1 |
Research on Organisational Change and Development and Strategic Change: Some Limitations |
Change is ubiquitous. Or is it? In the micro-events which surround our particular lives and in the daily trumpetings of the media change has an ever-present illusion of reality. Yet observe other men consciously attempting to move large and small systems in different directions, or attempt it yourself, and one sees what a difficult and complicated human process change is. And there is the problem of perspective. Where we sit not only influences where we stand, but also what we see. No observer of life or form begins with his mind a blank, to be gradually filled by evidence. Time itself sets a frame of reference for what changes are seen and how those changes are explained. The more we look at present-day events the easier it is to identify change; the longer we stay with an emergent process and the further back we go to disentangle its origins, the more we can identify continuities. Empirically and theoretically, change and continuity need one another, although as we shall see, many social scientists in their attempts to identify and explain change in organisations in terms of the micro-events of the day have artificially abstracted change out from the structures and contexts which give that change form, meaning, and dynamic. Change and continuity, process and structure, are inextricably linked.
A further conundrum is the choice of metaphors and images which exist in the study of change, each one pregnant with its own world view, model of man, and explanatory language. How are we to choose between growth and development, continuity and flow, life cycle and phase, contradiction, intrusion, and crisis? Are we to go along with the social analysts who would have us believe that cultures grow and perish in the same superb, natural, and aimless fashion as the flowers of the field? Or are we to accept the assumption that human history and social change are about chaps – and nothing else? Gellner's (1973) view is that history is about chaps, but he is quick to point out that it does not follow that its explanations are always in terms of chaps. Societies are what people do, but social scientists are not biographers en grande série.
Beware of the myth of the singular theory of social or organisational change. Look for continuity and change, patterns and idiosyncrasies, the actions of individuals and groups, the role of contexts and structures, and processes of structuring. Give history and social processes the chance to reveal their untidiness. Arguments over the true or basic sources of change, while interesting and worthwhile in the sharpening of academic minds and egos, are ultimately pointless. For the analyst interested in the theory and practice of changing the task is to identify the variety and mixture of causes of change and to explore some of the conditions and contexts under which these mixtures occur. That at least is the first stepping stone I offer the reader for joining me on this particular enterprise.
This research on continuity and change in ICI has provided an opportunity to study in a comparative and longitudinal mode the activities and strategies of very senior and middle managers trying to create significant change, and the role of internal and external organisation development (OD) specialists in facilitating change. The study began in 1975 after ICI had been using specialist OD resources at a variety of different levels and divisions in the company, and were pondering particular questions about why and how these specialists seemed to be more effective in some parts of the company than others. From this initial pragmatic question the research began to explore and establish the first theme in this book – the contributions and limitations of specialist-based attempts to create changes in organisational structure and culture. The first process under examination in the research has been the birth, evolution, development, and impact of groups of internal and external OD consultants in four divisions, and the corporate headquarters of ICI. However, as a result of the access and analyses provided by the initial study objectives the opportunity arose to explore a second and more inclusive process, the long-term processes of strategic decision-making and change in ICI in the differing social, economic, and political context of 1960–83. The use of the word strategic linked to change is just a description of magnitude of alteration in, for example product market focus, structure, and organisational culture, recognising the second-order effects, or multiple consequences, of any such changes. This second, broader analysis of continuity and change is seen through the eyes and activities of the main board of ICI, and the boards and senior managers of ICI's four largest divisions – Agricultural, Mond, Petrochemicals, and Plastics.
Given that the completed study incorporates an analysis of senior executive and specialist contributions to the process of formulating and implementing changes of such scope and impact that they can justifiably be described as strategic; and examines the role of line managers and specialists grappling with the introduction of more localised and circumscribed changes; and that OD techniques and ideas were applied to both the strategic and the more limited changes, then it is evident the data in this study are potentially connectable to a number of the various elements of the literature on the management of change. Unfortunately the writing and research on the management of change is a sprawling and none too coherent collection of bodies of literature, some proclaiming pragmatic precepts and transparent values, others coolly analysing and theorising without any commitment to statements of practice, and others being concerned with the lofty analytical formulation of strategic change and disdaining the examination of how such strategies are to be implemented in the political and cultural mosaic of large and medium-sized organisations. Of course, a further and predictable difficulty with a topic like organisational change, is that the study of change has attracted scholars from a number of academic disciplines who identify themselves as being more or less interested in description or prescription; a consequence of this is that these different bodies of literature do not, explicitly at least, talk to one another.
In what follows no attempt is made to find a herculean and grand coherence in the literature on the management of change where none exists. Rather the more limited objective for a review is offered of description, explication, and where appropriate criticism. The aim is to lay out some parts of the writing and research on organisational change and development in the hope that some of the strengths and weaknesses of that literature are clarified, and in the expectation that the identification of such patterns will serendipitously encourage the disparate elements in the literature to talk to one another. This chapter thus begins by briefly defining and reviewing the history of organisation development. Attention is then focused on identifying the strengths and limitations of the theory, research, and practice in the related fields of organisational change and development, and finally a short commentary is provided on the literature on strategic change in organisations. From this review will emerge some points of synthesis which will identify not only the particular and limited forms of theorising about organisational change, but also a recognition of the essentially ahistorical, acontextual, and aprocessual character of much research on organisational change and development. From these themes will appear a statement outlining the theoretical and methodological rationale for this research on the processes of continuity and change in ICI.
ORGANISATIONAL CHANGE AND DEVELOPMENT
The immediate theoretical and empirical context in which the present research places itself is the research and writing in the related fields of organisational change and development. Organisation development (OD) has almost as many definitions as it has practitioners. Thus, Bennis (1969:2) talks of OD as “a response to change, a complex educational strategy intended to change the beliefs, attitudes, values and structure of organisations”, while French and Bell (1973:15) emphasise the long-range problem-solving and renewal process objectives that OD denotes to them. Alderfer (1977) meanwhile sees OD both as a professional field of social action and an area of academic study, and emphasises its dual value objective in improving the quality of life for members of human systems and increasing the institutional effectiveness of those systems. Beckhard's (1969:9) definition perhaps best captures the idealized scope that people in the 1960s had for OD.
Organisation development is an effort (1) planned, (2) organisationwide and (3) managed from the top, to (4) increase organisational effectiveness and health through (5) planned intervention in the organisation's processes using behavioural science knowledge.
This plethora of attempts to find an acceptable definition to use as a focal point for defining a field of practice and study has by no means met with success. Kahn (1974) in an oft-quoted and penetrating article has argued that these different definitions merely reflect a series of preferred approaches in selecting different techniques, targets, and processes for creating planned change. According to Kahn then, OD is what each man wishes to make it, and in consequence the field of OD, while possessing its own house journal, the Journal of Applied Behavioral Science, and its networks of associations, lacks a body of ideas and concepts, a theory even, to give it coherence and identity. This, like most pertinent statements, conceals as much as it reveals. Concepts and theories about intervention processes (Argyris, 1970), contingent forms of organisational diagnosis (Lawrence and Lorsch, 1969), theories of change (Alderfer, 1976), and strategies for building and managing consultant groups (Pettigrew, 1975a) have appeared, but these have been faltering and few compared with what the practitioners in the field, given their experience, are capable of writing. Instead, in the late 1960s and early 1970s, there was a torrent of idealised literature, most of it written by external consultants extolling the virtues of OD in general or their particular packaged form of it, (Blake and Mouton, 1968; Reddin, 1970; and Taylor, 1972).
Part of the problem has been the absence of a counterpoint to these espoused theories of change and intervention. Few took up Vaill's (1971) research and proceeded to ask the simple question, what do individuals in organisations who think of themselves as doing OD actually do? Weisbord (1974), who did ask this question, and McLean et al. (1982), who followed him, found not surprisingly there were large mismatches between formal descriptions of consultants’ roles as portrayed in the OD literature and their actual activities. I am led to conclude, therefore, that merely using what the OD practitioners themselves define and report of their concepts, activities, and techniques is an inadequate basis to use as a starting point for this research. The practice of OD has far outstripped research on and in OD.
ORIGINS, HISTORY, AND DEVELOPMENT OF OD
There have been a number of comprehensive and useful efforts to chronicle the origins of OD (French and Bell, 1973; Greiner, 1977); and to analyse the emergence of certain parts of its technology (Back, 1972); and its values (Tannenbaum and Davis, 1969; Solomon, 1971; Friedlander, 1976). While this is not the place to effect another broadly based survey of the history of OD, given the empirical side of this book closely mirrors a good portion of that history, it is important to examine the ICI experience in the light of the wider development of the field. In addition, exploring its history is a further window to look through in attempting to specify what OD actually is.
The term OD appears to have originated in the United States in the late 1950s possibly by Blake, Shepard and Mouton while they were co-operating on a series of OD experiments at three Esso refineries: Bayonne, Baton Rouge, and Bayway. In 1957, the year before the Esso OD work began, Douglas McGregor, who as we shall see played a brief but important part in the birth of the ICI OD work, started using laboratory training skills at Union Carbide (French and Bell, 1973). It was no accident that both the Esso and Union Carbide work used group methods, for the seeds of OD's emergence were in offshoots of group therapy methods. Laboratory training, unstructured small-group situations in which participants learn from their own interactions and the evolving group dynamics had been institutionalised as a technology by Kurt Lewin, Kenneth Benne, Leland Bradford, and Ronald Lippitt when they created the National Training Laboratories for Group Development at Bethel, Maine, in 1947.
Kurt Lewin had, of course, played an important role not only in research on group dynamics but also in stressing how theoretical and conceptual formulations could be devised from practice. This kind of thinking led to action research work first of all when Lewin founded the Center for Group Dynamics at the Massachusetts Institute of Technology in 1945, and then when the staff of the Center for Group Dynamics moved to join Michigan's Survey Research Center, and created the Institute for Social Research. This happened just after Kurt Lewin died in 1947. The kind of action research usually referred to as survey research and feedback is particularly associated with the University of Michigan at this time and later led to well-known work by Floyd Mann, Rensis Likert, and others.
As French and Bell (1973) point out, although Lewin died only 2 years after the founding of the Center for Group Dynamics and before the first formal session of NTL, he had a profound influence on both organisations and his1 influence continues today, most notably in the writing of Chris Argyris. Greiner (1977) is careful to pinpoint Carl Rogers’ work (1942) as an influence in the group therapy side of OD and also in the theorising about consultant – client relationships which evolved with the training and development of internal and external consultants in the practice of OD. Rogers’ work encouraged therapists to reject traditional psychoanalytic techniques and “to advocate methods which focused on ‘here and now’ behaviour, with special attention to encouraging expressions of affection, trust, and openness as means for healing emotional wounds.” Greiner (1977:68). It was not too long before these Rogerian theories and the value emphasis they gave to listening, expressing feelings, trust and openness began to appear in the conduct of T-groups, but this time for “normal” people who possessed the usual assortment of behaviour problems.
If Lewin and Rogers were the intellectual heirs to OD, NTL and the Center for Group Dynamics were its early institutional embodiment, and the T-group and survey research and feedback its initial vehicles, then who translated these ideas into organisational settings? Following on from Douglas McGregor in Union Carbide, and Herbert Shepard, Robert Blake and Jane Mouton at Esso, came Sheldon Davis at the TRW Systems Group in 1961 and then a whole succession of well-known academics and consultants such as Chris Argyris, Warren Bennis, Richard Beckhard, Rensis Likert, Floyd Mann, and Edgar Schein.
During this period in the late 1950s and early 1960s, OD was already taking on board its core values. Solomon (1971) likens these to a philosophy of humanism where man is seen as basically good and is at the centre of all valuational processes. From this humanist dictum that man is the measure of all things, Solomon (1971) goes on to assert that institutions should serve man, and not vice-versa, that individual dignity is supreme and that participatory democracy really only works when individuals have a voice in all decisions that affect them, when present trends toward centralisation and bureaucratisation can be countered, and where there is a place for the non-rational and emotional in the conduct of human decision-making. Alongside these factors humanistic psychology propounded a view of man which saw the self moving toward actualisation, and growthful relationships implying the willingness and ability of partners to be equally vulnerable, honest, and open to change. Solomon (1971) was open enough to note that in his experience there was slippage between ideology and reality as training programmes drawing on these values were implemented. More generally he concluded that the above characteristics of humanism can lead to a fear of control, movement towards a narcissistic autonomy, and anti-intellectualism.
In a beautifully written article Friedlander (1976) outlines another version of the underlying philosophy and perspective of OD. He argues the grandparents of OD can be thought of as three basic value stances: rationalism, pragmatism, and existentialism. The rationalistic side of OD derived from its linkages with psychology and social psychology. These provided logic, consistency, and determinism where the basic purpose was to discover truth through the precise construction of concepts and knowledge. As I indicated earlier, the academic literature on OD hardly suggests that this value has been fulfilled.
Usefulness and effectiveness, the desire to improve practice by acting – doing, were the cornerstones of pragmatism. Existentialism meanwhile provided OD with a phenomenological mode of exploration – everything now was to begin with the actor's own subjective experience in the here and now, and shared feelings and mutuality were to be the hallmarks of good communication.
The elegant part of Friedlander's (1976) use of these three value stances was the way he drew on the uncomfortable mixture which they presented for OD in its “adolescence”. Thus, for example, the pragmatist's stance of “if it works, it's good” offends the rationalist's need for precise models and frameworks, and the latter is likely to accuse the pragmatist of becoming a victim of whatever works – whether it be a technique, a fad, or a method. The existentialist meanwhile will feel suspicious that whatever works may turn out to be a manipulation of the person, and both the pragmatist and rationalist will think that the existentialist's mode of feeling and doing will run the risk of pushing experience, intuition, subjectivity, and an idealized humanistic vision beyond the realms of what most people in organisational settings will be prepared to accept.
Greiner’s (1977) model of stages of OD evolution is a useful heuristic for revealing when and why some of these value tensions in OD began to reveal themselves in practical project work. He posits five stages:
1. Orthodoxy and Advocacy – the 1950s
2. Packaged Alternatives and Choice – the early 1960s
3. Evaluation and Doubt – the late 1960s
4. Pragmatism and the Eclectic – the early 1970s
5. Reconceptualization and the New Theorists – the late 1970s.
Greiner’s (1977) dates refer to the American experience of OD, the ICI experience followed a broadly similar pattern but in a different historical time frame. In discussing the birth of OD, I have already described Greiner's stage 1. The second stage was in some sense a response to questioning of the practical value of T-groups. Blake in particular turned his back on T-groups after the Esso experience, feeling that they did not produce lasting results in the back-home organisation, relied too much on external consultancy help, and excessively stressed human relationships over task performance (French and Bell, 1972; Greiner, 1977). Blake's response was the first and probably most successful of the 1960s packages, the managerial grid (Blake and Mouton, 1964). This was a multi-phase programme focusing on managerial style which integrated people with task accomplishments and which could be run without Blake's presence. Blake and Mouton left their university posts and set up their own company, Scientific Methods Inc. They were reputed to be dollar millionaires by the end of the 1960s. Other packages by Reddin (1967) and Likert and his colleagues (1961) also emerged at this time.
The late 1960s was a period of evaluation and doubt. Managers and academics began to question the utility of many of these person-centred organisational interactions on pragmatic and moral grounds. An academic industry developed at this time critically evaluating not only the T-group’s ability to produce behaviour changes (Campbell and Dunnette, 1968; House, 1967), but also the methodological standing and conceptual base of the motivation and job enrichment ideas of Herzberg et al.2 (1959, and Herzberg 1966). Perhaps more to the point, in a survey of OD activities in 149 companies, Rush (1973) noted that the 1969–70 economic recession took its toll on several company OD efforts.
Greiner (1977) describes the early 1970s as a period of pragmatism and eclecticism. Having had their fingers burnt by some of the general packaged solutions of the 1960s, the inability in many specific situations to link, in managers’ minds, people with productivity through devices such as job enrichment and team-building, many OD consultants, internal and external, turned to the more pragmatic and less person-centred intellectual approach in structural contingency theory. There was also a feeling that “anything that works” is worth hearing about (Greiner, 1977:72). And so the early and mid-1970s saw a burgeoning of new labels, new technologies, and new settings to apply OD values and techniques. The term OD began to be superseded by the likes of human resource management, and even quality of work life. As job enrichment receded, job design and sociotechnical systems surfaced along with team-building, structural interventions, open systems redesign, and career management. The development and content of these OD technologies is well documented in Beer (1976, 1980). The evolution of the concept of sociotechnical systems, and indeed the impact of the Tavistock Institute of Human Relations as a British influence on the emergence of OD, is described in Miller (1977).
The fifth of Greiner's (1977) stages is reconceptualization and new thrusts, and this he dates to the later 1970s. Unfortunately, Greiner (1977) is not able to specify what these reconceptualizations and new thrusts are to be, and the reader is left with a feeling that searching for an identity for OD in a triple concern for “behavioural processes in organisations,” helping the longer-term capacity of people in social systems “to cope more effectively with challenges put before them” and “inculcating organisations with humanistic values” are but a return to ideology, and an ideology far out of touch with, for example, the social, economic, and organisational conditions of the United Kingdom in the late 1970s and early 1980s.
RESEARCH ON ORGANISATIONAL CHANGE AND DEVELOPMENT
Having tried to define OD, questioned the value of merely using the espoused theories of OD practitioners as expressed in their writing as a base on which to build this research, and then given some form to our discussion by describing and characterising the origins and development of OD, there remains the task of examining the strengths and weaknesses of existing research on organisational change and development. In what follows I shall spare the reader the pain of completeness for completeness’ sake. Instead my objective will be first of all to mention the styles and perspective which some of the review papers in this area demonstrate, then to discuss some of the comparative research which assesses the impact of OD interventions, and then follow this with a critical look at a few of the more solid autobiographical accounts of change which exist. Finally, an exploration of the rare studies which have elements of a longitudinal design, and contextual and process analyses of change, will reveal something of the theoretical and methodological cornerstones behind the design and conduct of this research.
Of the current review papers examining the field and practice of OD, there seem to be three main styles of presenting the literature. There is the descriptive and interpretative approach of Friedlander and Brown (1974) and Alderfer (1977), the evaluative approach of Kahn (1974), Stephenson (1975), and Strauss (1976), and the scientific – rigorous perspective adopted by White and Mitchell (1976), Porras and Berg (1978), and King, Sherwood, and Manning (1978). Not surprisingly, the descriptive approach is the most informative, the scientific the most precise and predictable, and the evaluative the most interesting to read.
Strauss (1976) stores up most of his criticism to the end of his paper, subtly drawing the reader into the invective with an opening gambit that “in the end, OD is likely to be evaluated in terms of gut reactions rather than by dispassionate research” (1976:667). He goes on to discuss OD as a fad, its anti-intellectualism, essentially conservative objectives and methods, and tendencies towards manipulation and violations of individual privacy, but the strongest language is used to decry the hard sell. “For my taste OD has more than its share of evangelical hucksterism. As an academician, I am repelled by the cloying emotionalism and unsubstantiated claims which appear in some of the literature and much of the advertising” Strauss (1976:667).
Stephenson (1975) chastises much of OD for being obsessed with the people variable to the exclusion of structural, environmental, and societal influences, for imposing an exclusive set of humanistic values on people in organisations, for misrepresenting bureaucratic functioning by concentrating only on its negative connotations, and for hypocrisy – appearing to be concerned for people yet at the same time using them as organisational instruments.
Kahn (1974) goes straight for OD's soft underbelly, that part of its literature attempting to theorise about the theory and practice of change. “A few theoretical propositions are repeated without additional data or development: a few bits of homey advice are reiterated without proof or disproof, and a few sturdy empirical generalizations are quoted with reverence but without refinement or explication” (1974:487). Kahn's (1974) eloquence disappears, however, when he moves on to make some prescriptive statements about the kind of research designs which would produce knowledge and theory quite different from what he characterizes in the above quote. In his desire to set up model field experiments to control as many variables as possible, and establish whether or not the OD intervention did produce the predicted changes, he slips into the language of “master” and “slave” groups, albeit in their figurative, mechanical sense. The mind boggles; imagining Orwellian functionaries herding abject organisational citizens into master and slave groups while Dr Strangelove stands aloof waiting to administer the experimental OD treatment. Here at last are the conditions to provide tangible evidence for Kenneth Boulding's aphorism that “knowledge is always gained at the cost of truth” (1972:112)
The Friedlander and Brown (1974) and Alderfer (1977) reviews are a good deal more balanced and comprehensive. It is beyond the scope of this chapter to synthesise the former's 175 references and the latter's 104. Alderfer (1977) is able in part to refute the comments made by Kahn (1974) 3 years earlier about the lack of sophistication of research on OD, especially with the development of the academic interest in evaluation research (Guttentag and Struening, 1975). He is also able to point to a variety of new settings where new OD technologies are being applied although he notes that it is often the relatively successful businesses, often without unions, and stable suburban school systems without urban unrest where OD is being practised. Both reviews end with substantially similar conclusions which are germane to this study; Alderfer (1977) lamenting the lack of theoretical controversies and the relatively primitive form of theorising about organisational change and development, and Friedlander and Brown pointing out the continuing “failure to produce a theory of change which emerges from the change process itself” (1974:336). Shortly we shall argue one of the prime reasons for that failure has been the paucity of research on change which actually allows the change process to reveal itself in any kind of substantially temporal or contextual manner. Research on change continues to this day to focus on change episodes, and more likely a change episode, rather than the processual dynamics of changing.
The review papers by White and Mitchell (1976), and Porras and Berg (1978) each examine published research on OD, the former covering the period 1964–74, and Porras and Berg spanning the years 1959–75. Their findings corroborate some of the statements made earlier about the history of OD. Both papers note the heavy emphasis on individuals and groups as the change target in the period up to the early 1970s, and a corresponding lack of emphasis on intergroup relationships and procedural or structural features of the organisation. Porras and Berg (1978) also report that none of the 35 studies they looked at involved work with any groups other than managerial and professional staff. Rush (1973) provides data from a much larger sample of cases also indicating that OD for most companies was an exercise for managers. In a theoretical paper, Kochan and Dyer (1976) argue the reason for the relatively rare use of OD techniques with shop-floor workers is the poor fit, value and conceptwise, between OD models of change and the normal parameters which define relations between union and management.
Probably the most extensive work carried out on assessing the impact of organisational change and development in the late 1960s and up to the mid-1970s was the research at Michigan University. These studies by Bowers (1973), Bowers et al. (1975), and Franklin (1976), are important because of the attempts at precise operationalisation of variables, the use of control groups, and the comparative analysis. These studies are also of value because of the effort made to conceptualise the determinants of success and failure in attempts to create change. Franklin's (1976) paper compares 11 organisations with successful OD efforts and 14 organisations with unsuccessful OD efforts around 8 clusters of characteristics ranging from the organisation's environment, the nature of the organisation itself, the process of setting up the project and generating commitment, to the characteristics of the internal and external change agents. His findings are a direct rebuke to those who seek a singular pattern or theory of organisational change. “Thus it appears that a strong case cannot be made for characteristics that are either absolutely necessary or in and of themselves sufficient to determine successful or unsuccessful change in organisations” (Franklin, 1976:487). However, Franklin (1976) was able to report that there were three areas which did seem to differentiate between the successful and unsuccessful organisations. Thus organisations which had already been open to and involved in adjusting to change were more likely to be successful in their OD effort than a more stable and status quo-oriented organisation. Success also appeared to be associated with the existence of an acknowledged specific problem, the ability of the consultant to find these situationally specific problems and their causes, and to aid in the process of finding the appropriate intervention, and to plan it with care.
An undoubted weakness of the Franklin (1973) work which identified the importance of climate setting as a precursor to change was their interpretation of cross-sectional data to imply cause–effect relationships. In one sense, of course, Bowers (1973) did have longitudinal data, but these data were collected exclusively by questionnaires administered in conventional snapshot fashion, before and after the change. Bowers (1973) therefore was unable to disentangle how the cause–effect relationships he was positing unfolded themselves through time. Nevertheless, the factors contributing to success and failure in organisational change identified by the Michigan researchers will be of value in informing this research at least until the present work moves beyond reflecting about the change programme as the unit of analysis.
One of the criticisms made of accounts of OD interventions is that they are often autobiographical (Kahn, 1974). The implication of this criticism is, of course, that self-reports by the consultants and action researchers of change events are bound to be coloured by the investment of their time and egos in the project work, and the written discussions that appear will be selective in the presentation of facts, particular in the asking of questions, and rosy in the interpretation of outcomes. The mere existence of Mirvis and Berg's (1977) set of cases on failures in organisation development and change, and the tone of self-critical writing in some of them is one counterpoint to that argument. Evidence substantiating the dangers of autobiographical accounts is to be found in Blackler and Brown's (1980) book, Whatever Happened to Shell's New Philosophy of Management?
The Blackler and Brown (1980) book is a product of a retrospective analysis based on a small sample of interviews of the attempt by Shell UK Refining in the 1960s to introduce a new philosophy of management in their refineries in order to increase employee commitment, efficiency, and productivity. On the basis of a very limited set of interviews, Blackler and Brown (1980) felt confident enough to reject the conclusions of Hill's (1976) autobiographical account of the Shell work, and conclude “from very early on the philosophy exercise achieved but moderate success. There is no doubt about it we believe that the exercise failed, rather spectacularly, to introduce a new philosophy of management to Shell” (1980:5).
The factors Blackler and Brown (1980) list as contributing to the failure include an initial oversell to a limited group in the company, unfavourable antecedents to the change, an overlong analytical and diagnostic phase which led to unrealistic expectations, a failure to get the board of Shell UK to officially adopt the new philosophy, the movement of key people in the change process and their replacement with “nonbelievers”; and a variety of contextual factors, including the competing values locked into the company's reward system, and unforeseen environmental changes which disrupted the emerging pattern of relationships between management and the shop floor. Blackler and Brown (1980) were also highly critical of what they called the naive intellectual base of the Shell work, the belief that profound changes could emanate from better human relations management, and the change strategy adopted by the internal and external consultants, which they caustically dubbed social science as fluoridation. Fundamental also to the change's failure were the divergent interests of the three principal parties involved in the change. These parties were the Shell internal consultancy group, the Tavistock external consultancy team, and the junior and middle managers in Shell who seemed to be the biggest sceptics of the intervention. Ultimately, Blackler and Brown (1980) contend the internal consultancy group became “a character in search of an author”, the Tavistock team were “authors in search of a character” and “the play the Tavistock authors wrote was not quite what the actors at Shell had in mind” (1980:61).
To Blackler and Brown's (1980) credit they do allow space in their book for two of the consultants involved to provide a rejoinder to their analysis, but even without these less than impartial retorts there is cause for reflection on such categorical dismissal of the Shell work given the nature and extent of the authors’ data, the sketchy analytical use they make of the data, and the overwhelming cynicism of their presentation. Blackler and Brown's (1980) use of their small number of interviews is purely to select out quotes to reveal attitudes. There is little analysis beyond that of either the nature of the developing relationship between the major actors in the change process, or of why in their own admission the Shell programme seemed to take rather more in one refinery than another. For authors, rightly in my view, emphasising the impact of contextual variables on change processes, the differential impact in the Stanlow and Shell Haven refineries offered them a marvellous opportunity to look again at their data and begin the process of developing theoretical ideas which would inform the question, why? Equally well, the factors they usefully developed to help explain the apparent overall lack of success of the Shell work could have been related to other published material such as Walton's (1975. 1978) research on the diffusion of innovations, and the Mirvis and Berg (1977) cases of failures in OD. While Blackler and Brown (1980) have undoubtedly added to the interpretation of the Shell work initially provided by Hill (1976), their criticisms did not dissuade me from examining two other autobiographical accounts of organisational change efforts in the UK published by Klein (1976), and Warmington, Lupton and Gribbin (1977).
The Klein (1976) book is a narrative account of the growth, stabilisation, and demise of Lisl Klein's role as a social sciences adviser to Esso between 1963 and 1970. Klein (1976) admits the book is weakened by its autobiographical origins. Like the Hill (1976) and Blackler and Brown (1980) books, it is also limited by two important requirements of case study work. First, the importance of conceptualisation to place the empirical material in a broader frame of discourse, and secondly the necessity to relate the case studies to similar published materials. Klein's (1976) book offers very few links with other published cases of change, and where attempts are made in part two of the book to develop the more general themes of the politics of social science, and the dynamics of industry and social science relationship, these generalising frames of reference are inadequately connected back to her own case study material.
The book at an implicit level, however, does have a number of important things to say which can aid the present task of drawing out some of the general strengths and weaknesses of the existing research on organisational change and development. In the first place Klein's (1976) narrative account does, through its personal and political realism, provide a counterweight to balance against the normative espoused theory of the OD literature. By stressing the vulnerability of the internal consulting role, its essentially political character, and the continuing problems of territory and competition she had with other related specialist functions such as personnel, and with external consultants, she contributes to those authors (Pettigrew, 1975b) who have questioned the truth, trust, love and collaboration approach to change emphasised by Bennis (1969:77).
The factors Klein (1976) draws out to explain the demise of her role show many similarities with the list prepared in Blackler and Brown's (1980) analysis of the Shell work. These include a permissive rather than enthusiastic senior management, her isolation from the centres of power in the organisation which partly stemmed from her own entrapment inside an employee relations department at the middle of the organisation, and the frequent management development moves of key personnel which created a general and continuing problem between rapid career progression and organisational learning and change. The most significant general message of Klein's (1976) book, and the one that informs the themes and analytical perspective of this research, is the importance she attaches to environmental change as a contributor in the success or otherwise of specialist-led attempts to create change. Although she admits it was a conclusion born of hindsight, she asserts “social science in application involves a continuous interplay between the content of the work being done and its context. Success depends on successfully regulating the relations between the two” (Klein, 1976:9). As we shall see, the specification of the differing levels and forms of contextual impact, and the process of regulating that impact on major changes, will be one of the crucial themes of this research on organisational change in ICI.
The final autobiographical account of organisational change which is of relevance is the work by Warmington, Lupton and Gribbin (1977) in a large UK manufacturing organisation. This study limits itself at this stage to laying out the analytical assumptions and change strategy adopted by the authors and a group of internal consultants to create “open sociotechnical systems change”. Regrettably the authors are not able in this publication to include a detailed description and analysis of the actual process of implementing their change programme, in this sense the more important publication from this research is yet to appear. In the theoretical prelude to the cases which they do feel able to describe, and in their concluding practical assessment, they also mention a number of contextual variables which they consider were fateful to explaining the only limited success of the change venture they were part of. Again there are similarities with the accounts of change portrayed by Hill (1976), Blackler and Brown (1980), and Klein (1976).
A central problem of the Warmington et al. (1977) interventions derived from the exclusive language and problem-solving rationale built into their open sociotechnical systems approach, and in the other arm of their change strategy which emphasised a cell division concept of diffusing change. They had assumed that a combination of an intellectually powerful mode of problem-solving, plus a strategy of involving participants closely in the diagnosis and initiation of change in a number of separate but relatable cells or projects would, if the process were successful, lead local managers to press for wider changes. What Warmington et al. (1977) found was that “the process of implementing departmental improvements using rather unorthodox theories and methods is risky for local managers; its outcome is problematic and there are cases in which the projected improvements never eventualise” (1977:239). When this happened the momentum for the change receded and no amount of logical reasoning or pushing from the combination of internal and external consultants could revive interest either in the mechanism for achieving change, or the wider process of change itself.
Again these authors conclude with a highly contextualist argument which emphasises building into the change process a developed awareness by the consultants and skill in intervening in the political and cultural systems of the organisation in order to build up a nucleus of political support for the changes. In this regard, however, they pinpoint a basic lacuna in the processes of designing a large change programme. “We are in a sense in a vicious circle in that if it is to be effective the design of a change programme can only take place when the culture and power system and modes of behaviour of the wider organisation are fully understood; but these features cannot be properly understood until some processes of change have been introduced and the reactions observed” (Warmington et al., 1977:240).
In this author's view there are at least two broad ways of breaking out of the logic of the above vicious circle and many tactical alternatives within and around those two broad ways. The Warmington et al. (1977) change lacuna is predicated on the assumption that change has to be led by a group of change agents with a vested interest in change who somehow represent a better way of organising than those groups who currently are in control of the organisation and define and articulate its culture. In looking at the ICI experience we will be able to examine the alternative strategies of change where the leading edge for change was provided by the chief executives and senior managers of the company. This strategy does, of course, like any other action create its own set of implications and consequences, but not those bounded by the Warmington et al. (1977) change conundrum. The other route out of the change conundrum of, you have to act before you can appreciate how to act competently, is to develop more sensitive ways of conceptualising and theorising about change which recognise and help to define the immediate and broader context in which change occurs, and how that context may be mobilised by advocates of change in order to achieve their objectives. In my view the development of such knowledge and theorising about change which recognises both contextual variability, and a continuing process of dipping into the context in order to assess where, when, and how change is possible, requires a quite different strategy for research on change than the approaches dominant today. In this final section reviewing relevant parts of the literature on organisational change and development and strategic change I shall build on other research on change with a longitudinal and contextual character in order to establish some of the theoretical and methodological ground rules for this research.
CONTEXT AND PROCESS IN EXISTING RESEARCH ON ORGANISATIONAL CHANGE
With the exception of Klein (1976) and Warmington et al. (1977) most of the studies of change which have been mentioned so far in this review have been ahistorical, acontextual, and aprocessual. This is the tradition in the field and even Klein (1976) and Warmington et al. (1977) while contextual in part do not draw on historical materials or provide a processual analysis of change. As with so many other areas in the social sciences the empirical findings and theoretical developments in the field of organisational change are method-bound. For as long as we continue to conduct research on change which is ahistorical, acontextual, and aprocessual, which continues to treat the change programme as the unit of analysis and regard change as an episode divorced from the immediate and more distant context in which it is embedded, then we will continue to develop inadequate descriptive theories of change which are ill-composed guides for action. Indeed as I have implied already there is still a dearth of studies which can make statements about the how and why of change, about the processual dynamics of change, in short which go beyond the analysis of change and begin to theorise about changing.
These analytical issues, and the way they are represented in the field, are not black and white. There are alternative ways of defining longitudinal research each of which has different implications for whether a processual analysis actually appears from the longitudinal data. An examination of the Bowers (1973) work illustrated how time series data, if collected in a snapshot fashion, would not provide a processual analysis, and how even though Franklin (1976) had data which were amenable to processual analysis he chose to use those data to abstract out cross-sectional categories and factors. Equally well single case study longitudinal work such as Alderfer and Brown (1975) which explicitly uses the term changing as a focal point for the study is actually limited by not being able to relate variability in context to alternative processes of changing. And the important paper by Greiner (1967), which proclaimed the importance of historical analysis and the significance of antecedent conditions in explaining the limitations and possibilities of change, was itself hampered by not being able to relate alternative antecedent conditions to common or variable contexts, and thence to differing resultant change outcomes.
A further area of research which has attempted to provide contextualist propositions about change is the literature on innovation (Zaltman, Duncan and Holbeck, 1973; Kimberly, 1981). This literature, either by defining innovation as a discrete product or programme (Rogers and Shoemaker, 1971), or as a process (Knight, 1967; Shepard, 1967), has attempted to disentangle the variables particular to individuals, to organisational structures or broader organisational contexts which influence rates of initiation, adoption, diffusion and in some more limited cases, implementation of innovations. As Zaltman et al (1973:58) point out most diffusion theorists terminate their analysis at the stage of initiation, where an idea has been accepted by the authority system of an organisation and ignore the rather more intractable implementation stage. Zaltman et al. (1973) argue for linking the initiation and implementation phases together in a sequence given impetus in the first place by senior managers’ perceptions of a performance gap between what the organisation is doing and ought to be doing.
The structural contingency theorists Burns and Stalker (1961), Lawrence and Lorsch (1967), and Duncan (1973) all posit relationships between structural context and innovation potential. Zaltman et al. (1973:154) in a useful summary of this literature argue that different configurations of organisational structure are appropriate at the initiation phase of innovation than in the implementation phase. Given the high functional requirement for gathering and processing information at the initiation stage then there is a need for higher degree of complexity of structure, lower formalisation, and lower centralisation. At the implementation phase, however, a higher level of formalisation and centralisation, and a lower level of complexity are required in order to reduce the role conflict and ambiguity which could impair implementations.
There are a number of cardinal and basic faults with this literature on innovation which makes it a poor base to construct realistic theories of change. A central difficulty the innovation literature shares with the writing on planned organisational change is the highly rational and linear theories of process which drive these models. Thus from the planned change theorists we have phases of consultant activity which range from develop need, establish relationship, work toward change, stabilise change and evaluate (Lippitt et al, 1958); to diagnosis, strategy plan, educate, consult and train, to evaluate (Beckhard, 1969). And from the innovation theorists, this time not considering the activities of change agents but a characterisation of the innovation process as evaluation, initiation, implementation and routinisation (Hage and Aiken, 1970); or from Zaltman et al. (1973), the more complicated but equally rational and linear; knowledge–awareness substage, formation of attitudes toward the innovation substage, decision substage, initial implementation substage and continued sustained implementation substage. Zaltman et al. (1973:53)) even go on to say that “in the process approach, innovation is viewed as an unfolding process consisting of stages in which characteristic factors not only appear in greater or smaller degree, but also3 in a certain order of occurrence”. And even Warmington et al. (1977) who elsewhere in their book imply they see change in much more sophisticated processual terms than these argue “the formulation of a strategy for change can be seen as a specific stage in a sequence that includes investigation, analysis of facts, model-building, proposal formulation and implementation of improvements” (1977:181).
In this research on change in ICI it will be clear that this rational problem-solving approach to planned change and innovation is both an inadequate way of theorising about what actually happens during change processes and an overtly simple guide for action. The field of organisational change badly needs theoretical development along the lines of the literature on organisational decision-making where there are now a variety of process models of choice which include satisficing views of process (March and Simon, 1958); political views of process (Pettigrew, 1973); and garbage can views of process (March and Olsen, 1976).
Moving away for an instant from the process difficulties of the planned change and innovation literature back to other studies of change which draw on contextual variables one can see the promise in this approach. Other writers apart from the structural contingency theorists who have used contextualist explanations of change include Mohrman et al. (1977), Lewicki :1977), Elden (1978), Kervasdoue and Kimberly (1979) and Kimberly (1980). Of these authors, the first two define context either exclusively in terms of intraorganisation context or in terms of a combination of that and some notion of organisation environment, while the latter group with varying degrees of specificity draw on intraorganisational, organisation-environmental, and socio-economic contextual variables.
Mohrman et al. (1977) report a survey feedback and problem-solving intervention in a school district and ask the question why was the programme almost completely implemented in School B and hardly implemented at all in a high school? Their answer to that question equates very closely with Klein's (1976) observation that success in stimulating change is very much a function of the fit or lack of fit between the content of an intervention and its context. School B was, relative to the high school, small, fairly tightly integrated and with a culture which already supported the kind of group problem-solving activity the change programme was seeking to encourage. It also had a principal who provided both strong general and specific support for the programme and who required attendance at group meetings. The high school on the other hand was much larger, structurally much more differentiated, and had a culture with little history and capacity for working in groups on problem identification and solution generation processes.
Lewicki’s (1977) study was an attempt to use team-building techniques among small businessmen in two New England communities. Although similar strategies were used by the consultants in both communities, the intervention in the author's view was a success in “Marysville” but a failure in “Riverside”. Lewicki (1977) accounts for the different outcome in specifically contextual terms, “following the research by Shapero (1975) and others, we learned that a community or region like Riverside may not have the environmental qualities that lend themselves to stimulating community development and eventually economic growth” (1977:309). His more detailed explanation for success and failure, however, was a mixture of context and strategy. Thus Riverside was dominated by a single large employer whereas Marysville had a number of units in the business community of reasonably equal power. Riverside's existing sense of geographical identity, interdependence and mutual fate was probably lower than in Marysville, and the businessman who led the change attempt at Riverside was a marginal man in that community, who was too directive in his approach and too poorly tied to the central power structure of the business community to get the correct kinds of changes off the ground in the first place, or to sustain what had managed to start when the economic conditions altered and further threatened the change programme. Meanwhile the businessman who led the change effort in Marysville was well known and liked; had been explicitly selected by the community, had a business that stood to gain from the envisaged changes yet set in motion a process which involved high levels of participation and commitment in choosing salient, initially small-scale, and highly tangible change objectives.
This attempt to link circumstances and behaviour to account for the success and failure of change efforts is also seen in the work by Kimberly and Nielsen (1975), and Kimberly (1980). In the 1975 study Kimberly and Nielsen are able to demonstrate with longitudinal data that crucial to the impact of an OD effort on organisational performance was not only factors particular to the change programme itself, but also more significantly the corporate policies and market conditions of the firm. Kimberly's (1980) study of the birth of a new medical school, his work on French hospitals with Kervasdoue, Kervasdoue and Kimberly (1979), and Elden's writing (1978) all point to an aspect of context virtually ignored in the literature on change and development, the conditioning and enabling influence of the social and economic environment surrounding the organisation.
The Kervasdoue and Kimberly (1979) research is interesting because the authors started by posing their principal research question within the frame of reference of structural contingency theory. They were interested in examining the extent to which variability in rates of adoption of innovations in medical technology in US and French hospitals could be accounted for by variations in their structure. What they concluded was that in order to understand hospital innovation, it is necessary to go beyond the comparative structuralist paradigm and ask questions about sociopolitical, historical, and cultural factors in and around organisations. The reports of the Norwegian Industrial Democracy Programme by Qvale (1976) and Elden (1978) go a stage beyond the Kervasdoue and Kimberly (1979) observations in making normative statements about how the particular political, practical, and professional conditions which have evolved in Norway over the past 15 to 20 years were crucial preconditions for the evolution of the Industrial Democracy Programme. These conceptual and practical acknowledgements of the role that social, economic, political, and historical factors can play in facilitating and constraining change at the organisational level of analysis take us to a brief review of the literature on the formulation and implementation of strategic change, the final body of thinking and writing connectable to the theme of this research.
ELEMENTS OF THE LITERATURE ON STRATEGIC CHOICE AND CHANGE
One of the analytical conclusions of this study is that theoretically sound and practically useful research on strategic change should involve the continuous interplay between ideas about the context of change, the process of change, and the content of change, together with skill in regulating the relations between the three. Formulating the content of a strategic change crucially entails managing its context and process. Without sensitivity to and apposite action on the what, the why, and the how of introducing major change more than likely the change idea will either die shortly after birth or be emasculated at later stages in the processes of formulation and implementation.
But what has the existing literature on strategic change to say about either the content, the context, or the process of managing change? In fact the general literature on business strategy development almost exclusively concerns itself with the content of strategy, while treating context in terms of the crucial but still limiting notion of the competitive environment of the firm (Andrews, 1971; King and Cleland, 1978; Porter, 1980). Even more problematic, however, is the implicit view of problem-solving built into this strategy literature which is as rational as, if not more rational than, the view of choice and change found in parts of the writing on innovation and planned organisational change discussed in the previous section of this chapter. Thus as applied to the development of business strategy the rational approach describes and prescribes techniques for identifying current strategy, analysing environments, resources and gaps, revealing and assessing strategic alternatives, and choosing and implementing carefully analysed and well-thought-through outcomes. This rational picture of business problem-solving has as its concern the content or what of strategy – the outcome which is sought – and has nothing to say at an explicit level of how to achieve that outcome. In other words it has no process theory within it of how and why to create the strategic outcomes so perceptively and logically derived from the analysis of competitive forces.
But what are the potential sources available to try and develop an adequate theory of strategic change which incorporates thinking about the process of managing change? The organisation development literature has rarely been used to inform thinking about strategic change, even though recent writing by, for example, Beckhard and Harris (1977) and Beer (1980) could be used with profit to grapple with some of the practical problems of creating and managing strategic change. Perhaps another route to stimulate process theorising about strategic change is to draw on the work of authors such as Bower (1970), Allison (1971), Pettigrew (1973), Mumford and Pettigrew (1975), March and Olsen (1976), Mintzberg (1978), and Quinn (1980), who at least at the level of observation and description, if not always prescriptively, have sought to rescue research on strategic choice and change from its habitual focus on rational analytical schemas of intentional process and outcome, and to see decision-making and change in a variety of process modes.
We have Simon (1957) and March and Simon (1958) with their concepts of bounded rationality and satisficing to thank for highlighting how the cognitive, learning and search limitations of individuals can curtail maximising models of rational choice. However, March and Simon's (1958) tendency to project individual processes of choice into statements about organisational processes of decision-making means there is a liberal bias to reconstructing the organisation from the perspective of the individual, and not enough on demonstrating how the organisation structures the perspective and interests of the individual. No such theoretical fallacy is evident in Cyert and March (1963). They also conceive of a firm as a goal-directed, economising and learning individual but extend this view by conceiving of decision-making as a political process. In Cyert and March (1963) conflicts of interest are normal features of organisational life, coalitions form, and sub-groups work to generate support for their interests and demands. But although some of the language of political analysis is present in Cyert and March, and there is the prospect that power may cause decision outcomes, the theoretical apparatus to help describe and analyse how political processes cause choices to be made is absent.4 As Pfeffer (1982:7) has recently argued, implicit in the bounded rationality approach was the normative position that the bounds of bounded rationality could actually be pushed back if new decision technologies or information processing systems could be developed. In this way technology could compensate for the limited cognitive and information processing abilities of man, and more rational action would result.
Another group of scholars also interested in descriptive and prescriptive approaches to choice were the policy analysts (Lindblom, 1959; Braybrooke and Lindblom, 1963). In the complex and unpredictable world of macro-policy-making, how again could boundedly rational decisions be made under time constraints and where search costs were limited? The answer was through methods of successive limited comparisons in which policy-makers would move incrementally from the base line set by practice and precedent. Here again, presumably under the long shadow of the public interest, policy analysts are implicitly trying to be as rational as they can given the difficult contexts in which they operated, and the complex problems they had to solve.
Eventually the process theorists started to take an interest in strategy formulation in business settings where the dominant intellectual paradigm was a combination of a preoccupation with the analytical content of strategy with limited views of context, and an avowedly rational picture of business problem-solving (Ansoff, 1965; Andrews, 1971). Without necessarily over-identifying themselves with any of the particular process theories then emerging, writers such as Bower (1970) and Mintzberg et al. (1976, 1978) began to treat strategy not as an output but as a process, and confront the dominant rational/analytical paradigm with the results of their descriptive studies. Bower (1970), like Aharoni (1966), conceived of strategy developing at multiple levels in the organisation through processes characterised as chains of commitment leading to eventual confirmation. Phase metaphors describing the life history of investment proposals from definition to impetus were postulated, and aspects of the impetus phase were described in political terms. Top managers were seen to determine the structural context through which capital proposals would pass and be filtered. Mintzberg (1978), using extended time series data collected retrospectively, rather than through elements of historical and real time data, conceived of strategy as consistency over time in a stream of decisions. This approach allowed him to distinguish between intended and realised strategy, and to pinpoint strategies contributed after the fact, or as he puts it realised despite intentions, (Mintzberg, 1978:934). This was an important if fairly obvious observation, for it drove a stake right through the prevailing orthodoxy that strategies are and should be plans conceived in advance of making specific decisions.
Another of Mintzberg's contributions was to postulate both that strategies appear to have life cycles – periods of incubation, development, and decay; and that distinct periods of change and continuity can be discerned in the overall pattern of strategic development of the firm. This notion that firms may have patterned and periodic shifts in their strategy (Mintzberg and Waters, 1982) has been extended by comparative case study research conducted by Miller and Friesen (1980) using the metaphors of evolution and revolution, and by recent historical work on firms in the pre-nationalised British Steel Industry. In this work on steel firms Boswell (1983) identifies what he calls phases of strategic concentration around values of for a time growth, and then efficiency. This phenomenon of the deep organisational cultural roots of business strategies, and the tendency of strategic changes to occur in packages interspersed with periods of incremental adjustment, will be an issue we shall return to when examining strategic continuity and change in ICI.
In summary, this empirical process research on strategy made a number of descriptive contributions to the understanding of strategic decision-making and change. Strategic processes were now accepted as multi-level activities and not just the province of a few, or even a single general manager. Outcomes of decisions were not just a product of rational or boundedly rational debates, but were also shaped by the interests and commitments of individuals and groups, the forces of bureaucratic momentum, gross changes in the environment, and the manipulation of the structural context around decisions. With the view that strategy development was a continuous process, strategies could now be thought of as reconstructions after the fact, rather than just rationally intended plans. The linear view of process explicit in strategy formulation to strategy implementation was questioned, and with increasing interest in enduring characteristics of structural and strategic context (Bower, 1970; Burgelman, 1983), Chandler's (1962) dictum that structure followed strategy was modified by evidence indicating why and how strategy followed structure (Galbraith and Nathanson, 1978).
However, analytically the Bower and Mintzberg research was weakened by a number of factors. Chief amongst these was the treatment of decision-making just as front stage behaviour, and the lack of concern with non-decision-making, and secondly the lack of development of a specified process theory to explain the descriptions of process and outcome. Quinn's (1980) publication, however, tied itself to a very specific interpretation of process. Drawing on the work of Lindblom and Braybrook, Quinn interpreted his case study data on strategic change as displaying patterns of logical incrementalism. This is described as a jointly analytical and political process in which executives are described as, and are recommended “to proceed flexibly and experimentally from broad concepts to specific commitments, making the latter concrete as late as possible” (Quinn, 1980:56). Strategic change is thus seen as a cautious, step-by-step evolutionary process, where executives muddle through with a purpose.
Quinn’s (1980) style of presenting his ideas, moving easily from description to prescription, makes his book extremely attractive as a teaching medium, but the clarity of his belief in the prescriptive value of logical incrementalism means it is not always easy to disentangle what he has discovered empirically from what he would like to see. Certainly Mintzberg's (1978) findings that incremental change takes place in spurts, each followed by a period of continuity, tends to contradict Quinn's (1980) finding or view that strategies emerge in a continuous incremental and thereby additive fashion. Perhaps because Quinn's (1980) work is presented in the language of change rather than decision-making, and he is interested in prescription, he does make the terribly important point that in the management of strategic changes there are process limits to consider as well as just cognitive limits. These process limits, the concern with the timing and sequencing of action and events in order to build awareness, comfort levels, and consensus for strategic change, descriptively we will see was also a crucial part of the strategic change process in ICI.
Although the characterisation and description of process in Quinn's (1980) work is a good deal richer than anything offered by the bounded rationalists, the underlying rationalist perspective that there is an element of “conscious, foresightful action reasonably autonomously constructed to achieve some goal or value” (Pfeffer, 1982:7) is shared by Quinn and the bounded rationalists. No such rationalistic perspective is at the bottom of another process theory developed in the 1970s, the so-called garbage can view of action and choice (Cohen, March, and Olsen, 1972). In this view of organisation, individual and group interests are only partially understood and acted upon, actors walk in and out of decision processes, solutions are generated without reference to problems, and outcomes are not a direct consequence of process. In this anarchic picture of organisational life behaviour can be predicted neither by intention nor environmental constraint, instead decisions appear out of foggy emergent contexts when people, problems, and solutions find themselves sharing the same bed. Fundamentally, “rationality cannot guide action in this view, because rationality, goals, and preferences are viewed as emerging from action rather than guiding action” (Pfeffer, 1982:9).
This is a very attractive view of process precisely because it is such a counterpoint to the rationalists, but one wonders if the view of anarchy has not been overstated – there is greater consistency and continuity of action, clearer beliefs in self and group interest and action derived from those interests, and generally firmer cause–effect attributions in connecting perceptions, information, interests and preferences to action and outcomes than is implied by the garbage can theorists. In addition most of the empirical examples used to support this garbage can approach to choice, are taken from educational organisations (March and Olsen, 1976), where the theory may fit a context widely assumed to contain hapless citizens confused and conflicted about means and ends, confronted by multiple systems and centres of leadership, and unclear about their responsibilities and accountabilities.
The above necessarily synoptic account of the rational, boundedly rational, incremental, and garbage can approaches to choice and sometimes change, will have given a flavour both of the variety of perspectives now available to the student of strategy, some of the strengths and weaknesses and overlaps of each, and a few of the empirical findings that those alternative views of process have encouraged. A task that remains is to sketch out the meaning and utility of a process view of choice and change which combines a political and cultural analysis of organisational life. This task is part of the agenda for Chapter 2.
BEARINGS
The above review of the literature on organisational change and development and strategic change has found that literature wanting in a number of crucial respects. Principal amongst these has been the extent to which the theories and empirical findings in this area have been circumscribed by the limited frames of reference and methodologies and approach used to study change. In the main the research which has been reported has been ahistorical, aprocessual, and acontextual. In this respect the area of organisational change and development is merely reflecting the biases inherent in the social sciences generally, and in the study of organisations in particular. But in other respects the field of organisational change and development has emerged as a rather precious subculture of theory and practice which has not connected itself well either with existing writing or novel theoretical developments going on amongst sociologists and anthropologists interested in social change, (Zald and McCarthy, 1979; Geertz, 1973), or to other advances taking place in organisation theory and behaviour. A consequence is that the field of organisational change and development is characterised by limited attempts at theoretical development, few theoretical debates, highly focused kinds of conceptualisation, and very limited empirical findings. All this adds up to some rather poor descriptive theories of change which beyond a shopping list of prescriptive do's and don’ts, sometimes qualified by contextual riders, could hardly be described as adequate guides to informed action.
A particular limiting problem identified in the literature was the tendency to regard the change project as the unit of analysis, and change itself either as a single event or a set of discrete episodes somehow or other separate from the immediate and more distant context which gave those events form, meaning, and substance. The impression is created in this view of change that each change has a particular beginning and a finite ending apart from the more generalised processes around it. Regrettably this perspective still abounds today both in recent literature examining quality of work life changes (Goodman, 1979), and in that part of the OD literature which condenses organisational change processes down to the minutiae of relationships between consultant and client (Mangham, 1979; McLean et al., 1982). The literature on consultant–client relationships will be discussed in more detail at a later stage of this book as empirical material is being presented about the attitudes and behaviour of the ICI internal and external consultants. The consultancy literature is, however, yet another example of the prescriptive, rather idealised character of the change field (Blake and Mouton, 1976; Steele, 1975), and of a subset of that field focusing in on the change process from a highly limited frame of reference. Even fairly circumscribed attempts to demonstrate how consultant– consultant relationships might influence consultant-client relationships, (Pettigrew, 1975b), and how the strategic management of consultant groups influence change processes, (Pettigrew, 1975a), bring in levels of analysis rarely alluded to in the myopic concern with how consultant relates to client.
Aside from the above problems of limited focus, the above review has also pinpointed the inadequacies of the highly rational and linear process models which drive most planned theories of change and the literature on business strategy development. These planned theories, many of them with highly prescriptive and deterministic phases or stages (Lippitt et al., 1958; French and Bell, 1973), are, as we shall see, divorced from the actual conduct of change projects and processes. For the task of examining how strategic change actually takes place and some of the dynamics behind those change processes we will have to draw upon some alternative models of process (Pettigrew, 1973, 1979; Quinn, 1980) which are able to explain how the possibilities and limitations of change in any organisation are influenced by the history of attitudes and relationships between interest groups in and outside the firm, and by the mobilisation of support for a change within the power structure at any point in time. Such a view of changing does not carry along with it either the ideological baggage of progress identified by Kimberly (1981) in the literature on managerial innovation, or the necessity to see change just as a response of management to improve efficiency, but rather is capable of interpreting change as the legacy of struggles for power emerging through time.
Mindful, however, of my earlier admonition to beware of singular theories of changing, no attempt will be made here to simply substitute planned, rational linear theories with political process theories. The task is to identify the variety and mixture of causes of change, to examine the juxtaposition of the rational and the political, the quest for efficiency and power, the role of exceptional men and of extreme circumstances, the untidiness of chance, forces in the environment, and to explore some of the conditions in which mixtures of these occur.
This leads me to pick up the two final areas where particularly the OD and organisational change literature were found wanting, their use of contextual and processual forms of analysis and explanation. In this author's view the study of organisational change is now at the stage where theory and knowledge is required principally to understand the dynamics of changing in alternative contexts using a framework of analysis which can incorporate different levels of analysis with varying degrees of explanatory immediacy and distance from the change process under examination. In order to do this the field has to move beyond the useful but mechanical statements of contingency theory which emphasise the interconnections between a state of the environment and certain requirements for structure, behaviour or change, and begin to examine how and why changes occur in different organisational cultures and political systems, under different socioeconomic and business conditions, through time.
Such an analysis could properly explore the relationship and interplay between the content of change, the context of change, and the process of managing change. It would require frames of reference and methods of data collection sensitive to alternative antecedent conditions, variety of receiving culture for the change, alternative levels of analysis and explanation, differing change strategies, and alternative change outcomes. Above all, it will require time series, processual data in order to see how and why the above broad analytical factors work themselves through any particular sequence of events and actions.
In the chapter which follows a frame of reference and research design incorporating the above historical, contextual, and processual building blocks will be presented together with the detailed study questions for this research on organisational change and development in ICI.
1 The most detailed account of Kurt Lewin's life and influence is Marrow's book (1969), Practical Theorist.
2 For criticisms of the Herzberg approach, see House and Wigdor (1967).
3 My emphasis.
4 See Pettigrew (1973) for an elaboration of these reflections on the Carnegie decision-making scholars of the 1960s.