oshwc/ cfar0001.pdf).

Jackson, N. 1997. Reframing the discourse of skill. In Kenway, J., K. Tregenza and P. Watkins (eds.) Vocational education today: Topical issues. Victoria: Deakin University Press.

Jenson, J. 1989. The talents of women, the skills of men: Flexible specialization and women. In Wood, S. (ed.) The transformation of work. London: Unwin Hyman Ltd.

Karasek R. and T. Thorell. (1990). Healthy work: Stress, productivity and the reconstruction of working life. New York: Basic Books.

Karasek, R. (1979). Job demands, decision latitude and mental health: implications for job redesign. Administrative Science Quarterly. 24:285-308.

Kinney, Jospeh A. 1995. Violence at work. Englewood Cliffs: NJ: Prentice Hall. Kohn, M.L. and C.L. Scholler (1983). Work and personality: An inquiry into the impact of social stratification. Norwood, NJ: Ablex.

Landsbergis, P.A., A.J. Cahin and P.L. Schnall. (1999). The impact of lean producation and related new systems of work organization on worker health. Journal of Occupational Health Psychology. 4(2):108-130.

Lee-Treweek, Geraldine. (1997) "Women, Resistance and Care: An Ethnographic Study of Nursing Auxiliary Work." Work, Employment and Society. Volume 11, Number 1:47-63.

Lincoln, Yvonne S. and Egon G. Guba. 1985. Naturalistic inquiry. London: Sage Publications.

Manwearing, T. and S. Wood. 1984. The ghost in the machine: Tacit skills in the labour process. Socialist Review. 14, 2, March: 69-80.

Messing, K. and J.P. Revert. (1983). Are women in female jobs for their health? A study of working conditions and health effects in the fish-processing industry in Quebec. International Journal of Health Services. 13:636-648.

Michalko, R. (2002). The Difference that Disability Makes. Philadelphia: Temple University Press.

Moody, K. (1997) Workers in a Lean World. Unions in the International Economy. London: Verso Books.

National Association of Social Workers. (1999). Code of ethics. Washington, D.C.: NASW.

National Institute for Occupational Health and Safety. (2000). Worker health chartbook, 2000. NIOSH Publication No. 2000-127. Atlanta, GA: NIOSH.

National Institute for Occupational Health and Safety. 1996. Violence in the Workplace. Atlanta: Centre for Disease Control.

National Institute for Occupational Health and Safety/National Occupational Research Agenda. 1995. Special Populations at Risk http://www.cdc.gov/ niosh/nrspop.html.

National Occupational Research Agenda. 1999. Work Environment and Workforce http://www.cdc.gov/niosh/worken.html.

Newhill, Christina E. 1996. Prevalence and risk factors for client violence toward social workers. Families in Society: The Journal of Contemporary Human Services October: 488-495.

Northwestern National Life. 1993. Fear and Violence in the Workplace: A Survey Documenting the Experience of American Workers. Minneapolis, MN: Northwestern National Life.

--. 1992. The Changing Organization of Work and the Safety and Health of Working People. Atlanta: Centre for Disease Control, Publication No. 2002-116.

Oliver, Michael. 1996. Understanding Disability: From Theory to Practice. New York: St. Martin's Press.

Ontario Public Service Employees Union. 1999. Developmental services sector questionnaire--preliminary results. Toronto: OPSEU.

Pizzino, Anthony. 2002. Dealing with violence in the workplace: The experience of Canadian unions. In Violence at work. Causes, patterns and prevention, edited by Martin Gil, Bonnie Fisher and Vaughan Bowie. Portland, OR: Willan Publishing.

Pizzino, A. (1994). Report on CUPE's national health and safety survey of aggression against staff. Ottawa: CUPE Health and Safety Branch.

Rey, Lucy D. 1996. What social workers need to know about client violence. Families in Society: The Journal of Contemporary Human Services. 16:33-39.

Roehr Institute. 1995. Harm's way: The many faces of violence and abuse against persons with disabilities. Toronto: C. Allan Roehr Institute.

Richmond, T. and J. Shields. (2003) NG Restructuring: Constraints and Consequences. Paper presented at the 11th Biennial Social Welfare Conference. Ottawa, ON.

Santana, Shannon A., and Bonnie S. Fisher. 2002. Workplace violence in the USA: Are there gender differences? In Violence at work. Causes, patterns and prevention, edited by Martin Gil, Bonnie Fisher and Vaughan Bowie. Portland, OR: Willan Publishing.

Schultz, Leroy G. 1987. The social worker as a victim of violence. Social Casework: The Journal of Contemporary Social Work. 6:240-244.

Sobsey, Dick. 1991. Disability, sexuality and abuse: An annotated bibliography. Baltimore: P.H. Brookes Publishing Co.

Stanley, Janet, and Chris Goddard. 2002. In the firing line. Violence and power in child protection. West Sussex, UK: John Wiley and Sons.

Strauss, A. L. and Corbin, J. (1990) Basics of qualitative research. Grounded theory procedures and techniques. Newbury Park CA: Sage Publications.

Taylor, Steven J., and Richard Bogdan. 1989. On accepting relationship between people with mental retardation and nondisabled people: Towards an understanding of acceptance. Disability, Handicap and Society. 4:21-36.

Taylor, Steve, and Melissa Tyler. 2000. Emotional labour and sexual difference in the airline industry. Work, Employment and Society. 14:77-95.

Tjaden, Patricia, Nancy and Thoennes. 1999. Violence by co-workers: Finding from the National Violence Against Women Survey. Washington, DC: National Institute of Justice.

Traustadottir, Ranveig. 2000. Disability reform and women's caring work. In Care Work. Gender, Labour and the Welfare State, edited by Madonna Harrington Meyer. London: Routledge.

Tully, Carol C., Nancy P. Kropf, and Janet L. Price. 1993. A study of violence in field placements. Journal of Social Work Education. 29:191-199.

DONNA BAINES

McMaster University

Labour Studies and Social Work