Copyright MCB UP Limited (MCB) 2000June Tolsby: Department of Business Studies, Aalborg University, Denmark
Introduction
Living in a time where information technology (IT) seems to be a natural, integrated part of our work and daily lives, less and less critique seems to be framed concerning how this technology changes how we do our jobs and live our lives. But, how do we explain and identify the impact of this technology upon our jobs, and do we acknowledge that an understanding of these changes is important in making ourselves masters of this technology and not its slaves? A useful approach to find answers to these questions might be to study how the implementation of this technology affected employees. Orlikowski (1992) has pointed out that a valuable approach to an understanding of how IT changes behaviour is to study first-attempt implementations. In this paper, the empirical data analysed were taken from a study of a first-attempt implementation of an IT system in two administrative army units in Norway. The remainder of the basis for this paper lies on what we could call "insignificant statements". These statements are analysed in order to illustrate how "little things" can offer new insights and reveal views that otherwise would be left unnoticed. A good example of how insignificant statements provide interesting information can be found in the works of Wrege et al. (1997). They investigated some of Taylor's insignificant statements, and provided a completely new view on Taylor's work. This paper tries to do more of the same, and seeks new insights by looking at some "little things" or, in the words of Wrege et al. (1997, p. 59):"... how little things can offer the possibilities of new information on almost any facet of management history."
It has not been the goal of this paper to report from the vast amount of literature written about Taylorism, but rather to focus on more recent developments within the literature dealing with this topic, to find comparisons with respect to how the concepts of scientific management stand today, and to see if the signs found are applicable to the data presented in this paper. Within this scheme, a discussion about concepts was found worthwhile. Can any signs of Taylorism found in the data used be interpreted as true signs of Taylorism or should they be classified as neo-Taylorism, in which case, how do we differentiate between neo-Taylorism and Taylorism?
The story
The Norwegian Army consisted in 1995 of three hierarchical levels:
(1) the central level with the defence command and the army material command;
(2) the regional level with the five land commands; and
(3) the local level with its approximately 130 units.
This organisation was constructed in this way over the years to accommodate, and act consistently and according to, orders. This made the army a strict hierarchical organisation with great emphasis on structure and formal positions, and where obedience rests on the need for firm and unified action when the army is at war. Obedience has traditionally been a virtue in the organisation and an absolute requirement, a factor that features prominently in an officer's military record. For officers, the military record is part of their curriculum vitae, together with their formal education, and qualifies them for jobs in the army. The importance of the military record becomes apparent when it has "a blemish", resulting perhaps from reluctance to obey orders - this could disqualify any future military career (Tolsby, 1996).
In 1995 I studied how this organisation implemented a major administrative IT system. The IT system was going to be implemented at the 130 local units of the army, starting first with a few pilot units. It consisted of six different modules all capable of accessing the same database, combined with an office automation module. Figure 1 illustrates how the system was going to look like.
When I first reported from my research I found an organisation where the introduction of IT changed the familiar ways of working within the organisation, but where the dominant structures of the organisation continued to be upheld, despite the apparently egalitarian nature of the PCs and more information being available to more operators, no evidence was found that this promoted organisational democracy. The original research report also claimed that the technology adopted was problematic in terms of the custom-designed software, and that the way this project was organised reduced the possibility of employees influencing to a minimum, leading to IT representing a new form of Taylorism (Tolsby, 1996).
I return in this paper to some of the small, insignificant statements expressing employees' views on the consequences of the implementation of this IT system in an attempt to see if the implementation of the system imposed a scientific management scheme upon the employees, or some new form of Taylorism.
Small insignificant statements
The implementation of this IT system was advocated through a project organisation which was organised in the traditional way - with a steering group, a reference group and a project group. The project group was responsible for the local introduction and implementation of the IT system.
To ensure local acceptance throughout the implementation process the steering group enrolled trade union representatives. Their perceptions of this project were based on reports from members. Users of the IT system reacted positively with respect to how information could be retrieved much quicker and easier, and at the same time feared that the IT system would replace them and make them redundant.
The trade union representatives in the steering group felt they had little influence upon the implementation process. They felt like a "rubber stamp", without any real influence on the decisions taken in the steering group. In fact, one of the trade union representatives felt like a "hostage" in the steering group. Real influence was felt to be virtually non-existent since they were outnumbered in the steering group - they were only two members out of a group of 20. Since discussions in the steering group occurred at a general level, the union representatives felt unable to take care of the user's interests, which they felt were related more to the direct implications of the new IT system:"We really did not have any influence, we were there more as hostages. It would have been better if we had been able to participate in working groups. Then we would have felt that we could influence the decisions that were taken. I would have preferred to look at results and then say: This is not good enough and we will not put up with it. When the general agreement between the trade unions and employers was made there were many trade unions that were afraid of not being heard, so they wanted to be members of steering groups, project groups and work groups. Then one became trapped by one's ability to influence decisions. Why be a member of these foras when one is not being listened to? I would assume that a computer system that is wrong for the user would also be wrong for the employer (union representative and member of the steering group in this project (Tolsby, 1996, p. 64))."
Despite feeling unable to take care of their members' interests in this project, the union representatives seemed to accept the course of action taken by the project:"The EDBLF project started growing as all computer projects do. One started off to solve a particular problem but thousand others arouse for the project to try and solve. A result of the general gigantomania of the Army in relation to this project, became too much and difficult to handle. It reached a stage where there was too much money and prestige involved. It reached the point of no return there could be no other solution but to continue. The more funds and the longer the time spent on this project the more difficult it became to scrap the whole project (union representative and member of the steering group in this project (Tolsby, 1996, p. 68))."
According to the union representatives, they faced a situation where the project was developing in the wrong direction, owing to their perceived lack of real influence and the feeling that they were unable to do a good enough job as members' (users') representatives.
In this project the implementation process was to be carried out using a set of qualified specialists educated in using the IT system. These experts would then work as agents within the different local units, and help in educating local expert users, and thus creating a "cascade" of knowledgeable persons within each local unit. Obviously, this would depend upon a working IT system with a tolerable amount of faults, and a personnel willing to adopt the system. Thus the success of such an operation would depend upon employees' motivation and enthusiasm, an enthusiasm that could be fuelled by a well-designed user interface. The army decided on two things that decreased this enthusiasm:
(1) they chose to adopt an off-the-shelf application to implement as part of their office automation package; and
(2) they decided upon custom-designed software for the other modules within the IT system.
Consequently, the army found itself in a squeeze between old technology, based on a command-oriented interface, and the new technology, based on direct manipulation of icons. In particular, the introduction of the Windows user interface changed users' perceptions of what PCs could do and "look like". Thus the users in the army units got caught between a modern and an old user interface - both part of the complete IT system they had to relate to on a daily basis. Adding to this system inconsistency was the fact that the custom-designed software contained more faults than the office automation packages, causing further frustration and tension among the users.
Another hindrance to a successful adoption of this IT system lies within the military culture itself. High job rotation, instability and fear for future career all contributed to hostility towards the IT system (Tolsby, 1998). Despite this, are there other signs in this material that point towards a long-term effect of this IT system? When revisiting this research my attention was drawn to three statements that needed further investigation:"Because of the module I have had to learn more about the routines performed by person X. Since the personnel module was implemented, it has required knowledge of how the soldiers are handled within the module. The module causes dependence on the routines of others, since it requires others to add consistent data at regulated times (lieutenant employed at one of the local units (Tolsby, 1996, p. 54))."
We can see from this expression that both the requirements of the employee and how the job is performed changed as a consequence of the personnel module, causing dependence between employees. But more than this, the employee also expressed how consistent data needed to be added at regulated times. This illustrates how the design of the IT system required a particular organisational behaviour from the employee, and forced the employee to learn how other people within the organisation performed their jobs. A relevant assumption could be that, previously, employees had more autonomous jobs where less information about how other people conduct their jobs was needed. With the new IT system the employee needed to know how data were handled within the module while, under a manual scheme, papers were filled in and handed to the next person, requiring less organisational know-how and involvement from the employee. On the other hand, it could be argued that the new IT system encouraged employee involvement, and contributed to a more holistic understanding of the administrative routines and of the local unit as a whole. A moderate interpretation of this could be that the new IT system, instead of deskilling the workers, re-skilled them. Assuming that reskilling occurred, what implications did it have and to what extent does this resemble Taylorism?
From Taylor's own works we know that, through his time and motion studies, he broke the work down into small units and looked at how these could be optimised. In particular, one must consider his works with pig iron where he took the best skilled worker and made him work at optimum speed to find out how much pig iron shuffling it is possible to do in a day's work. In this particular case, Taylor's measurements related to the push-and-pull arm movements a workman does, and he used this as an indicator of the maximum amount of pig iron the optimum worker could handle in a day without becoming exhausted (Taylor, 1967). At first glance there appears to be little resemblance between scientific management and the consequences of this IT system. But there are also signs of an IT system bringing work, employees and organisation closer together - a scheme reducing job flexibility and employees' perception of their work. Interestingly, this new scheme was not employed for unskilled jobs at shop floor level, but for administrative skilled jobs. If we are able to accredit some of the changes taking place here to a Tayloristic scheme, this is in contrast to Heisig and Littek's (1995) findings. They claim that research on work within sociology has a weakness because the needs of Taylorism (i.e. distrust and the need for direct control) have predominantly focused on the shopfloor level, and not taken into account the increase in administrative jobs necessary to enforce mass production and fragmented work on the shop floor, and the fact that these jobs now are knowledge-based and do not function like unskilled jobs (Heisig and Littek, 1995). The present research, in contrast, claims that, despite having a knowledge-based level of jobs above the shop floor level, which is precisely the organisational level which provides the empirical data used in this paper, it is not possible to claim, unconsciously, that these jobs are free from Tayloristic elements. What then are the Tayloristic elements found? Typically, the new IT system conveyed a scientific way of representing the employees' work. It replaced a more lenient approach to how the job should be done; before, employees manually wrote information on personnel sheets, relying on previous experience. With the new IT system, they adapted to a prescribed way of how the work should be done, defined by the system and thus the rule-of-thumb approach was replaced by a scientific approach. Another important element for Taylor was the emphasis he put on training and teaching the employees how to do their jobs, in contrast to how the jobs were done before where the employee both chose his own work and trained himself (Taylor, 1967). The empirical evidence reported in this paper, shows how a whole "corps" of well-trained employees functioned as training agents for other employees, and illustrates how much emphasis was placed on training and teaching.
It should also be noted that Taylor claimed that there should be an equal division of work and responsibility between the management and the workers. Management should, according to him, co-operate heartily with the men to ensure that work is done according to the principles of scientific management (Taylor, 1967). Ambivalence is the word that best describes my perception of how management behaved when this IT system was implemented. They were reluctant to adopt the IT system for their own work, but quite positive about the adoption of the system by their employees. Intuitively, I assumed that management's reluctance to adopt the IT system could be explained by their fear of losing control over the organisation. However, reanalysing the research revealed that management lacked knowledge of what the different modules of the IT system could do, and how it could facilitate their work. Consequently, the management did not take on the new duties required of them and thus they did not conform strictly to the requirements of scientific management. Management's lack of skills and knowledge of how to operate the new IT system left them with too little insight to control and determine the labour process and its outcome (Zuboff, 1988). This might have led to the system having more resilience but, on the other hand, one could argue against this as Heisig and Littek (1995) do. They claim that managerial control and cohesion is actually part of a scientific management scheme and, consequently, that the management studied in this research exercised control (Heisig and Littek, 1995). But I disagree with Heisig and Littek here by claiming that we can interpret Taylor in a different way, namely that he strongly emphasised that scientific management requires an accepted division of work and responsibility between workers and managers. It is of course possible to argue to what degree such a division instills equality and respect between the two groups, but here I keep strictly to what Taylor has said. In my case, such an acceptance never occurred, since the management failed to take on the new responsibilities.
Since the employees seemingly adjusted to the new IT system, and management did not, in whose interest is it that the employees conform to a particular organisational behaviour? Management relations to this system were definitely not omnipresent. To what extent management attitude affected the employees' views of this IT system has not been explored, but I cannot rule out that this might also have had an effect upon employees and fuelled the negative attitude towards the IT system. But does this necessarily lead us to describe the new IT system as a system introducing Taylorism or neo-Taylorism? Can users' motives for using the system despite its apparent dysfunctions give us some answers as to what working scheme it imposed?"The most important thing for us was to become acquainted with the personnel module to make it work practically, and to avoid being criticised by individuals depending upon our data (lieutenant working with personnel and employed in one of the army units (Tolsby, 1996, p. 54))."
This expression shows how the employees' interest in the module was motivated by two factors. First, the employee had to do his work because others were dependent upon the data. Second, the employee was afraid of getting criticised by others. The quote clearly illustrates the lack of any intrinsic properties within the IT system itself motivating the employee to use it. In fact, the employees' motivation can be found in the way the system changed employees' work. The system created interdependence between the different jobs in the organisation and forced employees to work as the system prescribed. But is this a correct interpretation of the above quote? Heisig and Littek argue that the Taylorist model of work organisation is built on distrust, and that responsible autonomy represents an alternative work organisation model (Heisig and Littek, 1995). How do my own data conform to a work organisation based on distrust? Fear of getting criticised could be a sign of distrust, but also an illustration of how the culture within this unit worked. On the other hand, the employees' impression was clearly connected to the IT system, thus it should be fair to claim that it changed relationships between employees and in a negative way. What consequence did this change have?"It is a drawback that one can not be away from work (lieutenant working in one of the army units (Tolsby, 1996, p. 54))."
This small, insignificant statement illustrates how the new IT system changed employees' jobs. While previously it was possible for the employees to be away from work, this seems to have changed with the new IT system. Seen within a context where pressure was placed upon employees to understand how the system works and do things as required by the IT system, i.e. entering consistent data at regulated times, the IT system forced a new time schedule upon the employee. Consequently, the result was loss of personal freedom to work at one's own pace.
Taylorism, neo-Taylorism or something else?
Having seen some signs of the reskilling of employees, distrust and lack of managerial involvement, can we say that these changes have turned the studied army units into true Tayloristic work organisations? In order to establish an answer to this question it might be fruitful to investigate to what extent employees are made interchangeable. An important aspect of scientific management is how the break-down of work into manageable parts can more easily accommodate the substitution of employees. When both the amount of knowledge and skills required to do the job are limited it becomes easier to dismiss old employees and recruit new ones. In my investigation I show how union representatives claimed that members, at least initially in the project, were afraid of becoming redundant. This implies that employees believed that the new IT system would replace them. In my empirical data I do not see a direct link between the IT system and layoffs, and the fact that the army started laying people off at the time of the introduction of this IT system must be ascribed to reduced military budgets and changes in the global military climate following the fall of the Berlin wall and the collapse of the former Soviet Union. At least I did not experience any layoffs in the administrative level I studied in the two army units. Despite this, I have previously stated that employees, because of reskilling, could, to a lesser extent, make use of their previous knowledge and experience. I thus observed a more instrumental use of the employees which resembled Taylorism. But is it fair to claim the implementation of an IT system imposes scientific management? Do these concepts occur elsewhere today and do they resemble any of my data?
When studying more current literature dealing with Taylorism, it can be found that the concepts of scientific management are discussed in relation to total quality management (TQM) and re-engineering. One of those who have tried to couple Taylorism and TQM is Martin (1995). He concluded that Taylor and TQM advocates were trying to do the same thing, but that Taylor had to go through some phases that, for modern management theorists, had already been taken. Both of these methods, he claims, sought to empower the workers by improving their work techniques:"Looking at two ends of the continuum, scientific management and TQM both emphasised worker training as the way to achieve that end, and both schools of thought encouraged a group sense of responsibility in organizational decision making (Martin, 1995, p. 49)."
Martin claims that what have changed since Taylor's scientific management dawned are the workers' competence and sense of involvement.
With respect to re-engineering, Boje et al. (1997) claim that re-engineering was mostly used to justify downsizing and that re-engineering is reproducing the bureaucracy it sets out to destroy. They claim that re-engineering does this by using the same devices of monitoring, centralising information systems and instigating a redivison of labour. Accordingly, they claim that re-engineering presupposes an integrated system of processes while rejecting diversity and difference (Boje et al., 1997) or, in the words of Cooper and Burrell:"Re-engineering is the work of a central and elitist planning team, within a hierarchical administrative body, combined with legions of junior consultants claiming expertise. Behind the masks of reason and rationality is a reengineering of power relationships according to yet another mechanistic metaphor (Cooper and Burrell, 1988, p. 110)."
Pruijt (1998) is among those who claim that business process re-engineering (BPR) can be seen as an unproblematic acceptance of Tayloristic solutions. He points to the use of knowledge-based information systems to replace highly trained specialists with teams of generalists or, in a decentralised organisation, using centralised computer controls. But he does not write off re-engineering as Tayloristic; instead, he claims that a more appropriate label would be neo-Taylorism. According to Pruijt, neo-Taylorism has the following characteristics:
- a top-down streamlining of operations;
- unproblematic acceptance of typical Tayloristic solutions; and
- asserting that the outcome for workers is an upgraded work content (Pruijt, 1998, p. 262).
His conclusion is that change projects that are labelled as BPR increasingly will show a Tayloristic bias and, in our attempts to test this hypothesis, we should observe the changes occurring at the shopfloor level:"... changes in job design must be assessed in terms of skill and control (Pruijt, 1998, p. 266)."
Returning to my own data, I find some of the elements Pruijt describes as characterising neo-Taylorism. Both his claim about a top-down streamlining of operations and the claim of unproblematic acceptance of typical Tayloristic solutions are present in my empirical discussion above, manifesting themselves through the adoption of a top-down approach by the project organisation using a pool of change agents to reskill the employees, but also the lack of influence employees' union representatives had on the process. Additionally, signs were found of what Boje et al. (1997) claim is part of re-engineering - an integrated system of processes rejecting diversity and difference. The new IT system made employees conform to a particular organisational behaviour and reject their differences. Instead, the new IT system was meant to empower the workers by improving their techniques of work, which is in accordance with how TQM resembles Taylorism according to Martin (1995). Based on this evidence, I conclude that the IT system, instead of increasing employees' work engagement, actually decreased it. The new IT system introduced in parts of this huge organisation contributed to a deskilling of the employees, to a more task-oriented approach to their work, and increased dependence between the employees. Instead of increasing employees' personal flexibility and involvement in their jobs, the new IT system in fact contributed to reduced freedom and choice for the employees to do their jobs at their own pace at times which suited them.
End of story?
In August 1995 the project organisation responsible for the implementation of this IT system dissolved, and more permanent organisational units took over the support and continued its implementation. My research did not follow up this development, making it difficult to say how the continued development and adjustment of the system influenced the employees and potentially made them change their original view of the system. Nevertheless, these first reactions are still valuable since they present the initial impact and perception of an IT system. Such impressions of a system are most likely to colour the reception of the system in other parts of the organisation, and impact on the acceptance and implementation of a system. In order to understand how and why such impressions are formed and how they influence the organisation and its employees, we need to be aware of them. Only then can we construct a model of what is happeneing in our workplaces and what effects the introduction of IT systems might have. Even revisiting old research becomes important if new theory can contribute to a better understanding of the events taking place. Eventually this will enable us to make better IT systems that promote personal flexibility and involvement and where organisational understanding is based on our uniqueness as humans and important players in a community of practice (Wenger, 1998).
References
1. Boje, D.M., Rosile, G.A., Dennehy, R. and Summers, D.J. (1997), "Restorying reengineering", Communication Research, Vol. 24 No. 6, pp. 631-68.
2. Cooper, R. and Burrell, G. (1988), "Modernism, postmodernism and organizational analysis: an introduction", Organizational Studies, Vol. 9 No. 1, pp. 91-112.
3. Heisig, U. and Littek, W. (1995), "Trust as a basis of work organisation", in Charles, T. and Littek, W. (Eds), The New Division of Labour Emerging Forms of Work Organisation in International Perspective, Walter de Gruyter & Co, Berlin, pp. 17-56.
4. Martin, D. (1995), "From mechanical engineering to re-engineering: would Taylor be pleased with modern management?", Journal of Management History, Vol. 1 No. 2, pp. 38-51.
5. Orlikowski, W.J. (1992), "Learning from notes: organizational issues in groupware implementation", CSCW 92: Sharing Perspectives: Proceedings of the Conference on Computer-Supported Co-operative Work, Association for Computing Machinery, New York, NY, pp. 362-9.
6. Pruijt, H. (1998), "Multiple personalities: the case of business process reengineering", Journal of Organizational Change Management, Vol. 11 No. 3, pp. 260-9.
7. Taylor, W.F. (1967), The Principles of Scientific Management, The Norton Library, New York, NY.
8. Tolsby, J. (1996), "The intersection between organisation theory, information technology, knowledge and learning: a case study of the Norwegian army's EDBLF project", Masters' thesis for the European Education Programme on Problems in Science, Society and Technology (ESST), University of Oslo, Oslo.
9. Tolsby, J. (1998), "Effects of organizational culture on a large scale IT introduction effort: a case study of the Norwegian Army's EDBLF project", European Journal of Information Systems, Vol. 7 No. 2, pp. 108-14.
10. Wenger, E. (1998), Communities of Practice, Learning, Meaning and Identity, Cambridge University Press, Cambridge.
11. Wrege, C.D., Greenwood, R.G. and Greenwood, R. (1997), "A new method of discovering primary management history: two examples where 'little things mean a lot"', Journal of Management History, Vol. 3 No. 1, pp. 59-92.
12. Zuboff, S. (1988), In the Age of the Smart Machine, Basic Books, New York, NY.
| [Illustration] |
| Caption: Figure 1.; IT system modules |
My Research