Getting the Story Straight: Illusions and Delusions in the Organizational Change Process[1]

Buchanan, David A

ABSTRACT

This paper considers the methodological implications arising from competing narratives of an organizational change process in a large acute city teaching hospital. This qualitative case study was informed by a processual-contextual perspective, and relied on an interpretive, constructivist epistemology. Two forms of contradiction are revealed. First, differing accounts were offered of substantive dimensions of the change programme. second, the impact of change on organizational effectiveness was indeterminate. This study suggests that the unitary, authentic narrative is illusory. Political motivations underpinning account-giving, and phenomenological variations in the lived experience of change, make competing narratives a naturally occurring phenomenon, not a methodological aberration. These findings have two main implications. First, case narrative validation through triangulation should be abandoned in favour of the pursuit of polyphony and ambiguity. second, the researcher faces the choice of being either an arbiter of accuracy, or of holding the less comfortable, more challenging, but creatively constructive role of exposing organizational tensions, disputes and contradictions.

SINGULARITY AND POLYVOCALITY

The point of social science is not to get it right but to challenge guidingassumplions, fixed meanings and relations. (Alvesson and Deete, 2000, p. 107)

Postmodern perspectives challenge singular or 'totalizing' theories or 'grand narratives' explaining social, political and economic phenomena, arguing instead for socially constructed views of reality based on multiple voices and interpretations (Reed, 1996; Clegg and Hardy, 1996). Lincoln and Denzin (1994, p.584) argue that we are listening to 'not one "voice" but polyvocality; not one story, but many tales, dramas, pieces of fiction'. Alvesson and Deetz (2000, p.34) observe that interpretive research increasingly questions, 'the logic of displaying a consensual unified culture', focusing instead on 'fragmentation, tensions, and processes of conflict suppression', and noting that, 'much more attention has been paid to the politics of representation and the role of the report author'. Clegg and Hardy (1996, p.696) argue that research should expose competing narratives of organizational phenomena, and not silence views that do not conform to conventional norms, or which threaten political positions.

The aim of this paper is to consider the nature and implications of competing narratives at the level of the organization, in the context of an attempt to understand strategic change within a processual-contextual perspective. Processual-contextual perspectives are probably the most influential (but not uncontested; Morgan and Sturdy, 2000) theoretical accounts of change (Pettigrew, 1985; 1987; 1988; Pettigrew and Whipp, 1991; Wilson, 1992). The unit of analysis is the process of change in context, which includes the external environment as well as the internal history, culture, structure, goals and politics of the organization. The process of change viewed through this lens is complex and untidy, shaped by interactions between the substance and process of change, and internal and external context factors. Processual-contextual perspectives advocate multi-layered longitudinal research, reported as rich case study narratives, such as Pettigrew's (1985) account of organization development and change in the chemicals company ICI. The work of Dawson (1994; 1996) and Clark (1995) follows in this tradition, depicting the non-linear, politicized and iterative nature of change. These perspectives contrast sharply with what Collins (1998) calls the '?-step guides' typical of the oversimplified, practitioner-focused managerial literature of change.

The notion of polyvocality is clearly recognized, but does not appear to have been adequately explored, by researchers studying organizational change using processual-contextual perspectives. In particular, the identification and reporting of competing narratives involve for the researcher awkward methodological issues which this paper seeks to address.

Dawson's work has been selected as the focus for critique in this paper for four reasons. First, Dawson's case accounts of change processes are methodologically rigorous, sensitive to differing stakeholder perceptions, and are consequently rich and detailed. second, Dawson's model explicitly highlights the shaping role of organization political factors in change. Third, Dawson consistently observes that change is characterized by 'competing histories'. While evident in the work of other commentators working with processual-contextual perspectives, these three features find particular emphasis and coherence in Dawson's work. Fourth, however, Dawson's empirical case accounts do not appear to expose or demonstrate clearly the competing histories emphasized in his theoretical perspective.

Dawson (1994; 1997) describes the use of complementary research methods to generate a 'chronology of events' (1994, p. 190) in each organization under analysis. The research underpinning this paper also relied on a multi-methods chronological narrative strategy (Langley, 1999). One strength of 'fine grain' qualitative case research, Langley argues, lies with accuracy. Dawson (1994, pp.188-9) thus describes how his processual-contextual accounts were systematically validated through a combination of regular organizational visits, observation notes and interview transcripts, cross-checked with annotated summaries of findings prior to producing'the final analysis' (1994, p.190; 1997, pp.399 and 402). This strategy reflects an overriding concern with 'getting the story straight', producing a singular account, relating 'what really happened' in the change process in all its complexity.

Dawson's emphases on the politicized nature of change and on competing histories reflect Burns' (1961 ) observations concerning the concealment and regulation of organization political behaviour through dual linguistic codes. Setting, public or 'backstage', determines the language appropriate for the articulation of organizational and personal goals. Burns (1961, p.260) observed that, The linguistic division, which is also a moral one, is particularly marked in universities, where mutually exclusive sets of expressions exist for discussion in faculty meetings or committees, and in bars, common rooms, or parties'. Burns' observations are echoed in the distinction made by March and Olsen (1983) between the 'rhetoric of administration' and the 'rhetoric of realpolitik'. The rhetoric of administration speaks of planning, economy, control, efficiency and effectiveness. The rhetoric of realpolitik speaks of contending interests, values, struggle, access and control. Dawson's observations, and the findings reported here, however, suggest polyphony or polyvocality, rather than duality concerning accounts of organizational change processes.

Describing the characteristics of a processual framework, Dawson notes that the approach can 'be used to accommodate the existence of a number of competing histories of the process of organizational transition' (1994, p.4; 1996, p.26; 1997, pp.390 and 400). This framework 'calls testimony to the complex and political nature of these processes and to the problems of data analysis which seeks to construct pathways of organizational change' (1994, p. 170). From a theoretical viewpoint, '[t]he existence of competing organizational histories of change [is] likely to complicate the most skeletal of explanations' (1994, p.182).

In Dawson's early case accounts, the inherent complexities of organizational change processes are admirably, and interestingly, illustrated in a range of settings. While Dawson points to predictable discrepancies between the perceptions of, for instance, senior managers and shop floor employees (e.g., 1997, p.399), his accounts do not display competing substantive histories of change processes. In contrast, it is the clarity, order and rich detail of Dawson's accounts which impress. In an apparent denial of his own perspective, Dawson contrasts 'the dominant or "official version" of change', with 'a true representation' (1994, p.4), or 'an accurate account' (1996, p.26) of the change process. The Official version', he argues, 'may often reflect the political positioning of certain key individuals or groups within an organization' (1994, p,4 and 1996, p.26). As Dawson's 'accurate account' may also reflect political positioning, the criteria for choice between these two versions are unclear.

In a more recent paper, exploring the political dimensions of change in a General Motors fabrication plant in South Australia, Dawson (2000, p.54) relates how, 'different groups may recount stories that reflect their own interests and perceptions of how change unfolds and their own position and influence over this process'. However (p.55), he also observes that, in this case, Over time, a common organization story has emerged which represents the view of the dominant narrators'. This 'dominant narrative', of course, 'may only represent the tip of the political iceberg', silencing unmade decisions, options which are not explored, and the voices which are never heard, but which still form part of the process.

We know that Dawson's case accounts have been systematically Validated', but it is not clear whether he is offering us the 'dominant' account, or an 'accurate and true' one, or the version of events that emerged from analysis of a complex qualitative data set, or simply an account which appealed to the researcher. As Czarniawska (1998; 1999) observes, writing accounts of case research is a craft involving skills not so distant from those of the novelist as might at first appear. Without accusing Dawson of fictionalizing his case histories, it is appropriate to ask, why are competing accounts not more in evidence ? There are at least two possible explanations. First, the research methods used, with respect to the range of informants and the duration and intensity of researcher-respondent relationships (Buchanan etal, 1984; Buchanan, 1994), may be inadequate to expose multiple accounts of the processes under investigation. This is unlikely, given Dawson's theoretical framework and his attention to methodological detail. second, untidiness and contradiction may be sidelined in the production of credible accounts for feedback to organization managers, and for other researchers. This latter explanation finds support in Dawson's comment concerning the 'daunting task of trying to prepare the material in a digestible form for publication' (1997, p.401 ), thus potentially producing a 'tidy rewriting of history'(Forster, 1994, p. 148).

In the research reported here, one account of a hospital re-engineering programme is first presented, based on interviews and documentation. Then, drawing on other information sources, contradictions concerning aspects of the process and its outcomes are revealed. Those contradictions sit alongside apparent consensus on other aspects of the programme. However, it appears that there is no single 'accurate account' of these events and their results.

The question which this paper addresses relies on two assumptions. First, that the observations of Dawson and others concerning competing narratives of change are valid. second, that the experience of this change programme is not idiosyncratic in this respect. What, then, are the methodological implications for case research on organizational change?

RESEARCH METHODS

The primary aim of this research was to understand the lived experience of change agents involved in a whole-hospital re-engineering programme. Many had been seconded for several months from clinical posts, and faced fresh demands and challenges in their new role. The findings concerning this experience have been presented elsewhere (Buchanan, 2000).

The Leicester Royal Infirmary NHS Trust (LRI) was an appropriate research site for several reasons. It was subject to constant external pressures and had a history of organizational innovation. The re-engineering programme was the first of its kind in Britain. There was widespread staff involvement. The change agenda continued until 2000. The site was thus chosen for instrumental reasons, offering potential for learning about the lived experience of change agency. It was also of intrinsic interest as a unique national pilot investigation into hospital re-engineering (Stake, 1994; Van Maanen, 1998). Data sources included:

* hospital documents;

* two Masters theses produced by hospital staff on university programmes;

* doctoral thesis written by the Re-engineering Programme Leader;

* briefing meetings with three management 'gatekeepers';

* interviews with 20 change agents, nine male and eleven female;

* feedback from a manager involved in reengineering, but not in this research.

Data collection began in 1999, with a document trawl, and with interviews during the final quarter of that year. Two external evaluations (unpublished) of the re-engineering programme became available to hospital management towards the end of 1999. Conversations with management gatekeepers were ongoing during this period, and continued into 2000, when an additional manager was invited to comment on draft reports of the study, based on her own close personal involvement in the re-engineering programme.

How can observations from a single case study contribute to understanding ? While a 'sample of one' prohibits statistical generalization, several commentators argue that single cases can be used to inform theory through analytical generalization (Mitchell, 1983; Tsoukas, 1989; Dyer and Wilkins, 1991; Yin, 1994; Butler, 1997; Buchanan, 1999). Tsoukas (1989) argues that studies of patterns of events in single cases can clarify structural aspects of social configurations, associated causal or 'generative' mechanisms, and contingent factors leading to observed behaviours. Dyer and Wilkins (1991 ) note that single case analysis can expose new theoretical relationships and question established thinking. This case account may have relevance for the 4,500 people employed by the hospital, and its catchment population of around one million. The aim of the programme was to generate lessons for the NHS whose hospitals share with LRI the external context and its historical, cultural, structural, cultural, professional and political features. It is reasonable to assume that the findings generalize to similar organizations in the NHS, and that the argument derived can inform the analysis of similar changes using processua!-contextual theoretical perspectives.

OUTRAGEOUS IMPROVEMENTS

A processual-contextual account requires analysis of the external and internal contexts and the substance and process of change. The first step in the research process, therefore, was to construct a picture of those contextual, substantive and processual elements, and to identify the chronology of events, from the inception of re-engineering at LRI prior to 1994, to the assessments of its impact carried out between 1997 and 1999.

LRI is one of the largest acute university teaching hospitals in Britain, with:

* an annual budget of £140 million;

* 1,100 inpatient beds;

* 4,200 employees;

* 110,000 inpatient and daycase visits a year;

* 130,000 accident and emergency attendances a year;

* 400,000 outpatient visits a year.

* a patient catchment area with a population of over 1,000,000.

External context

The National Health Service (NHS), with an annual budget of £50 billion and one million employees, is one of the largest employers in Europe. It has endured half a century of political intervention, now coupled with intense media scrutiny. The escalating cost of the service has embarrassed governments whose macroeconomic policies have focussed on curtailing public sec tor expenditure. Change in the 199Os was based on The Griffiths Report (DHSS, 1983), which argued that the service should be managed as a business, and healthcare units were invited to apply for Trust' status, giving managers a degree of local autonomy.

Media coverage of the NHS focuses on crises ('winter flu epidemic') and tragedies ('no bed, patient died'). Political debate focuses on funding and quality of service. The (Conservative) government of the 1990s encouraged consumerism, published service levels, and invited patients to complain when standards were not met. Since 1997, a new (Labour) government has launched several initiatives, such as reducing waiting lists for elective surgery, and clinical governance, a systematic approach to performance management. In 1999, a 'modernization' initiative was announced, to improve service quality through increased funding and changes in working practices.

Internal context

LRI was one of the first hospitals in Britain, in 1985, to adopt the clinical directorate management structure first developed in Baltimore, at the Johns Hopkins Hospital, and subsequently at Guy's Hospital in London (Buchanan and Wilson, 1997). LRI thus had an established tradition of medical engagement in hospital management. A new chief executive (CEO) was appointed in 1991, and LRI became an NHS Trust in April 1993.

In 1993, the Chairman of the local health authority challenged hospitals in the (Trent) region to undertake initiatives resulting in Outrageous improvements' to patient care. LRI launched five projects. Three failed, and two were 'spectacularly successful'(Bevan, 1997). Time to diagnosis for non-urgent neurology patients was cut from twelve weeks to five hours, with the introduction of a single-visit clinic. Turnaround for patients requiring hearing aids was cut from one year to six weeks. Could similar improvements be achieved across the hospital?

LRI had a history of underfunding. In the early 1990s, many new staff were recruited, and were critical of the facilities. One respondent said, 'When I came to Leicester, there were a lot of people coming in, medical staff, nursing staff, support staff, all of whom were saying what we have here is not acceptable. This isn't modern healthcare. We want to change it. We want to improve it.' The Chairman of the health authority and the new hospital CEO also supported the need for change, reinforced by government policy. Their aspirations were high. Re-engineering was to create 'the hospital of the future' in two years.

Substance

To co-ordinate the pilots, the CEO recruited in June 1993 a manager with private sector experience. Her analysis described how the projects which had failed had 'tinkered', while those which were successful had taken a holistic approach, and benefits had been gained by process re-engineering (Hammer and Champy, 1993; Davenport, 1993). (A non-technical term,'tinkering'has been elevated to a change management principle by Abrahamson, 2000.) The pilots were not initially designed as re-engineering initiatives and that label was attached subsequently; although beyond the scope of this paper, the ambiguity surrounding re-engineering methodologies should be noted (Pruijt, 1998; Valentine and Knights, 1998). The new Outpatient Project Manager became Re-engineering Programme Leader in November 1993. LRI then submitted a bid to the NHS Executive for funds to support the wholehospital re-engineering project, and was given a grant of £4.5 million. The project was to run for two years, on Michael Hammer's personal advice. Around £2 million was spent on management consultants and a similar amount paid for staff secondments and the re-engineering team. The rest was spent on redundancy and early retirement, affecting 24 staff, and on office costs, training, travel and other expenses (Bowns and McNulty, 1999, p.40). Over 1,000 employees were directly involved in process redesign, hundreds moved into completely new roles (LRI, 1997, p. 15), and the work of most hospital staff was affected.

The initial aim was to identify generic processes, such as patient visit, test, and stay, whose redesign could then be applied across the hospital. However, the diversity of patient 'pathways', and the different professional groupings with different views and preferences, suggested an alternative approach based on patient flows through individual clinical specialties. The main changes involved the appointment of process managers and process directors within clinical specialties, managing multidisciplinary teams responsible for the patient flow, 'front to back', through each specialty, rather than performing narrow functional or clinical tasks. Administrative tasks were also redesigned, for example creating clinic coordinators whose role combined several previously fragmented tasks.

Process

The programme ran from August 1994 until May 1996, through two broad phases. First, three reengineering laboratories were formed to redesign the hospital's generic processes (visit, test and stay). Identified by managers as being 'the brightest and the best', around six people were sec onded to each laboratory, full time, for six to nine months, along with the Programme Leader and an external management consultant. seconded staff included therapists, doctors, managers, porters, nurses and clerks. The number of laboratories was increased to four in February 1995 and by that summer there were 100 projects running. The programme was controlled by the Trust Board which had reporting to it a Re-engineering Steering Group chaired by the Trust Chairman. In February 1995, a Re-engineering Management Group was formed, chaired by the CEO, including managers, clinicians and team leaders, reporting to the Steering Group. Also in 1995, a Re-engineering Team Leaders Review group was established, chaired by the Programme Leader, and involving team leaders, key directors, and the CEO. This structure appears cumbersome, but it had to monitor and co-ordinate multiple initiatives over the two year period.

For the second phase, from September 1995, the laboratories were disbanded, and responsibility for re-engineering was transferred to clinical directorates. Some members left the hospital, some returned to their jobs, and some were appointed to process management roles. A small central team became the 'Centre for Best Practice' responsible for disseminating findings. Around 140 re-engineering projects were undertaken, but these were not the only changes. Bevan (1997, p.95) identified 68 other initiatives linked to re-engineering between 1994 and 1996. Table 1 summarizes the 'milestones' in the programme.

THE CHEMISTRY OF STRATEGY

Why did LRI do this ? all hospitals in the NHS face a similar external context. Why had most of them not embarked on similar internal change programmes ? At least eight factors contributed to the LRI strategy in 1994:

1. The 1992 Outrageous improvements' initiative of the Regional Chairman.

2. The ambitious new CEO who responded positively to that challenge.

3. The personal aspirations of the new Outpatient Project Manager, who quickly became Re-engineering Programme Leader (within six months of appointment).

4. The BPR approach fashionable in 1993 and 'compellingly attractive' to the Re-engineering Programme Leader (Bevan, 1997, p.108).

5. The support of the Trust Chairman (commercial background) and Medical Director (role model for doctors), both charismatic and highly influential figures.

6. The personal advice of Michael Hammer on the re-engineering proposals, including a crash BPR course with Hammer in Boston, involving managers and three doctors.

7. External management consultants recommending a BPR approach.

8. A grant of £4.5 million from the NHS Executive.

The answer thus lies, first, with a group of individual champions with high, clear and overlapping personal aspirations. They created an implementation process appropriate to this internal context. The answer also lies with factors political, financial, social - in the external context generating particular pressures for change, including a fashionable management literature promoting the substance of improvements.

LRI strategy appears to have been the result of a fortuitous combination, a 'chemistry', of contingent factors. This was not just the result of a traditional, rational, strategic management analysis (although the bid to the NHS Executive for supporting funds may have created that impression). This was not the gradual, emergent accumulation and consolidation of a series of prior small-scale initiatives. This was not an anarchic decision process characterized by confusion and uncertainty. The strategy formulation process was thus consistent neither with traditional rational models, nor with the incremental model of strategy formulation (Mintzberg, 1994), nor with the 'garbage can' model of decision making (March and Olsen, 1976). The roots of competing narratives of the strategic change process were thus nourished by this cocktail of internal and external pressures and events, which variously facilitated and constrained a range of personal and corporate agendas, driven in turn by a mix of personal and political ambitions and organizational goals.

ACCURACY AND INDETERMINACY

Interview responses concerning the lived experience of change revealed a consistent pattern. The nature and timing of re-engineering were (broadly) beyond dispute. The implications for job roles and careers were widely acknowledged. There was consensus that the attempt to implement a rapid programme of'punctuated' change (Tushrnan, Newman and Romanelli, 1986) had given way to a more modest, although ambitious and enduring, process of continuous improvement. However, identifying the 'accurate account' of 'what really happened' appears to be elusive and indeterminate. This indeterminacy was revealed in two main ways. First, in feedback from a hospital manager. second, in three formal assessments of the impact of the re-engineering programme on organizational effectiveness.

Management feedback

A draft report was given in June 2000 to a manager closely involved in re-engineering, but not initially part of this research. She was invited to check the draft for accuracy, and to comment on conclusions and interpretations. Her comments challenged four aspects.

First, the account noted that 'the brightest and the best' (Sevan, 1997) were seconded to re-engineering laboratories. There was an alternative version:

If was interesting to note the reference to 'the brightest and the best' beingseconded to the teams. From the inside - several colkagues had been seconded to the re-engineering teams with the warning that· this was a last opportunity to prove their worth. Some managers certainly seconded known troublemakers and those who couldn 't cope in their current roles. Interestingly, some of these people did very well in the changed environment.

However, another hospital manager independently offered yet another account of how the re-engineering teams were established, reflecting an untidy negotiation process which changed as the project evolved, ratherthan a systematic selection or'nomination1 procedure:

What we did was, ive chose the leaders to start with, and we fried to pick fairly senior and credible people in the organisation, and then worked together to work out who would be good people to have on that team. Audit was a process of negotiation. So other people in the organisation would be saying, this person's good, that person s good. .And sometimes people were good and sometimes they weren 't. Buf when we go t to the second stage of the project, what we mere learning was, actually you had to have people who could influence, people who could go in and talk to a senior nurse, and be able to say to that person, you know, suggest making changes and not be frightened, and actually hating the influence and the credibility to do that.

Second, it was noted that around 140 réengineering projects took place between 1994 and 1996 (LRI, 1997; Bowns and McNulty, 1999). There was an alternative version:

One of the enduring criticisms is the amount, of projects that reengineering took credit for that had absolutely no link with reengineering at all. Memo s used to regularly appear on managers ' desks asking for a breakdown of every change in practice that ivas either occurring or planned. For example, the consultants in [our area] hadpkmnedto anpkmentan initiative to increase day case activity, and the new unit was set up and piloted by the nursing team. [The central re-engineering team] go t to hear about it and suddenly it ivas presented as a re-engneeringpilot and credited to the re-engineenng team. That was a common occurrence and one that did serious damage to the who-kprogramme. I was ? 'ery angry about that because it took credit from staff who had worked really hard to make a project happen. Some consultants still refuse to have anything to do with the Centre for Best Practice because of similar experiences.

Third, formal assessments found that one of the 'hidden benefits' of re-engineering concerned embedding a culture of change (LRI, 1997; Bowns and McNulty, 1999). There was an alternative version:

Despite some of the illusions and delusions of that period, rapid change did occur and certainly enhanced career opportunities for many. My experience has been that rapid change didn '( result in embedded change. Once change agents had done their work and delivered the goods and left, the leadership often wasn 't there to embed, sustain and develop any further. Several redesigned roles and services are reverting gradually to pre-re-engineering days. Maybe multiple initiath (es occur now because ire didn 'tget it right in the first place.

However, another hospital manager independently offered a different account, focusing on the supporting roles of key managers, and arguing that development continued in their absence:

I think a lot of things needto movefonvard because they'll evolve anyway. I think the bigger danger is not at that level, teamwork, but actually at the top of the organisation. Because I think that one of the most important things about re-engineering, when you compare it to what was happening in other hospitals, was the absolute consistency at the lop of the organisation, that this was apriority, and to keep it moving forward. And you saw that in [the CEO], you saw that in [the Medical Director], you saw that in [the Director of Human Resources]. What's really good is that, even though [the initiators] have gone, it's still moving forward. There are still lots of developments going on there, and to me that's a real test.

Fourth, interviews suggested that those seconded into change agency roles experienced significant personal and career development, thus creating a cadre of experienced organizational innovators. There was an alternative version:

Under personal development, opportunities to move from functional specialism to management career - helps develop a cadre of skilled innovators - it could be argued it also depletes the pool of skilled clinical staff who, I believe, are going to have far more opportunities in the future than managers no matter how skilled. Maybe things will completely turn around due to change fatigue and instead of change agents we'll be employing change police - people who prevent changes occurring.

However, one manager again contradicted this view with a different personal perspective:

I mean, people have said things like, it's five years' worth of personal development in six months. And it was very, very focused and it was extremely fast moving. J think with me it was just having the experience and the opportunity to learn and to create improvement, even though some of the things that I did were a bit, looking back, a bit misguided. Because I see a lot of other people around the NHS now who are also in full time change roles, but because I had five years with LRI, I've got things that nobody else in the NHS has got because I have that experience. You can read books, and you can go to workshops, but you can't get what I've got very easily.

Is this contradictory feedback based on an idiosyncratic standpoint? The manager had no personal stake in giving this feedback (and subsequently left the hospital), and also observed that her experiences and views were shared by many other staff.

Formal evaluations

How successful was re-engineering in achieving the aims of whole-hospital redesign, with 'previously unachieved' levels of efficiency and service quality? Two independent reviews were commissioned. One focused on 'macro [whole hospital] measures' (Brennan, Sampson, Hemsley and Evans, 1999). The other evaluated 'implementation and impact' (Bowns and McNulty, 1999). These were preceded by the hospital's own internal assessment (LRI, 1997).

The macro measures study tracked hospital performance from 1994-95 (baseline) until 1997-98 (post-implementation), and compared performance with other hospitals. The analysis covered operating and staff costs, productivity, inpatient activity, quality indicators (waiting times, re-admission rates), and other resource and output measures. This revealed that LRI was a comparatively efficient hospital in 1994-95, that on some measures performance had improved, while on others performance had decreased. Despite 62 pages of text and 54 pages of detailed statistical tabulation, the report offers no overall evaluation. Brennan et al. (1999, p.6) comment instead that, 'It will clearly not be possible to disentangle the effects of re-engineering from other general initiatives and improvements in efficiency at the macro level'.

The implementation and impact study took a broader view, considering qualitative and quantitative results. This report concluded that although hospital re-engineering was particularly complex, it had been a catalyst for change, and quality of care had improved. However, a subheading summarized the conclusion; 'changed but not transformed'. Bowns and McNulty (1999, p.41) argue that, 'there is little evidence of the dramatic transformation of the performance of the hospital, routine quality indicators remain broadly stable [and] the general picture is of marginal improvements in most of the main traditional indicators of efficiency'. They also conclude (p.4) that, 'the redesign of patient care processes has not resulted in sufficient savings to consider the initiative to have paid for itself.

The internal report claimed recurrent annual cost savings of £900,000 and capacity increases of 20 per cent in some processes (LRI, 1997, p.3). Of the 140 projects reviewed, 120 delivered positive results, in five categories: quality improvement, cost improvement, capacity, generalizable benefits (releasing staff time, creating more bed nights), and 'hidden' benefits. While 64 per cent of projects delivered quality improvements, only 30 per cent delivered capacity increases, and only 8 per cent achieved cost improvements. In diagnostic endoscopy, for example, time from first patient visit to diagnosis was cut from 30 weeks to 5 hours as the patient process was redesigned into a single visit, increasing the capacity of the service by over 130 per cent, creating more time for teaching and research, and increasing the proportion of time nurses spent on direct patient care from 10 to 70 per cent.

However, 57 per cent of projects achieved 'hidden' benefits which included the implementation of team working, performance management systems, skills growth, and improved decision making. The single main hidden benefit identified by this report was the creation of, 'the organizational conditions under which patient process based improvements can be sustained and new projects initiated' (p.8). The report argued (p.2) that, 'patient process redesign work continues. Indeed, the depth and breadth of improvements achieved through redesign have demonstrably accelerated following completion of the formal project'.

These contradictions imply that the description and evaluation of complex strategic change is unstable and indeterminate. The notion of one unitary, accurate, authentic account of the change process and its outcomes is a delusion. Such acts of 'account giving' (Bies and Sitkin, 1992; Read, 1992) are politically charged. Claims to the 'official' or 'accurate' account of strategic change and its consequences must therefore be treated as partisan and suspect.

The indeterminacy of the change process and its results may also influence whether a change programme of this nature can be sustained. In April 2000, LRI merged with two other hospitals in Leicester to become part of University Hospitals of Leicester NHS Trust. Anew Chief Executive and management team were appointed, including only a small number of those involved in re-engineering, and conspicuously excluding champions of the programme (such as the Medical Director). In September 2000, the new CEO announced a new hospital management structure, based on the traditional clinical directorate model. Process directors and process managers were invited to apply for traditional business management roles. Many applied for posts in other organizations. The inability unambiguously to establish the benefits of re-engineering may have contributed to its demise.

LET'S BE CLEAR ABOUT UNCERTAINTY

The evidence from this case study appears to confirm Dawson's (1994; 1996; 2000) observations concerning competing narratives of change, which potentially arise from three main sources, in addition to the contextual pressures and demands explored earlier.

One source concerns complexity. LeviStrauss (1968, p. 134) noted that it is common for informants to offer different accounts of social structures to anthropologists because social organization 'is too complex to be formalized by means of a single model'. A radical whole-hospital change programme costing £4.5 million, with (allegedly) 140 re-engineering projects and 68 other initiatives over two years affecting around 4,500 staff, is relatively complex. Processes of strategic organizational change thus also appear to be characterized by layers of complexity which are difficult to capture in any simple or singular interpretation or argument.

A second source concerns phenomenological variance in the lived experience of change, contingent on organizational roles and personal frames of reference; how individuals position themselves as sponsors, victims, survivors, spectators, or drivers. Interviewees in this case held a variety of positions across the organization before their involvement in reengineering. Some were involved reluctantly while others welcomed the role. This variance is compounded by the social context and timing of data collection. The researcher often gathers information, through documentary evidence and interview, concerning actors' understandings of events as they are occurring, in this case through documentation. Other evidence, collected later, when some of the consequences of the change are known, is likely to be more reflective.

Third, as Alvesson and Deetz (2000, p. 131) observe, empirical data are 'impregnated' with politics, with the struggle for the justification and dominance of particular values, points of view and interests. Organizational change is a value laden issue; radical changes raise the political stakes and heighten political behaviour by threatening established interests (Dawson, 1996; Buchanan, 1997; Buchanan and Badham, 1999). Pettigrew (1985) analyses change in terms of the management of meaning, implying the need for a range of symbolic actions, including the construction of accounts which portray change as a success. The internal hospital evaluation of reengineering (LRI, 1997), not surprisingly, highlights the benefits.

Given these intertwined sources of competing narratives and interpretations, it follows that researchers in this field should perhaps be more surprised by and sceptical of the absence of ambiguity and contradiction than by the presence of these features. This conclusion appears to have at least two main methodological implications for the study of organizational change processes, concerning research aims and methodology, and research reporting, respectively.

Research aims and methodology

'Getting it straight' is an illusory goal. The question, 'what really happened ?', is irrelevant. This study, along with other evidence and commentary, suggests that there is no one authentic 'true' narrative of change. This is not, however, a counsel of despair, as this argument is consistent with the view that narratives are an important, if overlooked, source of insight and understanding in their own right (Putnam, Phillips and Chapman, 1996; Butler, 1997; Gabriel, 1998; Czarniawska, 1998 and 1999;). As Putnam et al. observe (1996, pp.386-7):

"Narratives are ubiquitous symbols that are prevalent in all organizations. Also referred to as s tories, scripts, myths, legends and sagas, narratives are accounts of events, usually developed chronologically and sequentially to indicate causality. [..] They are the vehicles through which organizational values and beliefs are produced, reproduced, and transformed. They shape organisational meanings through functioning as retrospective sensemaking, serving as premises of arguments and persuasive appeals, acting as implicit mechanisms of social control, and constituttngframes of reference for interpreting organisational actions."

The researcher seeking to validate a single coherent account of change by data triangulation is in danger of generating, at best, a partial rendering of the processes under investigation or, at worst, a partisan version which reflects the views of a limited range of actors. (Hamel, 2000, for example, offers an exemplary univocal account of change at IBM.) One central research aim in studying organizational change processes should thus concern the exposure of competing narratives, polyvocality, contradiction, ambiguity, disagreement, and uncertainty.

Competing narratives are naturally occurring phenomena, not aberrations to be triangulated away methodologically. Exposing competing narratives, including 'official' accounts and otherwise subversive and silenced voices, requires the use of a broad range of data collection sources and methods, considering also the nature of the interpersonal relationships formed with respondents. In this case, multiple access points to the organization were used, including managers known personally to the researcher, as well as formal gatekeepers, interviewees, and unpublished reports; further investigation potentially would have disclosed still more contrasting views. The account(s) offered here cannot be regarded as 'definitive' with respect to the competing narratives of this particular hospital change programme.

Research reporting

The singular, coherent account which fails to expose conflicting views of the change process is deeply suspect. Dawson (2000, p.55) notes how the 'emergent dominant narrative', the 'official' managerial version, of change 'downplays conflicts and avoids notions of failure whilst simultaneously applauding success and emphasizing collaboration'. Such accounts of change confirm a view, support a position, back an interest, validate a perspective, legitimate certain decisions and ideas. The 'dominant' view of change, the 'official' version, is thus a politically motivated vehicle which aspires to justify decisions and lines of action, to deny complexity and subtlety, to silence other voices and interpretations, and to maintain power positions. The researcher seeking to establish a singular position is thus implicated in the internal politics of the organization under investigation. Pursuit of the single 'accurate' account also creates potential for contradiction between the accounts of respondents and researchers. Dawson (2000, p.55) argues that, The creation of these stories [describing change processes] are themselves a political process and as such, characters can be recast and story lines changed to fit current contextual conditions and future strategic objectives'. The 'validated' account may contradict organization members concerned to preserve the validity of their own experience, coloured by their roles in the social structures being studied.

Faced with uncertainty and disagreements over substantive aspects of change processes, the researcher faces a choice of being either an arbiter of accuracy, or of accepting the role of exposing tensions and contradictions. The former role is consistent with the institutionalized stereotype of the researcher as neutral, objective, external 'expert' trained to collate and to analyse, to uncover truth, to find correct answers, to 'get the story straight'. This entails crafting one 'best' interpretation from available data, constructing a 'sound argument' and a 'compelling case' in relation to a theoretical framework for presentation to an academic readership. Research access is often contingent on providing feedback to an organization in the form of practical recommendations, again based on evidence and 'sound argument'.

Exposing uncertainty and contradiction is a considerably less comfortable stance. This research strategy generates ambiguity, ambivalence, destabilization, and lack of coherence and closure. The research task is more challenging. The 'evidence' is inconclusive. The 'expert' role has to be abandoned. 'Experts' are not expected to encourage confusion and uncertainty, but to resolve ambiguity and disagreement. Inability to offer a coherent, accurate account of events may inhibit publishing plans (in some academic journals). The researcher may also have the problem of translating findings into evidence-based advice. An inability to offer clear and unambiguous practical feedback from the research to a host organization may discredit the researcher, and contaminate that site for future study.

The researcher traditionally claims (tacit) authority, implicitly assuring readers that what is being presented is valid knowledge based on training and experience in the systematic use of methodological procedures. Alvesson and Deetz (2000, p. 135) argue that, 'such an attitude appears increasingly problematic', as researchers require a combination of political awareness and openness to competing interpretations. For Alvesson and Deetz, this entails the development of a 'non-authoritative form of research and writing', being clear about the researcher's preferences, social position and role, and theoretical orientations. To enable readers to understand how accounts have been constructed, and to encourage and facilitate differential reader interpretations, the researcher as narrator has to adopt a more reflexive voice, exposing the conceptual lenses and filters used to represent theories, models and analytical frameworks (Chia, 1996; Collins, 2000), emphasizing that findings are less stable and clear-cut and are open to discussion and (re)interpretation. Readers should perhaps be reminded that, 'what is offered is one story - at best empirically sensitive and well-grounded, and full of insights and theoretical contributions but still open to other readings and informed by other perspectives, interests, or creative powers' (Alvesson and Deetz, 2000, p.136).

This shift in the researcher's position implies a different audience perspective. Readers are invited actively to seek elements and ideas and interpretations which are personally meaningful and useful, in a mode described by Stake (1994) as 'naturalistic generalization', ratherthan passively receiving the researcher's 'authoritative' view. Alvesson and Deetz (2000, p.135) describe this as 'activating the reader', which in turn leads to, 'a downplaying of researcher authority and a reduction of the asymmetry between researcher and reader'.

Rejecting criticism of postmodern perspectives as offering only the nihilism of indeterminacy, Hatch (1997) offers constructive advice; avoid One right answer', dispute categories, challenge assumptions and claims to truth, maintain critical distance, take nothing for granted, imagine alternatives. Postmodernism, in denying the comforts of fixity, stability and clarity, can be a creative tool for organizational innovation. Collins (1998) similarly advises organizational change researchers to adopt less partisan and more critical positions. The challenge, therefore, is to present competing narratives as opportunities for learning and innovation, not to elicit judgement. The sources of and influences on competing narratives should thus attract more attention than efforts to generate tidy, coherent accounts. Learning from the differing interpretations of actors and observers is of more value than attempts to resolve or reconcile such contradictions. Encouragement to challenge, to imagine, to rethink, and to generate different organizational possibilities is of more value to academic and managerial readers than a tidy summary of findings and a checklist of action points.

FOOTNOTES

[1] The creatively constructive comments and criticisms of Alan Bryman, Robert Chia, David Collins, Patrick Dawson, Juha Laurila and Jane Parker on drafts of this article are all gratefully acknowledged. The errors and flaws, however, are the responsibility of the author.

REFERENCES

Abrahamson, E. (2000). Change without pain. Harvard Business Review,78(4), 75-9.

Alvesson, M. & Deetz, S. (2000). Doing critical management research. London: Sage Publications.

Bevan, H. (1997). Managing today while creating tomorrow: actionable knowledge for organizational change in a hospital. Unpublished doctoral thesis, Henley Management College and Brunel University, January.

Bies, R.S. & Sitkin, S.B. (1992). Explanation as legitimation: excuse-making in organizations'. In M.L. McLaughlin, M.J. Cody & S.J. Read (eds), Explaining one's self to others: Reason-giving in a social context. Hillsdale NJ: Lawrence Erlbaum, 183-98.

Bowns, I.R. & McNulty, T. (1999). Re-engineering Leicester Royal Infirmary: an independent evaluation of implementation and impact. School of Health and Related Research, University of Sheffield, and Centre for Creativity, Strategy and Change, University of Warwick.

Brennan, A., Sampson, R, Hemsley, J. & Evans, M. (1999). Evaluation of business process reengineering at Leicester Royal Infirmary: final report on changes between 1994/5 and 1997/ 8 - macro measures study. School of Health and Related Research, University of Sheffield.

Buchanan, D.A. (1993). Recruitment mode as a factor affecting informant response in organizational research. Journal of Management Studies, 30(2), 297-313.

Buchanan, D.A. (1997). The opportunities and limitations of business process re-engineering in a politicized organizational climate. Human Relations, 50(1), 51-72.

Buchanan, D.A. (1999). The logic of political action: an experiment with the epistemology of the particular. British Journal of Management, 10(special issue), September, 73-88.

Buchanan, D.A. (2000). The lived experience of high velocity change: a hospital case study. Paper presented to the American Academy of Management Conference, Symposium on Strategy as Dynamic and Pluralistic, Toronto, August.

Buchanan, D. & Badham, R. (1999). Power, politics, and organizational change: Winning the turf game. London: Sage Publications.

Buchanan, D.A., Boddy, D., & McCalman, J. (1988). Getting in, getting on, getting out and getting back. In A. Bryman (ed.), Doing research in Organizations. London: Routledge, 53-67.

Burns, T. (1961). Micropolitics: mechanisms of institutional change. Administrative Science Quarterly, 5, 257-81.

Butler, R. (1997). Stories and experiments in social inquiry. Organization Studies, 12(6), 927-48.

Chia, R. (1996). Teaching paradigm shifting in management education: university business schools and the entrepreneurial imagination. Journal of Management Studies, 33(4), 409-28.

Clark, J. (1995). Managing innovation and change: People, technology and strategy. London: Sage Publications.

Clegg, S.R. & Hardy, C. (1996). Organizations, organization and organizing. In S.R. Clegg, C. Hardy & W. Nord (eds), Handbook of organization studies. London: Sage Publications, 1-28.

Collins, D. (1998). Organizational change: Sociological perspectives. London: Routledge.

Collins, D. (2000). Management fads and buzzwords: Critical-practical perspectives. London: Routledge.

Czarniawska, B. (1998). A narrative approach to organization studies. Thousand Oaks, CA: Sage Publications.

Czarniawska, B. (1999). Writing management: Organization theory as a literary genre. Oxford: Oxford University Press.

Davenport, T., H. (1993). Process innovation: Reengineering work through information technology. Boston, Mass: Harvard Business School Press.

Dawson, P. (1994). Organizational change: A processual approach. London: Paul Chapman Publishing.

Dawson, P. (1996). Technology and quality: Change in the workplace. London: International Thomson.

Dawson, P. (1997). In at the deep end: conducting processual research on organizational change. Scandinavian Journal of Management, 13(4), 389-405.

DHSS, (1983). NHS management inquiry. London: Department of Health and Social Security, Her Majesty's Stationery Office.

Dyer, W.G. & Wilkins, A.L. (1991). Better stories, not better constructs, to generate a better theory: a rejoinder to Eisenhardt. Academy of Management Review, 16, 613-19.

Forster, N. (1994). The analysis of company documentation. In C. Casseli & G. Symon (eds), Qualitative methods in organizational research: A practical guide. London: Sage Publications, 147-66.

Gabriel, Y. (1998). The use of stories. In G. Symon & C. Casseli (eds), Qualitative methods and analysis in organizational research: A practical guide. London: Sage Publications, 135-60.

Hamel, G. (2000). Waking up IBM: how a gang of unlikely rebels transformed Big Blue. Harvard Business Review, 78(4), 137-46.

Hammer, M. & Champy, J. (1993). Reengineering the corporation: A manifesto for business revolution. London: Nicholas Brealey Publishing.

Langley, A. (1999). Strategies for theorizing from process data. Academy of Management Review, 24(4), 691-710.

Levi-Strauss, C. (1968). Structural anthropology. London: Allen Lane, The Penguin Press, edited by C. Jacobson & B. Grunfest Schoef.

Lincoln, Y.S. & Denzin, N.K. (1984). The fifth moment. In N.K. Denzin & Y.S. Lincoln (eds), Handbook of Qualitative Research. Thousand Oaks, CA: Sage Publications, 575-86.

LRI. (1997). Re-engineering In healthcare: The Leicester Royal Infirmary experience, The Leicester Royal Infirmary NHS Trust, September.

March, J.G. & Olsen, J.P. (1976). Ambiguity and Choice in Organizations. Oslo: Universitetsforlaget.

March, J.G. & Olson, J.P. (1983). Organizing political life: what administrative reorganization tells us about government. American Political Science Review, 77(2), 281-96.

Mintzberg, H. (1994). The rise and fall of strategic planning. New York: Free Press.

Mitchell, J.C. (1983). Case and situational analysis. Sociological Review, 31(2), 187-211.

Morgan, G. & Sturdy, B. (2000). Beyond organizational change. London: Macmillan.

Pettigrew, A.M. (198). The awakening giant: Continuity and change in ICI. Oxford: Basil Blackwell.

Pettigrew, A.M. (1987). Context and action in the transformation of the firm. Journal of Management Studies, 24(6), 649-70.

Pettigrew, A.M.(ed.). (1988). The management of strategic change. Oxford: Basil Blackwell.

Pettigrew, A.M.. & Whipp, R. (199). Managing change for competitive success. Oxford: Basil Blackwell.

Pruijt, H. (1998). Multiple personalities: the case of business process re-engineering. Journal of Organizational Change Management, 11(3), 260-8.

Putnam, L.L., Phillips, N. & Chapman, P. (1996). Metaphors of communication and organization. In S.R. Clegg, C. Hardy & W. Nord (eds), Handbook of Organization Studies. London: Sage Publications, 375-408.

Read, S.J. (1992). Constructing accounts: the role of explanatory coherence. In Margaret L. McLaughlin, Michael J. Cody & Stephen J. Read (eds), Explaining One's Self to Others: Reason-Giving in a Social Context. Hillsdale, NJ: Lawrence Erlbaum, 3-19.

Reed, M. (1996). Organizational theorizing: a historically contested terrain. In S.R. Clegg, C. Hardy & W. Nord (eds), Handbook of Organization Studies. London: Sage Publications, 31-56.

Stake, R.E. (1994). Case studies. In N.K. Denzin & Y.S. Lincoln (eds), Handbook of Qualitative Research. Thousand Oaks, CA: Sage Publications, 236-47.

Tsoukas, H. (1989). The validity of idiographic research explanations. Academy of Management Review, 14, 551-61.

Tushman, M.L., Newman, W.H. & Romanelli, E. (1986). Convergence and upheaval: managing the unsteady pace of organizational evolution. California Management Review, 29(1), 29-44.

Valentine, R. & Knights, D. (1998). TQM and BPR - can you spot the difference ? Personnel Review, 27(1), 78-85.

Van Maanen, J. (1998). Different strokes: qualitative research in the Administrative Science Quarterly from 1956 to 1996. In John Van Maanen (ed.), Qualitative Studies of Organizations. Thousand Oaks, CA: Sage Publications, ix-xxxii.

Wilson, D.C. (1992). A strategy of change. London: Routledge.

Yin, R.K. (1994). Case study research: Design and methods. Thousand Oaks, CA: Sage Publications.

by David A. Buchanan

School of Business, Department of Human Resource Management,

De Montfort University, UK

Copyright TamaraLand Publishers 2003
Provided by ProQuest Information and Learning Company. All rights Reserved