Interview with Client's Quantity Surveyor - Establishing who's on site. page 108 ... to note two alternative ways of conceiving organisation, as a system, or as a local ...... novel wording of the contract and the absence of existing case law to assist.
Social Order on Construction Projects.
A thesis submitted to the University of Manchester for the degree of Master of Philosophy, in the Faculty of Economic and Social Studies.
1996
John Rooke
Department of Sociology
Contents. 1. Introduction: developing an alternative account of those phenomena usually treated by sociologists as social structure. page 5 2. Cultural groups and institutionalised oppositions in the construction industry. page 19 3. Finding describable order in members talk about construction contracts, by using membership categorisation devices. page 47 4. Doing social differentiation using moral identities. page 74 5. Conclusions page 93 References page 96
Appendices: 1. Membership categorisation devices: rules and definitions. page 100 2. Research project meeting - designing a questionnaire. page 104 3. Interview with Client's Quantity Surveyor - Establishing who's on site. page 108 4. Interview with two contractor's engineers - doing social differentiation. page 111
2
Abstract. The thesis is intended to lay the groundwork for an ethnomethodological description of phenomena usually theorised in sociology as social structure. A summary is made of some principes of ethnomethodological study and its relation to the project of creating a social science. This is followed by a conventional ethnographic account of cultural goupings in the construction industry. It is argued that such a study cannot be used as the basis for scientific theorising. The body of the thesis is concerned with the applicaltion of Sacks' membership categorisation device analysis to the problem of describing members talk about organisation and social differentiation. Sacks analysis is examined, adapted and applied to two transcripts of members talk. Cuff's extension of Sacks' analysis is introduced and his concept of moral identity is used to analyse the second transcript and a third. Finally, a problem of versions is briefly considered.
3
Declaration. No portion of the work referred to in the thesis has been submitted in support of an application for another degree or qualification of this or any other university or other institution of learning.
4
1. Introduction: Developing an Alternative Account of those Phenomena Usually Treated by Sociologists as Social Structure. The purpose of this thesis is to develop an approach to the phenomena usually treated by sociologists as social structure. This approach is situated in the tradition of ethnomethodological studies. It is my intention, drawing upon the insights of this tradition, to organise them in such a way as to demonstrate that they provide a coherent and comprehensive alternative to the conventional approach. By the phase phenomena-usually-treated-by-sociologists-as-social-structure, I have principally two kinds of sociological construct in mind: stratification, by which is meant: (i) forms of institutionalised inequality, such as class, raceethnicity, gender, etc.; and (ii) organisations, as this term is ordinarily understood . It is upon the latter I intend to focus here, specifically upon the organisation of construction projects, though forms of inequality will also be treated as and when they impinge upon the subject matter. What, for the purpose of analysis, is meant by the term organisation must await further clarification since, as will become apparant, this is one of the matters at issue. For the moment, it will be sufficient to note two alternative ways of conceiving organisation, as a system, or as a local acheivement. This project is inspired, in part by Sharrock and Watson’s (1988) insistence on the autonomy of ethnomethodological studies from mainstream sociological theorising. Such theorising posits a social structure whose properties are “givenness, constraint, stability,” which is balanced by a concept of social action which supplies the contrary qualities of “creativity, autonomy, fluidity” (p58). The dichotomy, once set up, demands that the theorist makes a choice between one concept or the other, or of some particular combination of the two.
5
In rejecting attempts to subsume ethnomethodological work into this dichotomous approach, Sharrock and Watson assert that, ”It is wholly misleading to treat ethnomethodology as though it were a different view of the same things that other sociologies are interested in, when it is, in truth, interested in very different things than they are, so much so that there is a sharp discontinuity between them, with ethnomethodology directing its attention to the very things which (under the rules of their tradition of theorising) other sociologies must systematically disregard.” (p59) It follows, then, that in attempting to deal with the-phenomena-usually-treated-bysociologists-as-social-structure, I am not treating the same phenomena as those sociologists treat. This seems a paradox, but in fact it is simply a recognition that different approaches to sociology constitute their phenomena in different ways. On the other hand, these different types of phenomena are related, they are constituted, as it were, from the same raw material. To clarify, the phenomena with which I will be dealing are: on the one hand, social structures as opposed to social actions (in the conventional conception); on the other, structured action (in the ethnomethodological conception). Both refer to, in this case, the organisation of construction projects (in the common sense conception of members of the industry of the construction industry). Garfinkel’s (1956, 1967) seminal project of operationalising Parsons’ theoretical machinary by testing for the hypothesised existence of culture has several important implications for sociology: 1. the logic of verstehen is worked out to its ultimate logical conclusion as an empirical/phenomenological method, holding intersubjectively established, objective meaning as its sole object of study;
6
2. this logic requires the practice of ethnomethodological indifference, both as method for and criterion of succesful study; 3. this study reveals culture to exist, not as structure, but as method (see also Eglin 1980b); 4. as method, culture is not only a topic for sociology, but also its resource; 5. this study also reveals communicative activity as sociology’s central phenomenon of interest; 6. these two discoveries, involving the recognition of the reflexive and indexical nature of meaning in language, have lead to the establishment of the criterion of unique adequacy as the highest criterion by which sociological accounts may be judged and to the creation of accounts which satisfy this criterion. Drawing upon these already achieved insights of ethnomethodology, it is possible to assume the following characteristics of the organisation of construction projects. 1. it has an observable, intersubjectively established, objective existence for members; 2. it exists not as structure, but as method; 3. it is provided for, described, instituted, manifested and displayed in and through language; 4. the methods which constitute the organisation therefore share the reflexive and indexical nature of meaning in language and are available for study by the verstehen method, requiring the practice of ethnomethodological indifference and satisfaction of the criteria of ethnomethodological indifference and unique adequacy.
7
It is the purpose of this study to explicate these features of the organisation of construction projects, in order to provide an alternative account to conventional accounts of social structure which systematically ignore these features. In this way, social structure is revealed as the achieved product of members work (Garfinkel 1996). It is however, the work itself which is the focus for ethnomethodological study and thus for this thesis. The primary means of analysis I will use will be Sacks’ (1972) concept of membership catagorisation device (henceforth, MCD). In the concept of MCD, Sacks provides a means of describing the-phenomena-usually-treated-bysociologists-as-social-structure which is empirically grounded in the study of member's talk. The methods thus described, may be characterised as members sense making activities. This is not the direction that was taken by Sacks himself however. Sacks’ interest, like Garfinkel’s, was to establish criteria for what should pass as an adequate account in sociology. This purpose is logically prior to all others, since the adequacy of answers to all other questions in sociology depend upon the answer to this one. For Sacks, the prime interest of the MCD was as an attempt to provide such an adequate account. Sacks ambitions for sociological description are set out in an earlier paper (Sacks 1963). Announcing his concern that sociology should become a scientific discipline, he states: “As scientists we seek to produce a literal description of our subject matter. In order to describe we construct (or adapt for our uses) a language, while to begin with our langauage may be crude, one rule must constantly be attended to : whatever we take as subject must be described; nothing we take as subject can appear as part of our descriptive aparatus unless it itself has been described.” (ibid p2) 8
The concept of a literal description is, however, problematic. Any attempt to produce such a description is faced, as Sacks points out, with the etcetera problem. This is the problem consists in the fact that every description is necessarily incomplete. That is to say that it can be added to, to make it more complete. He correctly identifies the roots of this problem in the nature of language as a researcher’s tool, rejecting Weber’s formulation, that it lies in the nature of the social world as an object of empirical study. His implied hope is that it can be overcome by an analysis of the language used in descriptions and an elimination of assumed common (and therefore undescribed) experience. However, the indexical nature of language is unremitting: all uses of language aquire meaning from the context of their use. This means that an appeal must always be made to shared assumptions and experiences; any analysis of language, however exhaustive, will leave something more to be said. It turns out, though, that this is not after all a problem for sociological description. Since all accounts are produced in and for particular contexts, their adequacy is judged by criteria supplied in and by those contexts: they are judged by their sufficiency for all practical purposes (Garfinkel 1967). Such a criterion is given by Sacks (1963) when he compares conventional sociological and common sense perspectives, through the metaphor of a ‘commentary machine.’ the commentary machine is a machine which performs a task, while simultaneously producing a description of what it is doing. “Suppose that the machine has a recurring cycle. It procedes through a course of its activities, and having finished a course, begins over again. From the common-sense perspectives, problems of the encounter arise as problems over the course of the encounter; a ‘description’ need only be ‘good enough’ to permit the encounter to to proceed. Possible misunderstandings may be left till they actually raise difficulties, and when 9
they raise difficulties they need only be solved as far as is necessary for the encounter to proceed. The fact that different [encounterers] in the former class come away with different reports is no cause for embarrassment; it but testifies to the uniqueness of each’s encounters. The latter, the sociologist, seeks to warrant his description of activities as scientific. He seeks to write a description which would be produced by a colleage observing another cycle, or which could be used by a colleage for analyzing the machine’s course of activities.” (ibid p8-9) This comparison poses the etcetera problem in a way which suggests a solution. The analysis of descriptive terms and the further description of previously assumed experience must necessarily be pursed by sociologists. This pursuit, however, is only necessary to the extent that it acheives the practical purpose of making a particular account objective, in the sense of being an account which would be produced by another member of the sociological community. It is in this sense that Sacks’ purpose in developing the MCD can be understood: “to construct a description that provides the reproduciblity of the conclusion a suicidal person may reach - I have no one to turn to” (Sacks 1972 p31) In other words, to facilitate a formal sociological generalisation which would stand as an account of the process by which a suicidal person (any suicidal person) reaches the conclusion, ‘I have no one to turn to;’ such an account to consist in a literal description of social life and to be reproducable, from the data, by any other sociologist. This can only be acheived by adopting the correct attitude to common sense accounts of social life: 10
“even if it can be said that persons produce descriptions of the social world, the task of sociology is not to clarify these, or to ‘get them on the record,’ or to criticise them, but to describe them. That persons describe social life (if they can be concieved as doing so) is a happening of the subject quite as any other happening of any other subject in the sense that it poses the job of sociology, and in contrast with it providing a solution to sociology’s problem of describing the activities of its subject matter.” (ibid P7) In his pursuit of literal description, Sacks abandoned the elaborate analytic machinery of the MCD, using only a small part of it in the subsequent development of conversation analysis (henceforth, CA). It seems the reason for the development of CA was simply that the mechanics of conversational turn taking were ubiquitously available for study, under the principles of unique adequacy and ethnomethodological indifference. This availability enabled the attainment of a degree of empirically referenced objectivity and precision of analysis, previously unknown in social studies. However, while this was an extremely productive move in terms of attaining Sacks' ends, it does seem to have led to something of a cul-de-sac. While a separate strand of ethnomethodological studies has developed the ethnographic tradition into a profoundly detailed and reflexive discipline (see, for instance, Wieder 1974; Garfinkel, Lynch & Livingston 1981; Anderson, Hughes & Sharrock 1989) Sacks' program of producing scientific generalisation has remained confined to a study of the mechanics of conversational exchange. It is with the intention of discovering whether such generalisation can be achieved in other areas of human experience, that I proceed. Within CA itself, there exists an attempt to extend the programme to the study of those-phenomena-usually-treated-by-sociologists-as-social-structure. This
11
attempt is based on examining the impact of institutional situations upon the formal properties of the talk which takes place within them. However, as Hestor and Francis (unpublished manuscript) observe: this leads to importing unexamined common sense constructs, in order to describe the institution in question. Such a move undermines the purpose of the whole Sacksian project. The intention here is entirely the opposite: while largely abandoning the substance of CA analysis, to preserve, in the study of a different topic, the spirit that led to its creation. It is in order to achieve this that I have turned to the work Sacks which preceded the development of CA, the MCD. The membership categorisation device is claimed to have a very different status from the concept of social structure, in that it is the analytic tool that members themselves use to make sense of the social world. This is made clear in Sacks’ articulation of what is involved in providing for the reproducibility of his data: “attempting (a) to locate the collections of membership categories in terms of which the search for help for suicidalness is formulated, and (b) to describe the ways such collections are used to determine whether there are eligible persons available (to give help).”(ibid. 31-2, my emphasis) This claim is endorsed by Garfinkel (1991), who cites Sacks’ work as among those studies which satisfy the criteria of ethnomethodological indifference and unique adequacy. This claim is very different to that made for conventional sociological analysis which, in competition with members own accounts, offers improvements, developments, or formulations of greater coherence, consistency, or clarity. The profound difference between these two claims is spelled out by Garfinkel in his rejection of the suggestion that the two approaches can be reconciled, with specific reference to the theorising activity of Talcott Parsons:
12
"For one example, a recurrently used method consists of designing a formal scheme of types, giving their formal definitions an interpreted significance with which to develop and explain the orderly properties of the types as ideals, and then assigning the properties of the ideals to observable actions as their described properties of social order." (Garfinkel1991,p12) To give more substance to this example: in chapter two, I use this method to describe a range of attitudes to contracts which I depict as existing on a continuum between ideal type constructs. These ideal types are generalisations generated as summaries of a variety of remarks made to me by managers. Their purpose is to provide the reader with a brief and easily digested summary of my experience of research. As such, they represent a very low level of analysis. Nevertheless, they are potential first steps in a programme of constructive analysis, whereby all the data gathered here is available for characterisation in terms of these ideal typical constructs. Such an analysis would reveal whether, to what extent and in what specific ways these recorded events reflected the specifics of these ideal constructs. For example, in the section on designers drawings, I reproduce some talk in which two engineers, working for a contractor organisation characterise the attitudes of design consultants and clients. It is possible to see within this talk, features which can be taken as representing the ideal types. These features are, in fact, precisely those which contributed to my initial formulation of the types. Thus: things just don't fit together properly. It's quite understandable, because every bloody contract is unique, isn't it. It has to be redesigned. This piece of talk, displaying as it does a sympathetic attitude to the designer, is precisely the kind of remark that led me to the formulation of the integrative type.
13
It can be contrasted with a remark by another contractor, which can be characterised as 'integrative': It's their own bloody fault, if they got the drawings right in the first place, there wouldn't be any opportunities [for claims]. By a reversal of the process of induction by which the analysis in chapter two has been achieved, it is possible to see these pieces of talk as manifested examples of structures which lie behind and account for the attitudes displayed. A second piece of talk, taken from the same section of chapter two will illustrate how the ideal typical structures can be seen to manifest in an interactive way to account for subtle variations of attitude: 1
[manager] Under the fifth edition, the contractor builds bloody everything,
2
I mean, y'know, if he's got a profit in it. If there's some profit built into
3
one particular bit of work, he'll do the bit of work
4
[interviewer] right
5
[manager] We've both had cases where we have gone to a client and said
6
hey you can save money if you do it differently, he's said, oh, thanks very
7
much and issued the instruction to do it differently and kept the saving
8
himself. That's happened to me
9
[interviewer] right
10
[manager] Usually you propose it on the basis of let's share the saving and
11
some honourable clients do that. But these days I would tend to expect
12
most clients to say that's a good idea, thank you very much and keep the
13
savings themselves. In which case, sod that, you know, if you've got some
14
of your prelims built into that item, then you've lost out haven't you,
15
straight away?
14
Here, lines 1-4 can be taken as a manifestation of a distributive structure, while the succeeding lines can be taken as a manifestation of an integrative one. (the integrative type includes within its definition a recognition that the contractor must recognise and defend against distributive attitudes on the part of the contractor.) When the analysis achieves this level of abstraction, it begins to obscure the common sense understanding that we are dealing with flesh and blood human beings. The speakers take on the aspect of passive vessels for the real social forces which shape their actions. By multiplying the number of ideal types used it would be possible to immeasurably enrich this analysis until, at least in principle, all the recorded talk could be accounted for in terms of manifestations of such ideal typical structures. Thus, it would seem that the real structures which lie deeply hidden behind our common sense understandings of social life are being revealed. The ultimate implications of such an analytic stance are, again, best expressed by Garfinkel: "With his concreteness/analysis pair, Parsons demonstrated [though see my conclusions for a commentary on Garfinkel's usage here] there was no orderliness in concrete activities. With his plenum, Parsons specified the analytically empty concreteness of organisational things is not yet real organisational things. Nor is it yet organisational things produced according to, let alone consisting of, methodic procedures - call these 'actual' organisational things - nor is it yet organisational things evidently." (ibid. p13) Thus, it is not even that our ordinary understandings are flawed, or even that they are wrong, or even nonsensical. They are simply irrelevant. A distinction has been introduced between common sense understandings as data and social
scientific analysis as actual analysis, which brooks no comparison between the two. In a total reversal of analytic direction, ethnomethodology sets out to investigate the orderly properties of members own analyses of their social situations. At the end of chapter two, I have appended some further remarks on ethnographic methods and their relation to social theory. The bulk of this chapter, then, takes the form of a straightforward ethnography; an attempt to summarise some of the real life constraints which for managers in the construction industry are experienced as the social environment within which and with which they work. What is not spelled out there, but which is equally true, is that these constraints are constituents of and are constituted by managers work. What follows, in chapters three and four, represents an attempt to investigate some of the methods by which those managers constituted, displayed and utilised those constraints in their discussions with me and with each other. Consequently, it stands also as an account of the methods I have used to formulate the ethnography. The analytic tools I have used in this investigation are essentially the MCD apparatus developed by Sacks (1972) and Cuff's (1994) extension of that apparatus by the introduction of moral identities. These two chapters involve an attempt to apply the principles outlined in chapter one of Garfinkel's (1984) Studies in Ethnomethodology. These principles embody a recognition that expressions and actions are indexical, reflexive and rational; and that they are produced and perceived as such by members. This is a matter of unremarkable routine for them. For ethnomethodological studies, however, it is the exclusive topic of interest and investigation. A first principle of this investigation is that it is the very fact of the orderliness of those expressions and actions which makes them investigable and reportable. The analysis of this orderliness is nothing more than an attempt to recapture and make explicit the implicit analysis by which it was produced by members. Thus, 16
the indexicality, reflexivity and rationality of members actions and expressions is also the sole resource for these studies. The investigation is, therefore, thoroughly empirical. It consists in a focused investigation of actual, real life, real time expressions and actions by members. From this investigation can be seen the local productions of immortal ordinary society. These productions are the totality of what empirical real life, real time society, as lived by members consists of. That they contain an analysis means that we have no need to add a theory. Any additional generalising, or predicting we may wish to do, itself has the status of a members analysis. In making such additions, we are no longer in the business of simple study; rather, we are joining in with the business of producing social order ourselves. We are producing nothing essentially different to our data, we are simply adding more data to the pile. The materials analysed in chapter three and presented in full in appendices two and three, have been selected because they display concerted efforts to categorise populations. The additional material analysed in chapter four (appendix four) was selected for its richness in examples of categorisation. This was intended to provide an opportunity for an application of the full range of Sacks' analytical machinery. Cuff (1994) uses only a small part of that machinery in his own analysis and points out that it required considerable work to make even this much applicable to his materials. He concludes: "We suggest that the difficulties we have encountered in seeking to apply Sacks' work on members' sense-assembly machinery to our materials does, however, have possible serious implications for its generalisability and hence for Sacks' suggestion that though this machinery might have seemed 'overbuilt' for the materials he himself analysed, it might be found to be widely applicable." (ibid,p94) 17
With this in mind, my intention was to provide the most favourable possible conditions for an extended application of the MCD analysis, in order to determine more closely which parts of it held promise for more general use and which parts did not. This strategy contains a danger, particularly with regard to the materials analysed in chapter three. These two transcripts, dealing as they do in extensive categorisation of populations, might seem to add support to the idea that social structures have an objective existence for members. Thus, while not supportive of the more extreme position attributed by Garfinkel to Parsons, they may seem to underpin an interpretation of Schutz which holds that conventional social science conceptions of social structure represent a logical improvement and development, a 'tidying up', of members own understandings. It is not my intention to promote this view and a summary of Eglin's (1980) work on the production of lists in interviews has been included to act as a corrective.
18
2. Cultural Groups and Institutionalised Oppositions in the Construction Industry.
Background to the Study: The circumstances under which the research reported here has been carried out are somewhat unusual. Working as a researcher on a project funded by the Engineering and Physical Science Research Council and based in a school of civil engineering, my work has necessarily been primarily answerable to the academic constraints the discipline of construction management. Much of the material in this chapter is taken directly from reports and papers written to satisfy those constraints. On the other hand, one of the three investigators who authored the initial research proposal has a background in sociology an the clear intention behind employing a researcher with a first degree in the subject, was to import some sociological ideas and methods into what is essentially a multi-disciplinary field, though with its historical roots in civil engineering. The research has been carried out under the title 'The Impact of Culture on Project Performance Under the New Engineering Contract'. Designed as it is to address academics in construction management and introduce them to sociological concepts, this title displays some significant features. First, the research intends to produce results which will be of practical use to managers - the focus of interest is on 'performance', with the implication that some light will be shed on issues which will enable performance to be improved. Second, the sociological concept of culture is being introduced into this management studies agenda. Third, a particular version of 'culture' is being implied which sits more easily with the conventional than the ethnomethodological approach. Indeed, the suggested solidity of the structural nature of 'culture', that it is a phenomenon whose
19
properties are analogous to those of a physical body, is conveyed in an extremely graphic manner by the use of the term 'impact'. (It is at the same time true that this rubric has been used as something of a cover under which to smuggle into construction management the view of culture as method outlined in chapter one. It is also the case that the metaphorical power of the term impact is not exhausted by its implications of physical or quasi-physical properties. On the contrary, it has been an important part of the research effort to make plain the ubiquity of what we might equally well call culture, people's sense making methods, or structured action. A failure to recognise this phenomenon, which amounts to a failure to recognise the nature of all social life, has a potential to effect a project like the New Engineering Contract in a way which might be closely analogous the effect of a physical blow.) What follows is a short ethnography describing the cultural milieu in which the New Engineering Contract has been developed and the changes it is intended to effect within that milieu. This is a necessarily superficial description, but it will I hope be sufficient to its purposes: first, to acquaint the reader with some forms of structured action and corresponding cultural groupings which are commonly recognisable to members of the construction industry as features of that industry; and secondly, to show how those same features are theorisable as social structure, using the conceptual apparatus of conventional sociology. Or, to put it the other way around, how the matter-of-factly arrived at abstract generalisations which structure the account constitute quasi-theoretical constructs which depend upon, but do not explicate, the real life actions to which they refer. To be more precise, these 'features' are generalised abstractions which either are, or are capable of being, used by members to describe, account for, move within, manipulate, maintain, or whatever, their social environment; at the same time they are the
20
building blocks from which, either theory is built; or the second order constructs from which theory is built are derived (Schutz 1973). Subsequently, it will be possible to take a closer look at some specific instances of the contextual demonstration and utilisation of those features by members, thus initiating the development of an empirical (as opposed to theoretical) analysis, as outlined in chapter one. In other words, the approach I will be taking to these features, will be to use them as signposts which point back to the situated activities from which their existence has been derived.
Relationship between designer and contractor. From the contractor's point of view, the designer is effectively part of the client's team. Those that design the project traditionally supervise its construction for the client. Relationships can take a variety of forms, for instance: designers can be direct employees of client, e.g. until recently, local authorities or designers can be employed by a consultancy which itself has a contract with the client. More recently, the relationship between designer and contractor has changed: it is often mediated by the employment of professional managers to administer the project for the client; 'design and build' contracts reverse the relationship, making the designers the employees of the contractors
Designer's drawings. Problems with drawings were the most common complaint of the contractors I spoke to.
21
1. Sometimes the contractor sees this as inevitable, because of the one-off nature of construction projects: "things just don't fit together properly. It's quite understandable, because every bloody contract is unique, isn't it. It has to be redesigned." 2. However, they sometimes detect deliberate over design by engineers.
This is
attributed to the fact that design consultants are usually paid a percentage of the total cost of the project. Thus, although their contractual position means that they bear the risk of design failure, they stand to benefit financially from the increased costs of over design. One contractor told me: 'that's why you see so many ugly, over-designed buildings about.' In this context, he illustrated for me, with a certain show of bitterness, the difference in status between designers and contractors. But he's professional, you see, he's paid per hour, cos he's like the lawyer, he can be trusted you see. Whereas contractors, well nobody likes paying us like that, we can't be trusted, emm... sssssshhhhhhh, what else does NEC do? I don't know. It's good training for our own guys I suppose Except under design and build contracts, however, where the contractors would bear the additional costs, over-design does not present them with a practical problem. Indeed, they usually stand to benefit from the additional work it generates. 3. Obscurity of drawings is a problem which does affect contractors adversely. One site agent showed me a sheet of drawings to explain how he had built to the wrong specifications. The instruction to build the section in question was in the form of a drawing, at the bottom left hand corner of a sheet crowded by drawings. Beneath it was a detail, showing how to construct one particular part of that section. This was the instruction which the agent followed. Underneath the detail, however was a written instruction not to build what was pictured there and 22
referring the reader to a second detail which was positioned above the instruction to build the section. It was this drawing which the contractor should have followed. Since the contractors had built the item pictured in the first detail, they wanted paying for the work they had done; the client refused. I am told that the contractors have no case in law. Contractors are often in a position to correct the shortcomings of drawings, but as one contractor explained to me, there are considerable pressures on them not to do so: [contractor] Under the fifth edition, the contractor builds bloody everything, I mean, y'know, if he's got a profit in it. If there's some profit built into one particular bit of work, he'll do the bit of work [interviewer] right [contractor] We've both had cases where we have gone to a client and said hey you can save money if you do it differently, he's said, oh, thanks very much and issued the instruction to do it differently and kept the saving himself. That's happened to me [interviewer] right [manager] Usually you propose it on the basis of let's share the saving and some honourable clients do that. But these days I would tend to expect most clients to say that's a good idea, thank you very much and keep the savings themselves. In which case, sod that, you know, if you've got some of your prelims built into that item, then you've lost out haven't you, straight away? Furthermore, if the design consultants are also administering the project, they may well be unsympathetic to any proposed changes: 23
Well consultants that have designed it you see, especially if they're on a percentage of the fee, don't always take kindly to being told that y'know, Fred Bloggs Contractor Ltd. can think of a design which is cheaper which then reduces the client's fee. And in fact there was a job which I bid a couple of years ago, which we came up with quite a dramatic alternative, which saved sort of 20%. It was quite profound, I'm not sure what happened in the end, but we bid it, we bid this as an alternative and when we rang up the consultant to say how how'd we go on? - you bastards. This was his reaction, because em, not only had we embarrassed him by coming up with a design 20% cheaper, his fees had gone down by 20%, because he's on scale fees you see, the consultant's usually paid a percentage of the costs of the job. When asked how the efficiency of the industry could be improved, the most common suggestion from contractors is that they should be brought into the planning process at a much earlier stage, so that they can make an input into design.
The economics of the construction industry. The culture of conflict manifests in the economic arrangements with which practitioners have to contend. Construction work involves a high level of unpredictability. In addition to the usual problems of project management - principally, those of one-off design and of short term relationships - the physical conditions under which work takes place are highly significant in this respect. Bad weather can, of course, stop work with both time and money implications. A more contentious problem, however, is unforeseen ground conditions (clause 12 conditions under ICE). Thus, one contractor told me:
24
We had a case on this job, very early days. Er again down to us taking the risk, or not as the case may be. It’s more of a test case than anything else I don’t know if you want to go too much into this, but essentially we had contractor design piles. And the way the enquiry was couched, we were deemed to take on board responsibility for ground conditions. But, contradictory to that, we were instructed to price in accordance with a site investigation report which was given to us. The site investigation report showed a filled site. Which broadly it was, because it used to be an old brewery here which was demolished in on itself. But what none of it showed was the fact that about three and a half metres down was sandstone []stone and yorkstone [base with a slab], all over the site. You can’t pile through four inches of yorkstone, so we had to excavate all around the site grub all the yorkstone out, reinstate it and then do the pile operations. And the architect said, it’s your problem. I asked if this situation was due to an inadequate initial site investigation, the contractor did not think so; it was just unfortunate, he said, that the test holes had been sunk at points that happened to coincide with the few places where there was no slab. It is a major issue who is to bear the risk for such apparently unforseeable occurrences. Usually, this risk is carried by the client. The generally accepted principle is that, if contractors are to accept a risk then they should price for it. This would involve some kind of calculation of the cost and likelihood of unforeseen conditions arising; a premium would then be added to the tender price to reflect the possibility of such circumstances occurring. In this case, the contractor feels aggrieved that he was asked to price in accordance with the site report which apparently ruled out any underground obstructions. In the event, the architect accepted that the contractor's case had some merit and it was agreed that the cost of removing the slab should be shared.
25
The industry is under-capitalised. Construction projects are very expensive. Coupled with the fact that out-turn costs tend to be higher than tender price, this often results in late payments. Contractors must finance the construction work until such time as they are paid. Since margins are narrow, contractors depend on a quick turnover in order to generate profit. Another way of easing cash flow is by withholding payment from sub-contractors. Some disputes may arise simply as delaying tactics, because one party is unable to pay what is due. The industry is undergoing a recession and is highly competitive. Competitive tendering is the crucial mechanism by which competition is carried out in the industry. To the extent that clients allow price to be the sole criterion in selecting tenders, this has the effect of producing unrealistic tenders. This is because contractors will compete, not only on price, but on their ability to generate additional income from changes. Perhaps the best way to introduce a discussion of this contentious area is to reconsider the two claims already discussed above. I related these from the contractor's point of view; however, there exists and alternative interpretation. The substance of the first claim was that the drawings which the contractor received were misleading, causing the contractor to build to the wrong specifications. The work then had to be torn down and redone, costing the contractor in time, materials and labour. I have been told that the contractor had no case for reimbursement for this expenditure. He is nevertheless, pursuing the claim. It is quite possible that a contractor, spotting ambiguity in the drawings and believing that they provided the basis for a successful claim, would deliberately build the wrong item in the hope of generating additional profit. Similarly, it might be argued that in the case of the unforeseen ground conditions, the contractor clearly understood that he had contracted to carry the risk of an underground obstruction and should have priced for it; he is now trying to exploit 26
a subsequently discovered ambiguity in the wording of the tender documents to evade his responsibilities. (The contractor's reply to this might be to say that, if he had priced for the risk of unforeseen ground conditions, he would have failed to win the tender.) Identifying a contractor's behaviour as 'claims conscious', as opposed to 'reasonable', is not a simple thing to do therefore, at least for the outsider. This is particularly the case since a necessary part of claims conscious behaviour is to assert convincingly that one is being reasonable. The procedures described below are ones which have been described to me by contractors as tricks which are commonly used to get more money out of a job. If the work is held up due to circumstances beyond the contractors' control, they can claim 'day works'. In part, this involves the client paying for labour and machinery allocated to the job which now must stand idle. When the contractor foresees such an event, he can move all his excess productive capacity on to that part of the works, in order to claim for it. One manager has described a situation where, not only excess plant on site, but any, old unused plant in the companies possession would be moved to a site where a delay was foreseen. Sometimes, delays can be foreseen at the programming stage. For instance, delays are often associated with the work of 'statutory undertakers'. These are providers of services such as gas, water and telephone lines, who usually must be given access to a site in order to install or redirect the various pipes or wires for which they are responsible. Thus, if heavily resourced work is programmed to commence immediately after a statutory undertaker is programmed to finish, this is likely to lead to the contractor being paid for resources which are standing idle. This is of particular value in times of recession, when the contractors chief problem is finding ways of profitably employing excess capacity.
27
A third way of generating extra profit is by weighting the prices in the bills of quantities. Issued as part of the tender documentation, the bills of quantities are analyses of the proposed construction project in a quantified form. Thus, typical items might include digging out and removing from site a quantity of earth, or building a specified area of brickwork. If mistakes can be spotted in the bill, this can be very lucrative. For instance, if a contractor feels that the necessary amount of earth to be removed has been underestimated, while the amount of brickwork needed has been overestimated: he can put a very high price on the earth moving and a low price on the brickwork. If his forecast has been correct and the quantities are adjusted accordingly, once work has commenced, the reduction in the quantity of brickwork leads to a small loss of payment to the contractor, while the increased quantity of earth leads to a large increase in payment. Some companies employ estimators whose sole job it is to search the tender documents for this kind of 'opportunity'. There is a tendency for contractors to submit tenders at prices which allow only a narrow, non-existent, or even negative profit margin, with intention that the profit can be made up once the contract has been won and work has commenced. Client's, of course, are generally unhappy when the out-turn cost of a project begins to rise above the agreed tender price. This is the basis of the 'adversarial relations' which are widely held to characterise the industry.
Attitudes. It is often commented that the construction industry is 'fragmented'. The temporary nature of projects and hence, of working relationships, presents particular obstacles to the development of trust and co-operation. In such an environment, individuals do not have to live for long with the consequences of generating ill-feeling.
28
The same applies to corporate behaviour. Thus, there is often no strong social pressure to counteract the effects of competitive tendering. In classical economic theory, the overriding market force is customer satisfaction, which generates more sales for firms. The nature of the industry militates against the effective operation of this mechanism, however. Since the nature of construction projects means that each unit on production requires considerable investments of time and resources and repeat orders are infrequent, the incentive to win more work by pleasing a client is relatively low, compared to the risk of sustaining a damaging loss. This is particularly the case when clients focus on price rather than quality. It may also be the case that particular organisational structures among contracting companies tend to encourage this fragmentation, as when they are organised on a national, rather than a regional basis. There follows an attempt to describe the range of attitudes to the client-contractor division using Weber's ideal types. The ‘claims culture’ may be described in terms of four types of viewpoint. Two of these refer to the client and two to the contractor. The integrative and distributive views can be seen as opposite poles on a continuum of attitudes. These viewpoints represent pure or ideal types (Weber1947) and do not describe the views of any particular individual: an individual may hold a mixture of views, or may express different views in different contexts. Nor do they necessarily exhaust the range of views which might be held. Contractor’s View - Distributive. The client is out to get the best possible deal by ‘screwing’ the contractor at every opportunity. It is therefore both necessary and just that the contractor should adopt the same approach. Since clients are stupid as well as greedy, by and large, contractors will always get the better of them. A combination of heavy price competition and inadequate design and method statements means that contractors must quote for low or non-existent profit 29
margins. This is so because other contractors will use various tricks to increase the out-turn price after their artificially low tender has been accepted and the contract signed. Any contractor who does not do this, will fail to get work. Contractor’s View - Integrative. The aims of any project are to do a good job, on time and at contract price. The best way to achieve this is through developing cooperative relationships with the client, which contribute to a pleasanter and more efficient working relationship. Nevertheless, profitability is an issue of survival and where it is threatened, overshadows other considerations. Integrative contractors will try to avoid unprofitable contract prices, but may be forced by competition to become claims conscious. As one told me: If you get to the end of a job and you've made a loss, you look at bloody everything, to see if there are any commercial opportunities you've missed. Client’s View - Distributive. The aim of procurement is to get the job done as cheaply as possible, the contractor’s profit is not a consideration. Contracts should be designed to bind the contractor hand and foot. Contractors will cheat the client if they can and they cannot be trusted. Client’s View - Integrative. The contractor’s profitability is essential to the health of the project and of the industry as a whole. Fair-minded contractors will act in a reasonable manner, if they are treated fairly. Nevertheless, many contractors will attempt to ‘put one over’ on the client and this possibility must be guarded against.
Finding Common Ground. The kind of difficulty experienced by a contractor attempting to take a reasonable stance towards a client is illustrated in the transcript reproduced below. [contractor] Contractors have created this problem themselves over the years, having been involved in contract management on the financial side of it, I’ve personally put together a lot of claims submissions. An’ I firmly 30
believe that what you put forward, you must be able to substantiate, you must be reasonable. As much as the engineer has to be reasonable, contractors have to be reasonable and therefore you should put forward a document that, you're happy with, that you genuinely believe is your entitlement. A bit of gamesmanship might be slightly due on that, because nobody is going to pay you what you ask for. But there are so many clients, [names large client] and others, who just say, well I only ever pay 25%. [...] Contractor's have historically produced claims that are fiction. [interviewer] Right. So then it becomes very difficult to produce one that isn't. [contractor] No, it's not difficult to produce one that isn't. [interviewer] But get it accepted. [contractor] It's very difficult to get a client to accept that if you say you’re entitled to a hundred thousand, look, maybe 93 thousand's the right answer. Not 25. This kind of mismatch between the expectations of a particular client and a particular contractor are common. As can be seen from the above, it is not that the speaker is naive in his predictions of what the other's expectations will be, it is rather that he wishes to impose his own way of doing things. In the following extract, a client describes what he sees as a reasonable attitude on the part of a contractor: [interviewer] Do you reckon that the relationships with contractors are non-confrontational?
31
[client] You say non-confrontational, but like any civil engineering contract, there are commercial issues which they want to pursue. Obviously, from the client end we've got to put a case up where we feel that they're not justly [ ] where we feel that there is no substance to those claims. You still have the same situation whether its a claim or a compensation event, where they are required to define what that compensation event is and then to ascertain under the contract whether it is or not and you do get a degree of creativity when there's a contract of this size. Obviously they want to work on this, work diligently, professionally and leave it in profit, they don't want to be working their buns off and leave in three years time and make a loss, which is understandable. Where the NEC works is that they can't let things go. They can't write a global letter saying, I've received this instruction, there is a cost implication, and I'll let you know the consequences in due course. [interviewer] And are Contractor B doing that anyway under the ICE ? [he has previously expressed satisfaction with the way Contractor B is operating under this contract.] [client] Contractor B are doing that anyway under the ICE. I Can't say we're not getting any letters that leave it open to them to come back later, it would be remiss of them to put themselves in that situation, under the contract they do have the ability to come back later. [...] [client] I can say with all confidence, if that job on cut and cover had been written on the back of a fag pocket it would work, cos the people making it work. You can't cite that as an example.
32
It is to be expected that contractors will take advantage of favourable clauses in the contract, as a fair minded client, he accepts this. On one project, I was able to attend monthly project meetings between the client and the contractor. I made the following observations about the way that differences in interest were dealt with, while maintaining good relations and an orderly flow of work. 1. Claims, which are of their nature contentious, are continually put off, until we can get a better picture of what's happening. 2. It was formally agreed at the pre-contract meeting to keep the administration on site. Thus, the monthly progress meetings, letters etc. stay at site level. 3. After most progress meetings, the contractor treats the client to a meal. This is a way of allowing those involved to get to know each other, making it more than just a business relationship. 4. Business is conducted with much humour. Contentious issues and conflicting interests are often referred to in a humorous manner.
Engineers and Quantity Surveyors David Seymour (Seymour1986) has identified two separate orders of legitimacy in the industry, the technical and the economic. These imply the recognition on the part of participants that there are norms which govern the way people seek to optimise their economic interests and there is a different set of norms which govern the processes of design and construction. They coexist but sometimes they conflict as when, for example, a QS insisted that a pump, ordered to site by an engineer for the technically necessary purpose of draining a trench, be sent away until it was authorised by the RE who would, thereby, accept the cost of its
33
hire on behalf of the client. The water, it seems, was entering from an underground source and was therefore not the responsibility of the contractor. It is significant that it was the engineer who responded to the norms of the technical order, while the QS responded to the norms of the economic. Thus, those orders which are implicitly invoked can be used to categorise members according to the nature of their work. The contract documents have different value relevance among the different occupational groupings and relations between these groupings have an important impact on the administration of contracts. This may be seen from another example. An engineer in the civil engineering division of one construction company related the following history: “There was a time when engineering was a centralised service... the agent asked for engineers out of the pool. The engineering service idea stemmed from our building activities, where line management - site agents - were not engineers, but tradesman or ex-quantity surveyors, and required an engineering service. Engineers were used on site in a service role, and had no executive authority. Many engineers took the veil, and went into line management, and didn’t touch engineering again. However, about five years ago, we in the civil engineering division got hold of our own engineers, and are now giving them their own careers; we are getting this service thing out of the system.” (Seymour1986:200) This increase in status for engineers (as engineers, rather than as managers) is accompanied by a change of status for quantity surveyors. “It’s the same with quantity surveyors. Due to our building ancestry, they were until recently, a race apart, independent. Their job was to get as much income from the client as possible, while the site agent’s job was to construct as cheaply as possible. Since we became semi-detached from
34
building, quantity surveyors have now become part of the management team on site. They didn’t like this loss of independence at first...” (ibid.) In the view of the engineer quoted above quantity surveyors, as long as they retained their independence, chiefly occupied themselves with the use of the contract to increase income from the client. When they became part of the team, however, this function no longer retained the same importance. For engineers, the commercial issues tend to be more peripheral, as their primary concern is with the actual work of construction. My own research has confirmed the significance of the two orders of legitimacy which Seymour identifies. However, if the balance of power was shifting towards the engineers in the company which he investigated, it is not the case that this has been the general trend in intervening years. The following except of transcript is taken from a discussion with two engineers who holds senior management positions in a contracting firm. Manager A is explaining to me why it is usual for claims to remain unsettled until the job has been completed: [manager A] I mean what's traditionally happened is the client has not wanted to make a decision, the contractors have wanted to keep their powder dry and see what they needed to ask, so that's why. Y'know, it's kept till the end of the end. It's all a damned good job creation scheme by quantity surveyors, erm, who are a growing infestation in the industry and have been for ten years or more. It's a great job creation scheme and we've had experience, we think we've got one surveyor in our organisation who's bloody good at this. An erm, no names, no pack drill, but his style appears to be just to put in any old interim account during the period of the job and [interviewer] yeh
35
[manager A] the job loses money and er he pulls it round at the end. Y'know, he should have been fucking doing it all the way through [laugh] [interviewer] right [laugh] [manager A] but, but we're both engineers, so we, so we've got a bit of a hang up about surveyors [manager B] let's face it, quantity surveyors were only engineers clerks and we as engineers have got ourselves to blame [manager A] yeah [manager B] if we've managed to make [manager A] um, we didn't like dealing with things like this and we've we've left it to them, as the accountants. However, the grievances are not all one sided. I have described above the kinds of complaints which contractors have about drawings. Client's quantity surveyors (PQSs) can also find these a source of difficulty. A PQS told me: "they change the drawings and don’t highlight on the drawing, it’s very hard to pick up. Sometimes there’s a space for revisions, but they just put general revision or something like that. You get a big lump of drawings through an’ I have to go through each drawing to see, oh, that’s different to that, that’s different to that. An’ you can be talking fifteen revisions sometimes, to spot, all I’m concerned about at the end of the day is the last revision, but there’s a long history through the other fifteen revisions to er whatever was tendered for. (...) Obviously they’re issued to the contractor as well. Now he’s going to obviously spot things where there’s more cost, but there will be savings as well, which obviously I can’t expect a contractor to.. I mean sometimes they do, sometimes they don’t. But er,
36
to spot savings as well as additions. Obviously, additions he’s going to spot." The PQS is describing a similar situation to the one discussed with the contractor above: a change has been made, but not highlighted on the drawings. The PQS's concern is, of course, different to the contractors'; he is attempting to find changes which reduce the cost of the work, so that he doesn't overpay the contractor. Nevertheless, he experiences the same handicap. He observes that he cannot expect the contractor to "spot" changes which save money. There is an ambiguity in this use of the word spot: does he mean 'notice', or does he mean 'inform me'? If the former, then it is implied that it is a common occurrence for contractors to build things they shouldn't: sometimes they get paid for this and sometimes they don't. On the other hand, it may be that it means he cannot expect the contractor to inform him of changes which reduce cost, because if he doesn't notice them and pays for the work originally specified, the contractor will benefit financially. If this interpretation is correct, then it is perhaps surprising, taking account of the jaundiced view of industry morals which is generally expressed, that the contractor should sometimes inform the PQS of such changes.
Engineers and Lawyers: The New Engineering Contract. Though this is rather more artificial, it is possible to represent the relationship between engineers and lawyers as part of the opposition between technical and economic orders. Essentially, the lawyers concern is with contracts and these are usually only referred to in order to decide economic issues; technical ones are the province of the engineer. The New Engineering Contract (NEC) is an attempt to change this by reducing the incidence of commercial disputes and putting the engineers back in charge. The basic principles on which the NEC is based are set out in the guidance notes to the
37
contract (ICE 1991), under three headings: flexibility; clarity and simplicity; and stimulus to good management. Flexibility refers to the attempt to make the contract suited to use on all projects, providing a familiar standard form which will be available whatever the distribution of risk or method for payment used. The benefits of having such a recognisable form are obvious. Unfortunately, since the contract is new and very different to the presently used forms, these benefits cannot be immediately realised. Clarity and simplicity refer to both the structure and language of the documents. The usual cross-referencing has been abandoned and the contract has been designed with the use of flow charts to facilitate the simplest possible structure. The contract contains many less words than the traditional forms. The language is not the usual painful 'legalese' popularly associated with such documents and has been commended by the Plain English Campaign. It is the language, however, that is the most controversial feature of the contract, suspicion that it is not legally sound has been voiced both by lawyers and other members of the industry. On the intention that the document should act as a stimulus to good management, the guidance notes have this to say: "Every procedure has been designed so that its implementation should contribute rather than detract from the effectiveness of management of the work by the parties. This aspect of NEC is founded upon the proposition that foresighted, co-operative management of the interactions between the parties can shrink the risks inherent in construction work. Developments in project management techniques and their implementation over the last 20 years have moved faster than the evolution of traditional forms of contract. With the NEC, it is now possible to build arrangements for the different parties to contribute to the management of a project upon 38
improved practices and to motivate all of the parties by means of the contract to apply such practices to their work." (ICE 1991) By and large, lawyers have not given the NEC a warm welcome, though of course there are some who have done so. In this section, I review four papers by lawyers in order to establish the particular lines of reasoning which seems to underlie the unreceptive attitude. I detect two: the first, is to do with lawyer’s role as interpreters of contracts; the second is to do with their role as designers of contracts. Lawyers will naturally see a construction project in terms of the law. A contract like the NEC, which attempts to move the emphasis away from legal concerns towards task oriented ones will nevertheless tend to be seen by lawyers in legal terms. Their role is to anticipate what might happen if the parties to the contract enter into dispute: to anticipate how contractual provisions would be interpreted in a court of law. As members of a legal culture, they are employed precisely for the insight this membership gives them. The direction construction contracts have conventionally developed has consisted in designing clauses which will defend the interests of one party or another, in the event of dispute. This has the effect of reducing lengthy and costly disputes, since the outcomes of disputes which arise are foretold in the contract clauses and the body of case law. One of the motivations behind the NEC contract, is also to reduce the occurrence of disputes. However, under the NEC, possible sources of dispute are treated in a fundamentally different way from that in more conventional contracts. Under the NEC, the intention is that disputes are to be avoided through good management practice, which reduces the incidence of conflict. While it does address the matter of how disputes may be equably resolved, its primary emphasis is on how they can be avoided. The aim is transparency: bad management is to be penalised, not insured against.
39
The idea that good management practice can be promoted through the wording of a contract is not addressed by Jones (1995). On the question of whether a contract can promote co-operative relations, he cites four factors which he considers to be of greater importance; ironically, one of these is “inadequate management by contractor.” (ibid. 13.2, our emphasis) Inasmuch as he considers the role of contract wording, in this context, he does so with regard to “‘atmospheric’ conditions.” (ibid. 13.5) such as those calling for “all parties to deal fairly with each other.” (ibid.) He objects to such a condition on the grounds of vagueness. Referring, however, to the existence of “a real mood for change away from conflict...” (ibid. 13.3). He observes that a new contract might have a psychological effect “even if the detailed working was much the same.” (ibid.) Cornes (1995) recognises the intention of the NEC to promote good management practice, but makes little comment upon it. He expresses deep concern over the novel wording of the contract and the absence of existing case law to assist interpretation. Cornes also asserts that the NEC is harder to understand than the legal wording of more conventional contracts. This view has much support among members of the industry. One detractor of the NEC asked, "Why change wording which everybody understands?" Though actually, opinion seems to be fairly equally divided as to whether or not the language of NEC is better or worse than that of existing contracts. Secondly, lawyers are themselves in the process of contract design. Perceived problems in the industry and the possibility that they can be alleviated by better contracts, leads them to propose their own solutions. In these discussions, the NEC is often used as a foil against which their own proposals may be made.
40
Thus Uff (1995) in arguing for a more radical approach to handling the time element in contracts, comments that: "Just as no available standard form contract available offers fixed performance periods, so all such contracts, including the NEC, deal with delays in the same conventional manner, in terms of "excusable" and "compansable" (sic) delays, and extensions of time. The language may change, but the effect does not." (ibid. p5-6) Uff's paper is unique among the four considered here, for its attention to management issues. Craig (1994) while also proposing innovations in contract design, is again concerned with the consequences of disputes. The first part of his paper compares three features of the NEC and Electrical Contractors’ Association contracts: completion; retention; and security of payment for contractors. In the second part, he makes some suggestions of his own: that the aims and objectives of clauses should be drafted into contracts. The first half of the paper is concerned with facilitating the settlement of disputes; the second with proposals regarding the design of contracts.
Adapting to the NEC. The objections of lawyers are not the only obstacles to the successful adoption of the NEC. Another is conservatism, people simply like to go on doing things in the same way and no innovation, however beneficial it may be, will be welcomed unreservedly. There are, moreover, practical reasons for this, as one member, working under an NEC contract remarked of his colleagues: “If they get problems, they run back to what they’ve been doing for 20 years... You’ve got hundreds, if not thousands of years experience between them... people don’t forget what they’ve learnt over the past twenty years, not easily anyway.” 41
In introducing new ways of working, the contract is, in effect, asking people to abandon their hard won and hitherto valuable experience. In addition, people may not recognise the NEC diagnosis of and/or its solution to the problem. they may be satisfied with the way things are. Many contractors, for instance, assert that they are not claims conscious and that their planning is efficient. Thus, when it is suggested that the NEC has the effect of reducing conflict and improving planning, they are not impressed. Many see problems, but doubt that the NEC provides a solution to them. Contractors will point to the low profit margins in the industry as the chief source of problems. Others mistrust the motives of those promoting the contract, seeing it as an attempt by client’s to 'get one over' on the contractor, for instance. Conversely, some clients mistrust the contract, fearing that it does not properly represent their views. In a previous section, I have dealt, at length with the distributive viewpoint It would be wrong, however, to suppose that it is only from the distributive viewpoint that reservations are expressed. What follows is taken from an interview with a contractor’s commercial manager (CM) who professes a strong integretive motivation. [CM] One of the reasons why I've been round here is because I get on with clients. It's absolutely paramount [] there's always a solution to every problem, and so many people ... fight with the client and then, that isn't the best. The best contracts are where there's co-operation, good work, and hopefully some sort of margin. I know that's very difficult at the moment to make any profit [] in this current client, but certainly not losing money, at the end of the day [...]
42
[CM] So, and working with a client, rather than against a client, really is, is beneficial to both parties and to the contract. [I] I mean this is [CM] And I know that's the thinking behind the NEC... but it [I] I mean I can tell you don't like the NEC, because you're, you know, its er... [CM] I don't like it because.. on this contract you've got too many staff thrown together... and I doubt if any of them have ever worked on the NEC. If they get problems they run back to what they've been doing for 20 years. They get a problem, then they think, 'Oh it's ICE, what we gonna do.' The other side, there shouldn't be sides, cos its one team to build this contract, but there are two sides and you find that they've got, there is an engineer mentality [...] don't wanna know. [...] The other big drawback and I don't know whether its unique to this contract, or not cos as I say it’s the first NEC I’ve been involved in, but, in traditional ICE, GCE Works1, whatever it is, if you get a problem, you, you get problems everyday [...] you solve the problem and the financial implications are are put to one side and you come back to them. So you're getting on with the, you're getting the progress, you're achieving what everybody wants and that's getting the works forward, but here, under the NEC, that doesn't happen. If you have a problem, you have to do that and you've also got to think about the financial implications as well, and deal with them at that stage, and that is very, very time consuming 43
[...] [I] You see the theory as I understand it is you get the resources up front right at the beginning of the contract and through the contract, then you save at the end of the contract, because you're not left with all the outstandinq.. issues [CM] You see, if people do their job properly under ICE there's no need for things to run on, if everybody's on the ball. The ICE conditions contain the word reasonable I think 54 times, it says, 'the engineer shall be reasonable'. I think on two occasions it says 'he shall not be unreasonable.' But it's a contract about people being reasonable, if people are reasonable and work together and acknowledge that problems will occur and they need dealing with and maybe there's a different way of tackling it. If that's dealt with at the time, with co-operation between parties, it shouldn't run on
Ethnography, Generalisation and Theory. The ethnography above, like all such descriptions of social life, is constructed out of a series of generalisations. I have derived these partly from my own observations and partly from generalisations offered to me by members of the industry. In undertaking both these activities, I am employing methods which is used by every member of society to understand and describe their social world. In employing descriptions supplied by other members, I am taking the topic of my study and employing it as a resource of that study. It is such facts as these, which Sacks intends to render strange by his metaphor of the commentary machine. As Wieder (1974) fully explicates, they confer upon ethnographic accounts a logical status equivalent to the activities of those they describe. It is true that at one point I have used a more formal method of generalisation, Weber's ideal typical method.
44
However, these generalisations are achieved in the same manner as the rest. As Schutz (1973) has shown, all member's generalising activity can be conceived of in terms of the ideal typical method. Such generalisations, though entirely adequate to the purposes for which they are usually employed, must be treated with caution if it is our intention to pursue a scientific study. They have inbuilt features which are generally recognised and allowed for by members in their use of them, but which tend to be obscured by sociologists' claims to the possession of a scientific method. One of these features is the indexical nature of language already highlighted in chapter one as the etc. assumption. Members regularly conduct their communication on the basis that intended meanings will be heard as qualified by all sorts of unstated assumptions. Thus, the statements of members (including ethnographers) do not provide an exhaustive typification of the significance of categories like engineer and quantity surveyor. For example, one commercial manager told me: “the other side have... the engineer’s mentality...they don’t want to know.” This is typical of the kind of complaint which is often heard. Engineer’s mentality cannot be regarded as a consistent feature of a group, however. Because, as the same manager also observed: “...that does happen, that clearly does, where you can’t get an engineer to make a decision and so nothing happens ... and that’s obviously what we want to get away from. But that comes from an engineer who shouldn’t be in that position...” Another feature is the problem of induction. How many instance must one witness before one can say that a generalisation has been proved? This problem of logic, has been fully explored for the purposes of scientific study by Popper
45
(1959) who rejects the inductive method in favour of the falsification. This method, of developing only hypotheses which are capable of being proved false, whatever its utility in scientific research is not one of the primary methods with which members conduct the day to day business of learning about the social world. Nor is it one which is practically available to the ethnographer. Again, as Garfinkel (1984) has shown, anyone who conducts their social interaction according to the principle of scientific scepticism invites, at best a creative response, at worst an angry one. Instead, ethnographers and other members use the documentary method to make sense of the social world. This method of seeing events as instances of an overall pattern is again sufficient to its purposes, but does not confer upon its findings any special scientific warrant. A third feature is the moral nature of language. Classifying people for the purpose of generalisation, or indeed for any purpose, involves the use of categories which carry with them implied moral judgements. Despite the best efforts of the ethnographer to neutralise these implications, success can never be assured. It is worth remarking that an earlier published version of this chapter (Rooke & Seymour 1995) was described by one reader as 'lawyer bashing'.
46
3. Finding Describable Order in Member's Talk About Construction Contracts, by Using Membership Categorisation Devices. In this chapter, I intend to demonstrate the applicability of the MCD to the analysis of the transcript in Appendix Two. This transcript is taken from a research project meeting where the design of a questionnaire survey is being discussed. David S. has a background in sociology and is based in a school of civil engineering. Mick and David R. are based in a school of architecture and built environment, their backgrounds are in quantity surveying. The passage begins with a question from David S (lines 1-3). Who will the questionnaire be directed at? If what follows this question can be regarded as both an answer to the question and as an MCD, then the question itself stands as a name, definition, or brief description of the MCD. I shall demonstrate that the material can be so regarded and that the question does stand in some such relation to it. I shall go on to analyse how the material has been created, not only as an answer to the question, but as an answer which stands in testimony to the objective reality of the relationships and identities to which it refers. Finally, I shall explore some aspects of the significance of this objectivity for members attempting to understand the population so described. Before entering upon this analysis, it may be wise to recall a point already made in chapter one and to dwell for a while on its implications. Sacks' MCD apparatus represents methods used by members to make sense of the social world they inhabit. One implication of this is that the rules examined here are primarily hearers' rules. In using these rules, the sociological analyst is acting as a member. The point is that the effect of the rules is to make noticeable, features of the behaviour of others. Thus, in allowing that the economy rule provides for one category 47
application to be sufficient for one member, it is not suggested that it has a mechanistic effect, in determining that only one categorisation will be used. Rather, it does two things. First, it provides that one categorisation will be found acceptable - to have applied one category is to have sufficiently done the job of categorising. Second, it makes it noticeable when more than one category is used. This permits the search for a further reason why this should be so. Similarly, the consistency rule provides that when subsequent categorises from the same device are used, this is an acceptable way of doing categorisation; and it is noticeable when a different device is used to do further categorisation. This does not exhaust the significance of these rules, however. To be viable as hearer's rules, they must also be rules to which speakers orient; I will be attempting to draw out the implications of this in what follows. It is not suggested that members explicitly (either consciously. or unconsciously) go through the kind of logical computation presented here, or in Sacks' own work, in carrying out their sense making activities. It is rather that the analysis makes explicit the implicit logic of these activities.
An analysis of data, using Sacks' MCD apparatus. First, I wish to establish the applicability of the MCD apparatus to the analysis of this material. In doing so, it will be necessary to adapt many of Sacks' formulations to make them suitable for the purpose in hand. Sacks defines an MCD as a: "collection of membership categories, containing at least a category, that may be applied to some population, containing at least a Member, so as to provide, by the use of some rules of application, for the pairing of at least a population Member and a categorization device member. A device is then a collection plus rules of application" (Sacks1972a: 32, the use of capital 48
and small letter Ms is to distinguish between Members of the population being categorised and members of the collection of categories). The rules by which categories can be paired to population Members vary in their degree of generality. Sacks offers some explicit, general rules of application, but his analysis also draws upon rules which are tied to specific categories, or collections of categories. For instance: the economy rule governs the use of categories in any device; the modified repeatable use rule, can only governs the use of categories in a Pn2 adequate device; the rule, ‘may be applied to indicate those who provide drawings of the proposed works on a civil engineering project,’ governs the use of a very small class of categories, of which engineer, designer and consultant constitute the chief, if not the sole members. I will commence by considering how the general rules which Sacks provides allow for an analysis of the data; I shall then look for more specific rules of application in the data itself. (A list of the rules and definitions which constitute the MCD apparatus, as presented in Sacks1974 is presented, in tabular form, in appendix one.)
The economy rule The general rules which Sacks provides differ in the degree of their generality. I will begin with the only rule which has a completely general application. This is the economy rule, that categorisation may be complete with the application of a single category. In this weak form, the rule is of little use for the present purpose, but a stronger variant, in the form of a preference rule was later produced (Sacks & Schegloff????). This states:
49
"On occasions when reference is to be done it should preferredly be done with a single reference form." (ibid. p16) Reading the data with this rule in mind, it is apparent that, on certain occasions, a single category has not been considered sufficient for categorisation; this indicates the operation of at least one other rule in the production of those categories. If the question (line 1) names a population - those to whom the questionnaire would be directed - and the categories are offered in response to that question, then the question may be seen as organising those collections of categories. Thus, any use of categories, while satisfying the preference for minimisation whenever possible, must simultaneously satisfy the constraint of providing an answer to this question. This may be regarded as a variant of the preference for recipient design (ibid.). The initial question does not only name the population to be categorised, but gives two examples of the kinds of categories which would be constituents of an adequate response (lines 2-3). These categories are presented in a three part list, the final part of the list being an open category, showing the list to be incomplete (Jefferson1990). In this case, the list can be understood as a restatement of the question; and the open category, 'like who,' as a request for the list to be extended. David R., responds with a category, “management teams,” (line 3) which does not extend the proposed list, but defines it. That is to say that the category has an inclusive and an exclusive function. It is more general than those categories proffered by David S. and encompasses them. It excludes from the list all categories which it does not encompass. In the terms of this analysis, the category management teams, names a membership categorisation device. The categories traded in the first conversational pair, may be represented as:
50
population this questionnaire would be directed at = management teams
architcts
quantity surveyors
like who?
Again, we have in line 12: We're looking at architects, architects stroke designers This follows a discussion of a dichotomy which has been set up between architects and quantity surveyors. Here, Mick is expressing dissatisfaction with the category architects. What the phrase 'architects stroke designers' does is substitute a more general category on one side of the dichotomy. Confirmation of this can be found in lines 14-16, where David R. takes up the substituted category and demonstrates its generality by providing two other categories which are subordinate to it. We are therefore dealing with a hierarchy of generality, which is represented below:
design team/ designers
architect
services engineer
chemical engineer
whatever
The two sets of categories may be combined, by substituting designers for architects, as Mick does in the transcript:
51
population this questionnaire would be directed at = managemnt teams
design teams architect
quantity surveyors
like who?
whatever
services
chemical
engineer
engineer
The construction of the chart is made possible and is solely expressive of the double categorisations in the transcript. These double categorisations become analytically noticeable as a relaxation of the preference for minimal reference rule (economy rule). Conversely, the relaxation of the rule is a recognition and display of the hierarchy of generality shown in the chart. A rule is operating which is something like this: Show the location of each category in relation to other categories. Thus, in the first example, by re-naming the category of people 'this questionnaire would be directed at', David R. locates it in relation to the category of people whom the questionnaire would not be directed at (operatives and senior management, for instance). I have noted that the top level of this hierarchy (the total population) represents the name of an MCD. However, it is also true that 'design team,' on the second level, can be seen as the name of an MCD. In fact, given that an MCD may contain only one category, the analysis allows for the complete interchangeability of categories and collections of categories; any category may be regarded as an MCD in its own right. Whether it is so regarded in the analysis is dependent on
52
its function in members activities at any particular time. Thus, the category 'management teams' has been taken as naming the MCD in operation, because it is this category which names the population to be categorised. There are two other examples of double categorisation in the transcript. The first, includes some categories already mentioned, the categorisation of the varieties of engineer. That this is a double categorisation is not made explicit, however, until lines 33-36: And it is those people we are concerned with, so project architect, er, clients.. quantity surveyor, engineer, whether.. engineers, I should say, because you've got civils, structural, mechanical, electrical and services Any of the categories, 'civils', 'structural' etc. may be referred to simply as engineer. Secondly, the dichotomy which is referred to in line 78 as 'the basic them and us': the client's representatives and the contractor's representatives. This categorisation makes problematic the representation of the MCD as a simple hierarchy of generalisation. It is first introduced in lines 35-38 as the culmination of a passage which is the most category rich in the whole transcript. The passage may be represented thus:
people with authority to do things project architect
quantity surveyor
engineer(s) contractor's representatives
contract manager
quantity surveyor
53
The category 'quantity surveyor' appears twice in this diagram, demonstrating that the category 'contractor's representative' is not a more general one. If, however, the diagram were to be arranged with 'quantity surveyor' above 'contractor's representative', this would not do either; it is clear from Mick's statement that 'quantity surveyor' is part of the class of 'contractor's representatives'. The relationship between the two categories can be represented as a four square grid:
client
contractor
quantity
client's
contractor's
surveyor
quantity
quantity
surveyor
surveyor
designer
contract
other
manager
Another principle is needed in order to represent the categories as a hierarchy. Such a principle can be derived from the material: the division between client and contractor is 'basic', and the other categories should be taken as sub-divisions (lines 72-74). Taking these two principles - generality and the basic nature of the client contractor dichotomy and tidying up a little by omitting the open categories and the various types of engineer, it is possible to provide a general chart for the organisation of construction projects:
54
population this questionnaire would be directed at - management teams
contractor
client
design
quantity
contract
planning
team
surveyors
manager
engineer
architect
engineers
quantity surveyors
Sacks (1972) argues that the possibility of double categorisation presents a problem for sociological description: in that any particular categorisation cannot be warranted as adequate simply by ascertaining that that categorisation is an accurate one. It is apparent from the above analysis that such double categorisation can also provide a resource for sociological description.
The Consistency Rule. The consistency rule is not as general as the economy rule, being applicable only to Pn-adequate devices. It also presents some difficulties in interpretation. The first difficulty concerns the nature of a Pn-adequate device. Sacks defines two types of Pn adequate device, a Pn1 device: "can categorize each Member of the population N in such a way that one does not get for any Member of the population the pairing (population Member + no category member)" (ibid. p33-4) Pn2 devices are:
55
"(A) composed of programmatically relevant categories and (B) have for each category, either (a) some proper number of incumbents or (b) some proper minimum number of incumbents" (ibid. p39) A programmatically relevant category is one, the non-incumbency of which can be proposed as a fact. This is to say that the pairing (category member + no population Member) is in some way noticeable. B adds a normative dimension to this noticability. Sacks does not however supply an explicit general definition of a Pn adequate device. This makes it difficult to ascertain when, in fact, the consistency rule is applicable. He does state that the two definitions are linked by a modification of the repeatable use rule. This states: "The application of a Pn1-adequate device to a population N involves determining for each Member of the population, in any sequence, which category is appropriate, where any category may be repeatably used." (ibid. p36) The Pn2-adequate device is Pn adequate by virtue of the following modification: "any device as specified above (A, B), may categorize any population N, where the device's categories are treated as a unit with reference to repeatable usage. The population then is construed in terms of the number of complete or incomplete units that its Members may be partitioned into." (ibid. p39) It is not clear what the effects of this modification amount to. Does the exclusion of the pairing (population Member + no category member) still apply? If it does, this excludes many kinds of team, which otherwise would seem to typify the Pn2 device. Thus, for instance, a football team, while its categories might be normatively programmatic (i.e. it would fulfil the definitions of Pn2-adequacy, if
56
this did not encompass the definition of Pn1-adequate) would not be Pn1-adequate to defining any population. If, on the other hand, the exclusion is dropped, it is difficult to see what Pn adequate devices might have in common, other than the applicability of the consistency rule. The consistency rule itself might help settle the problem, unfortunately, it too is ambiguous. It states: “If some population is being categorised, and if a category from some device’s collection has been used to categorises a first Member of the population, then that category or other categories of the same collection may be used to categorise further Members of the population.” (Sacks 1972, p33) But is this, in fact a rule? Sacks admits, the rule is “weak,” in that “it may exclude no category of any device” (1974, p219). The question here is, what is the force of a rule that excludes nothing? We may say that the nature of a rule is that it tells us what to do; it belongs to a class of utterances which we might call instructions. There are other types of utterance in this class, orders and advice are two others. The difference between these types and a rule is that a rule is general. (Thus, a prototype rule might be, if x, then do y - the instruction is preceded by a set of conditions which are to be met before the instruction becomes operative.) An instruction is a kind of indication: it points out to us what we should do. But to point towards something is to point away from something else. The key word here, emphasised by Sacks, is may.
Clearly, this usage is intended
by Sacks to indicate that the rule is somehow not deterministic. That is to say, it allows for the possibility that categorisation will not proceed using the same device. But can the word ‘may’ have any part in a rule? Let us look more closely at this. One way in which the word ‘may’ might be used to produce a non-deterministic rule, is if it is used to indicate that the rule is a preference rule. 57
Garfinkel distinguishes between constitutive (basic) and preference rules in the following manner: “The possibilities that preference rules can deal with are provided by basic rules. For example, that any of the pawns or knights but no other pieces may be moved on White’s first move in chess is provided by the basic rules of chess. But which among these alternatives the player can choose is a matter of the player’s election.” (1963 p192) In this interpretation, ‘may’ is taken to indicate that a category from the same device should be preferred to one from another device. However, Garfinkel’s analysis implies that formulations of such rules should be categorical. The element of indeterminacy is introduced because members have a choice whether or not to orient towards particular preference rules. To formulate a preference rule in the manner I have suggested produces a problem. The question arises, how would a speaker orient towards such a rule? It would mean this: ‘if one category from a device has been used, prefer to use another from the same device for a subsequent categorisation (if x, then prefer y).’ But how does one prefer a course of action, except by doing it? Unless it is specified under what conditions one should not do y, how else could one follow such a rule? A possible way out of this is to formulate the rule as a categorical preference rule, i.e. If some population is being categorised, and if a category from some device’s collection has been used to categorises a first Member of the population, then use that category or other categories of the same collection to categorise further Members of the population. This certainly seems to be a viable rule, but there is no reason why it should be seen to hold only for Pn adequate devices. Its application would seem to be entirely general. I shall refer to this interpretation as the consistency preference rule. 58
Another way of understanding Sacks’ formulation, is that it is probabilistic (if x, then y is more probable). This is certainly viable, though it is not a rule at all, but an analyst’s empirical prediction. It would allow a hearer to predict (though not precisely) subsequent categories from a first, but it offers no guidance to speakers. This ability to predict, is what makes the consistency rule corollary work as a hearer’s maxim. This hearer’s maxim tells us: “if two or more categories are used to categorise two or more members of some population, and those categories can be heard as categories from the same collection, then: hear them that way.” (ibid. p219-220) This leaves us with the question, what rule is it that the speaker orients to, which makes possible the analysts prediction?
A possible formulation would be this: if
you have used a category to label a member of a population, do not subsequently use a category from a different collection, unless it is clear that the subsequent category does come from a different collection. This would go some way to making the analysts prediction workable and it seems reasonable that speakers might orient to such a rule; since, given a hearer’s propensity to use the analysts prediction, it would avoid confusion. But it would not necessarily provide for the use of consecutive categories from the same collection: it may simply be that speakers consistently made clear that a second category belonged to a different collection to a first category. The phenomenon can be accounted for by the consistency preference rule. Comment [jr1]: Page: 58
Another job which the word ‘may,’ might do in a rule is to grant permission. Superficially, to grant permission is to indicate a course of action without pointing away from other courses of action. To grant permission, is not to issue an instruction. However, granting permission is something we may only do against a background of restraint. It only makes sense to grant permission, if permission can be withheld. Under this interpretation, what is ruled out is the possibility 59
that, having used one category from a particular collection, one would not be allowed to use another category from the same collection. This interpretation has the advantage of being applicable only to Pn1-adequate devices (including Pn2 devices, if the full Pn1 definition, is deemed to be included in the Pn2 definition). Only a Pn1-adequate device assures the user that an appropriate category will be available for subsequent categorisations. This interpretation may be worded thus: If some population is being categorised and if a category from some device's collection has been used to categorise a first member of the population, then that category, or another is always available to be used to categorise any further Members of the population, such that a pairing (category member + population Member) is always achievable. I shall refer to this as the consistency Pn-adequate rule. As we shall see, the consistency Pn-adequate rule plays an important part in Sacks' argument. It does present problems, however, if used alone. For instance, it is difficult to see how it can fully provide for the workability of the hearer's maxim and it is noticeable that Sacks makes no reference to Pn1-adequacy when he introduces this idea (Sacks 1974). A further difficulty is considered below. Thus, we have two workable interpretations of the consistency rule: as a categorical preference rule; or as a permission granting rule. If the former interpretation is to be adopted, it is available for devices which do not exclude the possible pairing (population Member + no category member). However, as the rule is seemingly applicable to any device, the categorisation of a device as Pnadequate becomes meaningless. If we adapt the latter interpretation, it becomes applicable to Pn2 devices, if they are regarded as a special type of Pn1 device. The rule retains its exclusive applicability, but this means that many normatively programmatic devices cannot be considered as Pn2 devices.
60
Applying the Consistency Rules In this section, I will explore the consequences of applying each of the interpretations of the consistency rule discussed above. The consistency Pn1-adequate rule seems to do two jobs in Sacks' argument. First, the existence of more than one Pn1-adequate device poses a problem for sociological description. Given that one categorisation may be sufficient to categorise a member and given the existence of at least two Pn1-adequate devices with mutually exclusive categories, a decision must always be made as to the appropriateness of any particular categorisation, which is not dependent upon the accuracy of that categorisation. The Pn1-adequacy of these two devices necessitates that this should be so, not on some occasions, but on every occasion when categorisation is done. This directs our attention to the methods which members use to select categorisation devices. However, although Sacks uses the Pn1-adequacy of the devices 'sex' and 'age' to demonstrate the logical necessity of such decisions, it is not the case that only Pn1-adequate devices require such decisions to be made in practice. Second, in the specific case of a potential suicide telephoning a suicide prevention centre, the Pn1-adequacy of the device R helps to account for the fact that callers often use this device to categorise the person answering the phone. Here, though, it seems that the use of the consistency preference rule strengthens Sacks' argument further. Using only the consistency Pn1-adequate rule, the argument consists of: (1) the caller has commenced and conducted a search for help using the collection of categories belonging to device R; (2) since a category belonging this collection is available for categorising every member of the population, the caller continues to use device R. The account would be considerably strengthened by (3) the caller has, on the occasions where categorisation continues in this manner, chosen to follow the consistency preference rule. 61
There is no necessity to choose between the two interpretations of the consistency rule. However, the consistency Pn1-adequate rule is only applicable to a limited number of devices. The device 'management teams', clearly does not qualify as such a device. The analysis can be easily reconstructed, however, to render such a device, by adapting its collection to the general form of collection R. The first feature to note about collection R is that it is constructed in terms of rights and obligations. The second is that its Pn-adequacy depends upon its collection of categories containing at least one category which is applicable to people who might be expected to have no rights or obligations regarding the suicides search for help: the category of stranger. We might refer to such categories as 'null categories'. (A third feature - which will later become relevant - that its collection of categories is organised into relational pairs, is of no consequence at this point in the analysis.) Thus, following the form of Sacks' definition, I propose a collection M. Any category is a member of collection M if that category constitutes a locus for a set of rights and obligations concerning the management of construction projects. Such categories as general labourer and member of the public belong to collection M as null categories. The collection, 'management team', is a sub-set of collection M. This collection can now be used to make noticeable, the introduction of any other collection, such as age, sex, etc. Unfortunately, the materials used here contain no example of this. The existence of other Pn1-adequate devices does, however pose a question. Why is the device used here the appropriate one? Sacks' analysis supplies an answer, he observes: "In focussing on collection R, it might be asked; How is it that when a search for help is being engaged in, the relevance of collection R is provided for? The formulation is, however, erroneous. For as we shall 62
observe, the very propriety of a search for help is provided for by the rights and obligations that are organized by reference to collection R. Our initial task is then to show how features of collection R provide for the propriety of occurrence of a search for help." (ibid. p40) The possibility of designing a questionnaire survey to investigate management attitudes is dependent upon the existence of some such collection as M. The logic of designing such a questionnaire is as follows. The management of construction projects involves the exercise of rights and responsibilities. These rights and responsibilities are differentially distributed among various categories of manager. Certain of these rights and responsibilities depend upon the possession of specialised knowledge. Since each category of manager has different rights and responsibilities and has undergone a different typical education, training and career path, it may be possible to identify differences between these categories of manager over and above those of rights, responsibilities and specialised knowledge. Thus, it is the rights and responsibilities which are organised by reference to collection M, which provide for the possibility and rational nature of the enterprise of designing the survey. This is as far as collection M will take us. I now wish to propose that the transcript may be more perspicaciously analysed by assuming that two non-Pnadequate devices are in evidence. The first of these, is based on collection P: a collection of categories which denote a specialised competence with regard to the management of a construction project. The second, collection C, consists of two categories which act as a locus for a set of rights and responsibilities with regard to the management of a construction project, derived from a contractual agreement. We are finally in a position to return to the transcript. Lines 1-16 proceed with the use of collection P. Collection C is first introduced in lines 16-19: 63
The important thing is that they are interpreting the client's brief. [] So to a certain extent they've had [] to define what the brief comes out as, and what the actual design is. It is apparent that the category 'client' has a different logical status to those used in the preceding talk: that is to say it has a different use. Up to this point the categorisation has proceeded along the lines of: there's this kind of person and that kind of person. Here, the relationship between 'the client' and the categories in the collection 'design team' is being specified: the design team has rights and responsibilities with regard to interpreting the client's brief. Subsequently, the categories of collection P so far produced and others which are added at various points, are divided between the categories client and contractor. It can now be seen that the double categorisations discussed earlier represent, not merely a departure from the economy rule, but also from the consistency preference rule. Analysing the transcript in this way yields two charts. First, collection P, again organised as a hierarchy of generalisation: collection P
quantity surveyor
design team
project
contract manager
engineers
architect
The second chart is a very simple one: collection C
contractor
client
64
It is the combined application of these two collections which gives the chart at the end of the section on the economy rule. The form of these collections, which is borrowed from Sacks' formulations, raises a further possibility: that this way of formulating collections itself has a general application. In the next section I shall examine the transcript contained in Appendix Three, in order to test this possibility and to further extend the applicability of the economy and consistency preference rules.
Doing an Interview. Given the consistency preference rule, it is noticeable that the first 12 lines of the transcript (Appendix Three) contain categories from a large number of collections. This can be regarded as an attempt to establish which collection is applicable: the QS is attempting to find out which collection he should talk to. This problem is made explicit in 2-4: [I] Er... right... what else? Whose the project manager? [QS] The project manager being the actual project manager? [I] Well yeh, under the contract. It is not clear which collection the category 'project manager' belongs to. As the interviewer, I specify a collection. Although this collection is organised according to rights and responsibilities derived from a contractual agreement, it is not collection C, which is organised only according to the broad distinction client/contractor. The collection introduced here, I shall call it collection R1, consists of specific roles defined under the NEC contract. It involves the introduction of a category produced and then discarded in the research meeting transcript and ignored in my analysis of it: 'project manager'.
65
My specification of an MCD enables the QS to answer. The answer to such a request for identity involves further categorisation. He supplies three further categories: company name, profession and location. Here, the economy rule is relaxed also. This is to accommodate preference for recognition (Sacks & Schegloff 1979). The transcript is insufficiently detailed to record try markers, but this is unnecessary for the analysis here. Whether the categories are presented in the manner recorded by Sacks and Schegloff, or as a straightforward three part list is immaterial. The latter possibility arises because of the unfamiliarity of the conversational participants. In Sacks' and Schegloff's' material the speakers are obviously known to each other and refer back to previous talk between them. This transcript is taken from my first meeting with the QS; furthermore, as a researcher from Birmingham, I am not likely to be familiar with local companies. Thus it may be that in loading his answer with categories, the QS is providing for the maximum possible chance of recognition in one utterance. Name and place are Pn1 adequate devices and are therefore always available for any instance of categorisation, it is worth pausing to consider their are used here. First name: every person and company has a proper name, which is more or less unique to them. It is a feature of our culture that giving a proper name is usually an acceptable answer to the question 'who?'; though in circumstances such as those described here, this will usually be followed by further categorisation. As well as the convenience of this ubiquitous collection (its always being close at hand, as it were), it also facilitates further action on the part of the questioner. The practical uniqueness of the categories it contains provides a way of initiating communication with the categorised individual or company: attracting their attention by using their name, or looking them up in the phone book, for instance. It should be possible to formulate the tendency to use such favoured collections in terms of preference rules.
66
Place is less commonly used, though also Pn1-adequate. Its might also facilitate future communication, by indicating where they might be found. Its use may also be indicative of the mobile nature of construction work: unlike manufacturing, construction is not carried out on a companies premises and any project will involve companies and individual from a variety of locations. It is used again in line 7, this time to locate the individual architect involved and in lines 46-47, in connection with the structural engineers. The categorisation 'architects' may be taken as an application of collection P. My next question promotes the suggestion of a fourth collection (line 7): this one organised as to presence on site, may be called S. This collection has two categories: those who are permanently on site and those who are just visitors. Lines 6-8 represent a stabilisation in the application of collections: [I] And is he on site all the time: [QS] No he's just a site visitor, he comes from [location]. [I] Who is on site After the profusion of collections referred to in lines 2-6, we now have three lines in which only one collection is used. Having received an answer to line 6, I ask a further question employing the same collection. It is followed by a long answer by the QS, this seems to indicate that he has now solved the problem of which MCDs are appropriate. In the subsequent talk, he employs all the collections thus far raised. Below, I will examine how he deploys these as a solution to the problem of providing appropriate answers. First, though, I wish to examine his introduction of a further collection in lines 910: [QS] Who is on site is the three clerk of works, that's electrical, mechanical and er, building. 67
This is followed by a self-correction (lines 10-12): Or, at least that's what they're called under a normal contract, obviously it's supervisor's representatives, or project manager's representatives on site. Again, we see the answer to a request for identity, necessarily framed in terms of one collection and answered by the introduction of another. The collection thus introduced, I shall call it R2, as it is directly analogous to R1, represents specific roles defined under 'a normal contract'. The QS displays his translation of the 'normal contract' categories into the NEC ones. Clearly, there is some difficulty involved in this, as two alternative categorisations are offered. A tentative solution is offered in lines 14-15: I think he's actually called the supervisor, or the supervising officer. I mentioned above, the ubiquitousness, the close-at-handness, of the collection of proper names. A similar phenomenon is apparent here. The QS reaches for a familiar categorisation and then translates it into a newer, more correct one. He is literally learning a new language. Having categorised these individuals under R1/R2, he can tackle the further problem of answering the question by categorising them according to S (lines 1321): And they're, I wouldn't say they're full time, as you can see they're not here at the minute, but they have been here this morning and they're probably half a day every day. They do have other [client's name] jobs they have to er go to sites, but I don't think they have other site offices, to me they're the only offices they have, then they're just visiting other sites to see what []. All their administrative stuff is down here, so they obviously work from here [] electrical guy 'cos there's not a great deal occurring there, but
68
they're obviously down here. So they're the three guys that are on site permanently The categorisation, when finally arrived at, is definitive, 'they're the three guys that are on site permanently'. Yet a tremendous amount of work has been done to arrive at this formulation. Sacks has established that the existence of more than one Pn1-adequate device presents a problem for sociological description. Here we see another, analogous to those considered by Garfinkel (1984): even given the appropriateness of a particular collection, its application may be problematic. Nor is it that the collection may be applied wrongly. The categorisation is correct for all practical purposes, but its correctness depends upon a series of etc. assumptions encompassing all the work done in lines 13-21. In composing a questionnaire, or an ethnography, these details tend to be generalised out of the picture. A similar point can be made about the QS's self categorisation (lines 2126). In line 26, the QS makes explicit that he is using collection C, in order to organise his answer to line 8. The previous talk has centred on the client's representatives, now he is moving on to the contractor's team. Again, the unprompted use of this device indicates its close-at-handedness. Having completed the categorisation of those on site, the QS begins to categorise those off site. This is prompted by my expressed interest in talking to other people (lines 32-41). He begins with 'a contracts guy', a member of the contractor's organisation off site. Having dealt with the contractor's team, he then moves on to the engineers. These are grouped together, in the manner explicated in the examination of collection P in the section on the economy rule. A final feature to note about the categorisation in this transcript is who is left out. Subcontractors, craftsmen, labourer, for instance, are all categories which might be used to categorise the population on site. Not only have these categories not 69
been used, those members to whom they are applicable have not been categorised at all. They are taken to be excluded from the population. Clearly, some such collection as M, has been used to decide who-we-are-interested-in.
Producing lists in interviews. I observed above that a danger existed in my approach: that the MCDs thus explicated, could be taken as further evidence of the existence of social structures. I now intend to explain why I do not believe that this is the case. In order to do so, I will draw on Eglin's (1980) account of the production of lists in interviews. Eglin observes that: "the interview between ethnographer and informant does not take place in a social vacuum. Interactional work by the ethnographer is necessary to gain access to the interview setting. [...] The ethnographer is constrained to provide an account of his presence, identity and proposed activity. [...] the ethnographer's account of himself to the informant must include some version of what the interview is to be about. One inevitably, as an interactional necessity, structures the field prior to investigating it. The ethnographer's desire to avoid imposing an alien structure on the native domain remains an ideal only. Not that it is ever finally clear what the interview is or was 'all about'. Rather ethnographer and informant rely on each other to talk against a background of some version of 'what it is all about'. 'What it's all about' is an unspoken resource of the encounter" (ibid. p70) My interviewing technique is very different to the one employed by Eglin, it is largely unstructured, involving the use of questions only as a means of getting the informant talking, rather than as a set agenda which must be got through. One instruction which I give at the outset is that: 'I want to know what's on your mind,
70
what you think is important.' Also, the interviews were carried out with an entirely different research objective than the one pursued here: to discover the problems and attitudes of managers on construction projects and their receptiveness or antipathy to the NEC. Unlike Eglin, I was not attempting to elicit a list for the purposes of my analysis and the transcripts display the particular reasons why categorisation was done on that occasion. Nevertheless, a great deal of selectivity has been involved in presenting these particular transcripts. They were chosen for the profusion of categories contained in them. Thus, Eglin's comments remain valid, albeit in a post hoc fashion. In order to truly establish what part these categorisations play in members everyday lives, it will be necessary to extend the analysis to more randomly selected material. The first transcript is not of an interview at all, but of a research project meeting. However, this does nothing to mitigate Eglin's reservations. True, it has the virtue of allowing me to portray members at their normal work, but only at the expense of selecting members who are even more deeply involved in the research process than the interviewees. Eglin's remark about the indeterminacy and interactively achieved nature of 'what it is all about' is, I think, strongly borne out by the first few lines of the QS interview. Eglin's major point is that there is no evidence that these collections exist inside members heads (or even, in this case in society at large); rather, they can be seen to be produced in an occasioned manner, to answer a specific demand: “But what does the production of a list tell? That the informant is unpacking a cognitive arrangement in his head which requires only the right question for its elicitation? Or is it rather that the informant is exploiting his own ‘methodological’ ability to ‘do’ a list, since that is what I said I wanted?
71
Compare a shopping list. Would it be correct to infer that , because a member can construct a shopping list, the items on the list are mentally stored in some fashion corresponding to the structure of the list? I propose, rather, that lists are always and only produced for some purpose, their structure speaking to the occasion for which they are produced, at the same time as constituting that occasion.” (ibid. p71-2) This occasioned and occasion constituting nature can be seen in both transcripts. These collections are composed in response to questions. In the research project meeting transcript, it can be seen that considerable work has gone into the production of the list. It appears in dribs and drabs, partly in response to David S.'s prompting. This is despite the commitment that members display to the belief that such a list is an achievable and desirable phenomenon. In the QS interview, though, the list is more easily achieved, this is attributable to the fact that the informant is categorising members who appear as everyday features of his working life. In this case, however, the collections used to organise the list, although available, are not immediately evident as objectively applicable. Achieving this applicability takes interactional work. Although it has been possible to indicate some preferences among MCDs, these indications must remain tentative. All this reinforces Eglin's further point about the dangers of abstraction: "In abstracting from utterance such labels as "public health", "psychiatry" and "surgery", I (as semantic ethnographer) am not only ignoring the control of those items by the phrase "doctors connected with", but am failing to appreciate their status as not simply 'pre-coded entries on a memory drum' but as elements in a formulation" (ibid. p72) My recording of these collections as formal representations of members actions creates the possibility that they can be divorced from the context in which they 72
were produced and treated as objective, context free, descriptions of the organisations which are described in that talk. That this would be a mistake should be clear from the evident work involved in their production. To abstract the products of this work from the constraints and requirements of this productive activity would be to render them meaningless. It is true that the consummate achievement of the analytic direction taken by Sacks, the 'simplest systematics' (Sacks, Schegloff & Jefferson 1974) is both context free and context sensitive, but such an achievement is far in advance of what has been attempted here. The development of an equivalent machinery, if such is possible, would require a far more wide ranging analysis and be, in its final constitution, necessarily free of empirical content. Eglin's final point is that, having identified a an outline of the domain, the ethnographer adapts this outline as a principle to structure further research, often ignoring inconvenient findings. This reinforces the injunction recognised above: that future research must be based on more randomly selected materials, if it is to provide a fuller picture of the role which MCDs play in the mundane lives of members.
73
4. Doing Social Differentiation Using Moral Identities. Cuff shows that Sacks' relational pairs can be adapted to the study of problems of versions by supplementing them with moral identities. The standardised relational pair, in Sacks' analysis: "constitutes a locus for a set of rights an obligations concerning the activity of giving help." (Sacks 1972, p37) As used by Cuff and as I shall use it here, it will be simply treated as: A locus for a set of rights and responsibilities between two parties. Cuff notes that a moral import is already built in to this definition and goes on to argue that: "the question of moral adequacy is a central concern for tellers and hearers, as displayed by the way that moral assessments of the appropriateness of the actions of the parties involved can be seen to be built into the identifications used to make sense of what is told/heard to be happening." (Cuff 1994, p53) Cuff formulates these assessments as the attribution of moral identities. Such identities can be classified as either good or bad. In performing such a classification: "we are not saying necessarily that members actually employ the adjective 'good' or 'bad', but are simply suggesting as analysts, that these adjectives provide useful glosses for indicating in broad terms how they might be assessing and, thereby specifying the incumbents of the general categories" (ibid.) Towards the end of this chapter, I will engage in an analysis of versions using moral identities. First, however, I will demonstrate that they have a more general
74
utility in the analysis of the-phenomena-usually-treated-by-sociologists-as-socialstructure. Specifically, I wish to demonstrate how moral identities can be used to explicate the ways in which members make and display status difference. The bulk of this chapter is based on the material presented in Appendix Four. Later, I will reintroduce the transcript from Appendix Three, in order to examine the use of moral identities in that material. The Appendix Four transcript consists of three extracts from an interview with two managers in a contractor's organisation; both have engineering background and from this vantage point they are engaged in describing some aspects of three features of social order in the construction industry. The first is the relationship between client and contractor; the second, the role of quantity surveyors; the third, the relationship between contractor and engineering design consultant. This involves making social differences between the categories used. In these, moral identities can be seen to be used in three different ways, to produce status-laden accounts of social difference. Before looking at the way these descriptions are produced, in order not to seem to be pre-empting the argument, I will examine my own use of categories, showing how the categories I have used to describe these managers have been derived. I shall show how these categorisations can be derived from the text. I have used three categories for the interviewees: 'contractors', 'engineers', and 'managers'. To anyone with a knowledge of the industry, the opening lines (1-3) are sufficient to establish the category 'contractor': If you get to the end of a job and you've made a loss, you look at bloody everything to see if there are any commercial opportunities you've missed. However, this is insufficient for my purpose, as I am not addressing members of the industry. The question then arises, how do I know this and the answer cannot be given without invoking the categories I have learned in my research. This is 75
how the kind of circular argument described by Estor and Francis (unpublished manuscript) is set up. Further evidence is supplied in the story told in lines 42-50, which is told from a contractor's point of view. The point of the story is given in lines 48-50: Erm, and people are saying, tut, tut, tut, the avaricious contractors and their QSs, but what that meant was in fact that the tender prices were 30% too low. Clearly, contractors are portrayed as being unfairly accused of wrongdoing. This is not conclusive, however, because it is possible that a sympathetic client, or consultant, for instance, would take this point of view. Absolute confirmation is not to be found until lines 69-71: clients generally think that it is the contractors taking advantage, err picking up the money on the variations, whereas in reality, I think we would prefer a contract with no variations. Here, the use of we in line 70, is clearly indexical to the category contractors. Employing the documentary method, any reader of the text can now understand the preceding as well as the succeeding text as being an interview with contractors. Establishing the correctness of the categorisation engineer is simpler to establish. It is made explicit in line 17 that: we're both engineers The category 'managers' is more problematic to establish from the text. It has been used loosely here as a kind of neutral category, to avoid making pre-emptory categorisations. It can, with reference to chapter two, be read as shorthand for
76
'members of the population under study'. Nothing in the analysis hangs on a distinction between these two categorisations.
Using Moral Identities Moral identities may be crudely divided into three classes, good, bad and neutral. I shall look first at identifications which attribute good or bad qualities. As I observed above, these identifications can be used in several different ways. First, some categories of person are inherently more virtuous than others. Thus, in line 19, we have: [manager B] let's face it, quantity surveyors were only engineers clerks Here, we have a relational pair, quantity surveyors and engineers, with the correct relationship between the two categories specified: 'quantity surveyors were only engineers' clerks'. (leave aside for the moment that the statement is in the past tense.) The term 'only' specifies a low status for quantity surveyors. The term 'clerks' does a similar job in the context of a discussion of professional categories. The remark is made more pointed by the use of a possessive: that quantity surveyors are engineers' clerks definitely specifies the status relation between the two categories. Earlier in (lines 6-8) we have: It's all a good job creation scheme by quantity surveyors, erm, who are a growing infestation in the industry The alternate pair part on this occasion is 'the industry'. Thus, it is a wider category than 'engineers', encompassing all categories of the MCD other than quantity surveyors. The moral identity of quantity surveyors is even more clearly specified. They are not simply inferior to engineers, their influence is positively malign. It is not clear from this remark that there is anything that could be termed a correct relationship between quantity surveyors and an alternate pair part. Such 77
a usage is common enough, similar formulations are to be found, for example, in discussions of criminals, or in the nastier forms of racism. This is not to imply that these categories necessarily constitute master statuses (Hughes1971), it may be allowed that individuals may reform, thus ceasing to be criminals or quantity surveyors. In the event, it turns out that what we have here is a more complex account. It is, in fact, historical. A low status profession is increasing its influence, its previously correct relations, with engineers and with the industry as a whole, have been overturned. A further consideration of, the identification of historical periods will be given later in this section. A second use of moral identities, the use identified by Cuff, involves commenting upon the manner in which individuals conduct their incumbency of a particular category. Lines 5-13 in extract two provide an illustration: [manager A] We've both had cases where we have gone to a client and said hey you can save money if you do it differently, he's said, oh, thanks very much and issued the instruction to do it differently and kept the saving himself. That's happened to me [I] right [manager A] Usually you propose it on the basis of let's share the saving and some honourable clients do that. But these days I would tend to expect most clients to say that's a good idea, thank you very much and keep the savings themselves. A more complex example of this second use can be found in the passage considered previously (lines 3-14): [manager A] I mean what's traditionally happened is the client has not wanted to make a decision, the contractors have wanted to keep their 78
powder dry and see what they needed to ask, so that's why. Y'know, it's kept till the end of the end. [...] It's a great job creation scheme and we've had experience, we think we've got one surveyor in our organisation who's bloody good at this. An erm, no names, no pack drill, but his style appears to be just to put in any old interim account during the period of the job and [interviewer] yeh [manager A] the job loses money and er he pulls it round at the end. As sociologists are well aware, one way of giving an account of the social order is to do so ironically. As we see here, other members too use this device. I will first show how it can be seen as one possible solution to a problem of logic; then I will go on to examine the details of its use on this occasion, demonstrating how the irony is achieved. As we have seen, the category quantity surveyor is portrayed in wholly negative terms by the two managers whose talk is recorded here. There can be no such thing (in present times) as a good quantity surveyor. Yet there are still differences between quantity surveyors; some quantity surveyors do the things that quantity surveyors do with more efficiency or effectiveness than others. The question is, are these better quantity surveyors, or worse quantity surveyors? They are better at being bad. To say they are better may seem to be endorsing quantity surveyors values. To say they are worse, may seem to imply they are less efficient or effective than others. To say they are 'good', but to use the term ironically, is one solution to this problem. There are, of course, others. This solution is achieved in the following manner. In lines 1-6 a situation is described, it is not perhaps an ideal situation, but it is the-way-things-are. In lines 6-7, we have:
79
It's all a damned good job creation scheme by quantity surveyors Now 'a damned good job creation scheme' might be a very good thing indeed, if one holds a certain set of values about the-indignity-and-negative-social-effectsof-unemployment. But a job creation scheme by a particular section of the workforce could imply an entirely different judgement: it could imply the pursuit of a particular sectional interest. To put it in colloquial terms, the implication might be: these people are on the make. That this is the correct interpretation is revealed in the talk immediately following (lines 7-8): erm, who are a growing infestation in the industry There is no way in which a growing infestation can be regarded as a good thing. So when we get to (lines 9-10) we think we've got one surveyor in our organisation who's bloody good at this there can be in no doubt that, 'bloody good' is not to be taken literally. The following talk serves to reinforce this interpretation. A third use of moral identities that can be observed in the materials concerns the characterisation of relationships. There is a sense in which the uses described above can also be termed characterisations of relationships. In attributing a moral identity, one is describing how an individual, or category of individuals relates to others. This is more clearly brought into view, when we treat these categories as parts of relational pairs. Here, I want to point to a different phenomenon, however. Members may express views as to a relationship as a whole, encompassing both parts of the pair. Thus, marriage, lifelong friendship, gessellschaft, etc. may be attributed moral identities, characterised as good or bad things. This use of moral identities still involves reference to relational pairs, but 80
not in the same way. The locus of rights and responsibilities which defines the pairing has a different logical position. This can be most clearly seen if we review the two uses we have considered so far. If we begin with the second usage considered here, the one examined by Cuff, we see how the rights and responsibilities inherent in the pairing are used as a point of comparison with the actual behaviour of individuals. This is, in itself, different to the first usage, where the rights and responsibilities themselves are held up for examination. Thus, there are a set of rights and responsibilities between QSs and engineers which are the correct ones and a set which are in de facto operation, but are 'bitter and twisted' (line25). Indeed, it seems wrong to refer to these as rights and responsibilities at all, as these terms should be used to define de juste and not de facto terms. It is this contradiction which sets up the problem of how to characterise a 'good' quantity surveyor, discussed above. What is needed, for the purposes of this analysis, is a means of solving the problem other than the one used above, however. If the analysis is to conform to the principle of indifference, it is insufficient to affirm the position of the engineer, by ironicising that of the quantity surveyor. This is not because I wish, at stage to enter into an analysis of versions, something which can be given only the briefest attention in this analysis (though such an analysis would make the problem even more urgent). It is rather that it becomes difficult to analyse the role of the surveyor at all, without some such mechanism. Returning to the text, it appears that some such concept as role is in use. Thus (lines26-29): I mean it happens on big road jobs, the QSs look at the bill an look at the bill an drawings and seek to arrange the programme and arrange the method of working to exploit the client's problems. Y'know, they look at the way that the quantities of excavation and fill are em, are billed, by the 81
client, if they can spot mistakes there they will angle their programme to exploit those. This is simply what they do. We know from earlier talk that this method of working is wrong, but in the actual analysis of their activities, the manager has dropped any reference to value judgements. Thus, if QSs can hardly be described as having duties, they do at least have socially sanctioned ways of behaving, which they can be expected to conform to. We can assume that, in a similar way, they have expectations of how others will behave towards them, which are similarly fulfilled. It seems, then, correct to say that the relational pair is a locus for a set of reciprocal expectations, of which rights and responsibilities are but one kind. Given this redefinition, we can now go on to look at characterisations of relationships. The point is that, in the second usage, the locus of expectations is used as a reference point, by which to judge the behaviour of individuals; in the first, it is used to attribute relative status to two parts of the pairing; in the third, the expectations themselves are assessed for their moral worth. This can be most clearly seen from the comparison in extract three (lines 18-31). On the one hand, is posited a project under NEC conditions, where: there's an incentive there to, to be co-operative and to look in a pure engineering sense, y'know pure engineering is solving a problem at minimum cost. Erm, so it, y'know, this incentivise people to do this on both sides, which is good training for the guys coming up. The moral identity implied by 'co-operative' and 'pure engineering' can be seen, in the light of the passages already analysed to be good. This is reinforced by the remark that the contract is 'good training for the guys coming up'. The roles and relationships thus established are compared with those that will be engendered by:
82
the bitter and twisted claim training which they might have on a fifth edition Highways Agency job where the QSs are running the job. It can be seen that an important element in this comparison is the comparison between two relational pairs, each encompassing the same two categories, but different sets of expectations. In the NEC job, the expectation is that the engineer will be running the job and that engineer's attitudes will prevail, ICE job, it is the QSs and their attitudes that dominate. The way this comparison is managed, however, is quite complex. It is begun with the question (lines 18-19): what else does the NEC do? The NEC is, of course, a contract, but are different ways in which a contract can be viewed. It can be seen as simply a set of documents; alternately, an agreement between two parties; or as a project - the actual temporary organisation that is set up to carry out the construction work. As a set of documents, the NEC can do nothing, it has no agency; it simply embodies an agreement in an objective referable form. In this sense, then, it provides the rules of the temporary organisation, its articles of association. This in turn can be characterised as a set of relationships. The import of what I have suggested so far is that an empirical description of such a set of relationships entails the positing of a set of interconnected relational pairs. It is quite possible to posit the connection between the agreement/contract documents and the organisation based upon them as causal in nature and I might be tempted to do this is it were my intention to treat-these-phenomena-as-socialstructure. However, Egon Bittner's advice is pertinent here: “formal organisational designs are schemes of interpretation that competent and entitled users can invoke in yet unknown ways whenever it
83
suits their purposes. The varieties of ways in which the scheme can be invoked for information, direction, justification and so on, without incurring the risk of sanction, constitutes the scheme’s methodical use.” (Bittner1973,p272) The competent and entitled user who speaks in this transcript, posits one way in which those rules will be used. It is implied that the agreement that is embodied in the documents, includes that the parties will work in a more co-operative manner, in a manner determined by considerations of engineering judgement. That agreement is put into practice in the project organisation. How this works out in practice is of great interest, both in terms of the construction management research I am undertaking and for ethnomethodological studies. It is incidental to the thrust of this thesis, however, and must await explication elsewhere. The point I wish to make here is that there are many different ways of referring to a relationship, specifically here, a contractual relationship. A similar point needs to be made about attitude. The misconception that psychological descriptors such as attitude stand in causal connection to behaviour has, I think, been thoroughly exposed by Ryle (1973) and needs no further comment here. It will be sufficient to point out that, while the attribution of a cooperative attitude to one part of a relational pair, constitutes the attribution of a moral identity to a category, or the individual incumbent of a category, its attribution to both parts of a pair constitutes the attribution of that identity to a relationship. The manager immediately passes to another topic of traditional interest to sociologists, that of socialisation (line 19): It's good training for our guys
84
This leads, via a request for clarification to the characterisation of correct relations expressed above. From here, the comparison procedes. With the arguements above in mind, it can be seen that lines 1-25, in addition to attributing moral identities to categories and incumbents, also provide the resources for the attribution of moral identity to a relationship. This is made clear in lines 24-25, which can be seen as a summary of what has gone before: y'know all that's engendered by the fifth edition competitive attitude This summary adds nothing to what has gone before. Indeed, the moral value of the identity 'competitive attitude' can only be found by looking back at the previous talk. What it does do is enable the comparison which follows. This is a comparison of historical periods. The present situation is compared with that or (line 26): less than twenty years ago This period is summarised in lines 27-28: there was a bit less competition, a bit more money around, a bit more give and take
Moral Identities in the QS Interview. I shall now review the QS transcript analysed in chapter one, where two more examples of the use of moral identities may be found. These are also analysable in terms of moral identities, but display in addition some interesting features of their own. The first appears in lines 26-28: The contractor, he has er ... two QSs, er obviously, my opposite, plus er [] I think he’s only about 20, 21 years old, but he knows his stuff, y’know, he’s not too bad.
85
The moral identity ascribed here is a good one, successively formulated as 'he knows his stuff' and 'he's not too bad'. The way in which this ascription is achieved however, differs from those considered above, in that it is derived from what might be called mitigating circumstances. These are established by applying two Pn1-asequate devices to the individual who is being assessed: stage of life and age, the latter being a more precise formulation of the former. The second example (lines 49-54) displays the use of a device identified by Smith (1978) and considered by Cuff, though not in fact used in his analysis, the contrastive structure. Cuff rightly criticises Smith for the ambivalence of her own analysis, which leaves open the possibility that what is being attempted is far more than the explication of how an account is constructed: that what is intended is the subversion of the account she analyses. Nevertheless, when separated from any such intention, the device can be seen to be analytically useful. It consists essentially in setting up an acceptable way of behaving as a point of comparison with an actual way of behaving, which is then portrayed. In Smith's materials, however, it is used to demonstrate that someone (identified as K) is mentally ill and thus requires a different application from the one seen here. For the purposes of Smith's informant, it is required to be shown, not that K is properly identified as good, or bad, but that her behaviour places her beyond such categorisation. That she is, in fact, irrational. Nevertheless, as can be seen, the contrast structure is equally viable in the attribution of moral identities: They’re very poor at attending site, they don’t attend very much at all, I don’t know why, they should do, but they wait till a problem occurs and then they come out, rather than coming down every, I mean I would expect a structural engineer to come down at least every two days. That’s my experience to date, but they don’t, they don’t even come once a week, to be honest.
86
Three contrast structures can be identified in this passage: 1a. They’re very poor at attending site, they don’t attend very much at all, 1b. I don’t know why, they should do
2a. they wait till a problem occurs and then they come out 2b. rather than coming down every
3a. I would expect a structural engineer to come down at least every two days. That’s my experience to date 3b. but they don’t, they don’t even come once a week, to be honest.
All three make the same point, in the same way. A correct way of behaving is identified (1a, 2b, 3a): frequent visits to site. This acts as a contrast to actual behaviour (1b, 2a, 3b), infrequent visits. It is noticeable, however that this contrast would not work as an attribution of moral identity alone, unless the speaker could be sure that the correct and incorrect behaviours described would be identifiable as such to the hearer. In this case, the moral identity is explicitly attributed: 'poor'; 'they should do'; 'I would expect'. The purpose of the contrast structure is to supply a reason for this attribution.
Versions It was my intention in embarking on this study to tackle some problems of versions, in an attempt to develop the important, though still underrated work of E. C. Cuff (1993). The attempt to develop Cuff's work led me inevitably to its antecedents in the work of Dorothy Smith and Harvey Sacks. I became more and
87
more focused upon Sacks' concept of the membership categorisation device. The reason for this is, I believe, adequately explained in my introduction to this thesis and vindicated by my arguments thus far: the utility of Sacks' analysis for the explication of those-phenomena-usually-treated-by-sociogists-as-social-structure. The analysis of versions is an important part of such an explication, as I propose to show here. In order to do so, I will analyse a problem of versions which is apparent in the interview with the engineers analysed above. One of the versions it involves is nicely summarised in lines 32-52: It all happened in the mid-seventies didn't it, with people like [contractor's name] and [contractor's name], who first got the, virtually y'know commercial boys in, I suppose and screwed the client and clients employed them. Then in anticipation of being able to make 5% on claims, the [contractor's name]s of this world bid everything 5% below cost and that, to compete, everybody did the same thing then then then it just escalated. And you get companies still doing this, y'know, the [contractor's name]s of this world still operating on that sort of, that sort of line. But it's a very, it's a bit like farming, y'know very high input, hopefully very high return business, you've got to put in an awful lot of QS time to produce these savings and overall, are they real savings? I mean the department of, the Highway Agency was saying, wasn't it, last year, over the past year or two all their outturn costs have finished up 30%, on average their outturn costs, their final account values are something like, yes it was 30%, wasn't it? [manager B] must have [manager A] higher than average tender values. Erm, and people are saying, tut, tut, tut, the avaricious contractors and their QSs, but what that meant was in fact that the tender prices were 30% too low. Because there
88
are not excessive profits swilling around the industry. Not on average. Occasional jobs make a lot, occasional jobs don't, go bad, so At this point, manager A is interrupted by manager B, who provides an alternative version: [manager B] Either that or it was a direct result of er multiplicity of changes, as the work [manager A] Oh yes, very much so yes [manager B] I mean that is the largest single source of increased cost [manager A] yeh [manager B] whether it be on a design and build, where the client hasn't done the brief properly. Details, spent insufficient time on his brief and basic specification. Or, on conventional forms of contract where they just change their mind as time progresses [I] right [manager B] There is a ratio, I'm sure somebody would work it out. I'm sure somebody somewhere has done the equation, but I mean the outturn cost is certainly in some way proportional to the number of variations [manager A] um um mm [manager B] there's no two ways about it [manager A] probably the biggest single [manager B] clients generally think that it is the contractors taking advantage, err picking up the money on the variations, whereas in reality, I think we would prefer a contract with no variations. Because that enables you to get stuck in, build it, work efficiently without stop, start
89
[I] right It might be argued that this is not a problem of versions at all, since manager A agrees with everything which manager B says. Nevertheless, the talk following the interruption paints a very different picture from that which precedes it. I take the lack of explicit disagreement to confirm Cuff's observation that members routinely avoid conflict in their formulations: "We suggest that members might have good reasons for not making explicit formulations in their talk of what they are hearing; such formulations can lead to interactional troubles in that they might come off as 'too sharp' or 'too pointed'." (Sacks 1994, p60) Thus, I hear B's intervention as a neatly carried off piece of correctional work which, while not directly contradicting A's remarks, puts them in perspective. In this way, he his steering the talk, rather than initiating a dispute, reminding A, perhaps of a feature of the situation he has overlooked in his account. If this can be regarded, then, as at least an incipient problem of versions, it is instructive to notice what the difference between the two managers is about. Up until line 47, at least, the two appear to be in complete agreement, with manager B replying 'must have' to A's request for confirmation. Thus, the picture of the cut throat nature of competitive tendering is not being questioned. What follows constitutes an excuse for contractors behaviour in forcing up outturn costs, post tender. The tender prices were too low. It is not the moral identity thus attributed to contractors which is thus challenged however. If anything, B's version shows them in an even better light. It is rather a factual matter that is being questioned, whether higher outturn costs can be properly attributed to the tendering process, or whether on the contrary, the reason is client sponsored changes. This difference requires a different mode of analysis to the one used by
90
Cuff. To be sure, he recognises that the "moral adequacy" of an account depends upon both: "(1) how does it come off as a 'proper' description of what is happening in the world: and (2) how does it display the teller - as 'impartial' or as 'biased' or as 'sensitive' or as 'callous' or as 'involved' or as 'disinterested' or whatever?" (ibid. p40) The direction of his analysis, however, supported by the nature of his own materials, leads away from questions of facticity, towards that of value judgements: "For tellers to produce morally adequate accounts, what seems to be important is not so much their assembled nature, but the possibility that selection will display bias or one-sidedness." (ibid. p43) I am not proposing that it is the assembled nature of accounts that is of consequence in judging versions. Indeed, contra Smith's apparent stance, the concept of an account that has not been put together in some way is a nonsense. I am suggesting a third possibility for the subversion of accounts; one which Cuff's materials do not present him with. It can be seen from the transcript that B employs at least two devices to undermine the facticity of A's account and substantiate his own: that client originated changes are the most important reason for higher than tender outturn costs. First, he appeals to higher authority. He refers to a ratio, and to an equation, implying a mathematical certainty in the relationship between changes and outturn cost. Secondly, he asserts the firmness with which he holds his own position (lines 6465 & line 67): the outturn cost is certainly in some way proportional to the number of variations
91
and there's no to ways about it There is another way in which the transcript can be seen to be posing a problem of versions. This is in the way that the clients point of view is referred to, first in lines 48-49: Erm, and people are saying, tut, tut, tut, the avaricious contractors and their QSs, but what that meant was in fact that the tender prices were 30% too low. Because there are not excessive profits swilling around the industry. and again in lines 69-79 clients generally think that it is the contractors taking advantage, err picking up the money on the variations, whereas in reality, I think we would prefer a contract with no variations. Because that allows you to get stuck in, build it, work efficiently without stop, start It can be seen that each recognition of the alternative version is met with an immediate rebuttal and that the concern is with establishing a good moral identity for contractors. This highlights a further feature of the way members do status differentiation. The attribution of moral identities is not an uncontentious activity. In these two passages, the managers are effectively telling us that 'this is my point of view, others might tell you differently, but this is how it is.' Thus, an adequate sociological account of the matters discussed in this chapter must include an account of the social situation from which these phenomena have been described.
92
5. Conclusions. This thesis was prompted by the feeling that there was an area of investigation regarded by conventional sociology as its proper territory and into which ethnomethodology had barely ventured. This, I have termed, the-phenomenausually-treated-by-sociologists-as-social-structure. An important part of this area is the sociology of organisation. Despite some important studies, most notably Bittner (1973) and Zimmerman (1971), and the largely abortive innovation of studies of institutional talk in CA, terms such as organisation and structure are more usually used in ethnomethodological discourse to refer to the organisation and structure of members talk. Despite the importance of these uses, to stress for instance, the structural nature of action (as opposed, perhaps, to the structuration of action), they tend to direct attention away from a field of phenomena recognised and used by members as a feature and resource of their daily lives. I began by discussing how Garfinkel took Parson's structural notion of culture and showed how the phenomenon this described consisted in structured reflexive action. One of the implications of this analytic move is that Parson's social structure is likewise constituted. Indeed, properly understood, Garfinkel's insight leaves no room for any other possibility: the two halves of Parson's dichotomy are interdependent and if culture resolves into action, there can be no place for a separate action system. Nevertheless, the absence of a thoroughgoing explication of this necessity has permitted those who wish to re-incorporate ethnomethodology into mainstream sociology to posit it as a micro-sociology, properly complimented by conventional macro-sociology. The logical insupportability of this position notwithstanding, it has been a constant theme in the mainstream literature. Sharrock and Watson (1988), have sufficiently demonstrated this logical point. Part of my intention here has been to further explicate their position with some empirical work. 93
Sacks' MCD machinery has provided a means of achieving this. Using it I have been able to show that such so-called 'macro' phenomena as organisation, status difference and historical period are empirically and analysably present in members talk, rather than standing outside and above it. It is true that many of the conclusions I am now able to draw from this investigation and analysis are tentative and incomplete, but I hope to have, at least, established some principles to guide further study. I now list those conclusions: 1. I have explicated something of the method of presenting a conventional ethnography, showing how it involves selection and generalisation in order to produce a coherent picture. This work is a necessary first step in the construction of a sociological theory and the process involved, in its tendency to hide the empirical data, indicates an important limitation of such theorising. 2. With an ethnomethodological reversion towards an empirical approach, Sacks' MCD analysis can, with some adjustments, be used to analyse the instances from which these generalisations are derived. These adjustments consist in: (i) a reformulation of the consistency rule; (ii) Cuff's extension of the machinery to encompass the concept of moral identity; (iii) an adaptation of Smith's contrastive pair analysis to supplement this extension. 3. More precisely: attention to the consistency and economy rules makes noticeable the constitution of the collections in use in members talk. 4. The investigation of this constitution of collections reveals how their organisation can be visually represented in the form of a chart and suggests that general forms for the constitution of such collections might be findable. 5. Likewise, it seems also possible to explicate preference rules for the initial selection of such collections.
94
6. Moral identity is an important way of analysing social differentiation and inequality, capable of dealing with all the examples of this phenomenon I have thus far examined. These include status differences, not only between standard categories of person, but also between individual incumbents of those categories, and between types of relationship between categories. This last includes such phenomena as organisational forms and historical periods. 7. Problems of version, however, though only briefly touched upon here, prove in some instances to be intractable to analysis using moral identities. These factual versions require further study. 8. These devices, despite their analytic utility and their unique adequacy in being demonstrably devices which members use in the production of the analysed material, cannot be represented as existing as structures in the world. They are, rather, examples of structured actions in the world.
95
References. ANDERSON, R. J.; HUGHES, J. A. & SHARROCK, W. W. 1989 Working for Profit: the Social Organisation of Calculation in an Entrepreneurial Firm, Avebury, Aldershot. BITTNER,E. 1973 ‘The Concept of Organisation,’ in G.Salaman and K.Thompson, People and Organisations, Open University Press, Milton Keynes. CORNES,D. 1995 'Legal Implications of the NEC,' paper delivered at the School of Business and Industrial Management, 3rd February. CRAIG,R. 1994 ‘Drafting Contracts and Constructing the Team,’ RICS/West Midlands CPD lecture, 22nd November, 1994. CUFF, E.C. 1993 Problems of Versions in Everyday Situations, The International Institute for Ethnomethodology and Conversation Analysis & University Press of America. EGLIN, P. 1980a Talk and Taxonomy, John Benjamins B.V., Amsterdam. EGLIN, P. 1980b ‘Culture as Method: Location as an Interactional Device,’ in Journal of Pragmatics, 4:121-135. FRANCIS, D. & HESTOR, S. unpublished manuscript, 'Ethnomethodology Institutionalised?: Conversation Analysis and "Institutional Talk"'. GARFINKEL, H. 1952 The Perception of the Other, PhD thesis, Harvard University, Cambridge, Massachusetts. GARFINKEL, H. 1963 ‘A Conception of, and Experiments with, “Trust” as a Condition of Stable Concerted Actions,’ O.J.Harvey (ed.) Motivation and Social Interaction: Cognitive Determinants, Ronald Press, New York. GARFINKEL, H. 1984 Studies in Ethnomethodology, Polity Press, Cambridge.
96
GARFINKEL, H 1991 ‘Respecification: evidence for locally produced, naturally accountable phenomena of order, logic, reason, meaning, method, etc. in and as of the essential haecceity of immortal ordinary society (I) - an announcement of studies,’ in G.Button (ed.) Ethnomethodology and the Human Sciences, Cambridge University Press. GARFINKEL, H. 1996 'Ethnomethodology's Program,' in Social Psychology Quarterly, 59.1 GARFINKEL, H.; LYNCH, M. & LIVINGSTON, E. 1981 'The work of a discovering science construed with materials from the optically discovered pulsar,' in Philosophy of the Social Sciences, 11:131-158. HUGHES, E. C. 1971 The Sociological Eye: Selected Papers, Aldine Atherton, Chicago. ICE 1991 The New Engineering Contract: Guidance Notes, (consultative document) Thomas Telford, London. JEFFERSON, G. 1990 ‘List-Construction as a Task and Resource,’ in Interaction Competence: Studies in Ethnomethodology and Conversation Analysis, No 1, International Institute for Ethnomethodology and Conversation Analysis & University Press of America, Washington, D.C. JONES,N. 1995 ‘Legal Implications,’ in The Latham Report: Course Documents, Birmingham, Construction Study Centre. POPPER, K. R. 1959 The Logic of Scientific Discovery, Routledge, New York. ROOKE, J. & SEYMOUR, D. 1995 'The NEC and the Culture of the Industry: Some Early Findings Regarding Possible Sources of Resistance to Change', Engineering, Construction and Architectural Management, 2:4. RYLE,G. 1973 The Concept of Mind, Penguin Books, London.
97
SACKS, H. 1963 ‘Sociological Description,’ in Berkeley Journal of Sociology, 8. SACKS, H. 1972 ‘An Initial Investigation into the Usability of Conversational Data for Doing Sociology,’ in D.Sudnow (ed.) Studies in Social Interaction, Free Press, New York. SACKS, H. 1974 ‘On the Analysability of Stories by Children,’ in R. Turner (ed.) Ethnomethodology, Penguin, Harmondsworth. SACKS, H. & SCHEGLOFF, E. A. 1979 'Two Preferences in the Organization of Reference to Persons in Conversation and Their Interaction,' in G.Psathas (ed.) Everyday Language: Studies in Ethnomethodology, Irvington, New York. SACKS, H., SCHEGLOFF, E. A. & JEFFERSON, G. 1974 'A Simplest Systematics for the Organisation of Turn Taking for Conversation', in Language, 50:696-735. SCHUTZ, A. 1973 Collected Papers 1, M. Nanatanson (ed.), Martinus Nijhoff, The Hague. SEYMOUR, D. E. 1986 The Place of Organisation Theory in the study of Construction Management, PhD Thesis, University of Birmingham. SHARROCK, W.W. & WATSON, R. 1988 ‘Autonomy among social theories: the incarnation of social structures,’ in N.G.Fielding (ed.) Action and Structure, Sage, London. SMITH, D. 1978 'K is Mentally Ill: the Anatomy of a Factual Account,' in Sociology, 12.1:pp23-53 UFF, J. 1995 ‘The Unrealised Potential of Construction Lawyers,’ paper delivered to Society of Construction Lawyers, inaugural meeting of West Midlands branch.
98
WEBER, M. 1947 The Theory of Social and Economic Organisation, The Free Press, New York. WIEDER, D. L. 1974 Language and Social Reality, The Hague, Mouton. ZIMMERMAN, D. H. 1971 'The Practicalities of Rule Use' in J.D.Douglas (ed.) Understanding Everyday Life, Routledge & Kegan Paul, London.
99
Appendix 1: Membership Categorisation Devices: Rules and Definitions.
1) categorising members
MEMBERSHIP
a) a collection of membership
CATEGORISATION DEVICE
categories b) rules of application
economy rule
categorisation may be complete when each member has had one category applied to them
PN-ADEQUATE DEVICE,
contains a category for
TYPE 1
every member
consistency rule
if one Member has been categorised, others may be, using same MCD
category relevance rule 1
any category of an appropriate Pn adequate device may be used on each Member,
100
repeatable use rule
application involves determining for each Member, which category is appropriate, where each category may be repeatably used
PN-ADEQUATE DEVICE, TYPE 2
composed of programatically relevant categories, each category having either: a) a proper number of incumbents; or b) a proper minimal number
programmatic relevance
non-incumbency of a pair position (or category?) is an observable fact
modified repeatable use rule
application involves treating categories as a unit, thus population is construed in terms of the number of complete and incomplete units
category relevance rule 2
a device containing programatically relevant categories can generate not only the observation that a particular category contains a certain number of incumbents, but also that a particular category has no incumbents: its categories are programatically relevant
101
COLLECTION R
standardised paired relational categories which constitute a locus for a set of rights and obligations concerning the activity of giving help these categories are programatically relevant
standardised
known to all members
rights and obligations
any member has the right to search for help, i.e. to look for incumbents of relevant pair positions 2.
incumbents of these categories
have a right to be asked for help in preference to incumbents of other categories 3.
they have a right to be asked
for help in preference to doing suicide
Rp
set of those to whom it is proper to turn to for help the categories in this set are preferentially organised
Ri
those to whom it is improper to turn
102
categories based on knowledge
COLLECTION K
relevant to dealing with a problem (suicidalness) contains two categories: professionals and laypeople
2)categorising assertions definitive
provides the relevance of the MCD used, therefore defining the device recipient uses to form a response
ambiguous
set of consistently present proper alternative devices is available
proper alternative
can be used to: a) formulate a response; b) account for response, even if it proves to have been mistaken
contrast set
categories of which determine the correct MCD, central categories are: joke;serious.
103
Appendix 2: Research Project Meeting - Designing a Questionnaire Survey. 1
[David,S] This questionnaire would be directed .. at: architects, quantity
2
surveyors, like who? Do we have to be
3
[David,R] Management teams.
4
[David,S] Right, can we try to list the sorts of people who, we’re calling
5
these different roles in the project. That begs some questions, but er ..
6
who are we directing them at. And what’s the basis of that: directing at
7
those rather than other roles []. Cos at the end of the day we’re looking to
8
be able to say things like, y’know, ‘architects do this, have the following
9
priorities at that point in the course of the project, whereas quantity
10
surveyors see it differently, have a totally different agenda, or whatever.
11
We’re looking to do that kind of thing, right?
12
[Mick] Yes we are. We’re looking at architects, architects stroke designers,
13
I would say
14
[David,R] I mean we’re looking at, I think, so far as the design team is
15
concerned, anybody who has a ... some sort of input into it. So it could be
16
services engineers as well, a chemical engineer, whatever. The important
17
thing is that they all, to a certain extent, are interpreting a clients brief. []
18
So to a certain extent they’ve had [] to define what the brief comes out as,
19
and what the actual design is. So I think we have to ask them about
20
design team. So that’s structural engineers .. and so on. Now, whether or
21
not we actually need to ask individual [].
22
[Mick] I don’t think you do. I think that you need to talk to those people,
23
all of the people who are involved, but only those who are involved in the
24
making of those critical decisions, right? Who is going to carry the can,
104
25
as it were, for particular decisions. So I think, that if you’ve got, say
26
you’ve got er .. mechanical engineer, there might be three of them
27
[David,R] Say you’ve got a project architect and the staff
28
[Mick] Yeh, well its only the project architect
29
[David,R]
30
panel.
31
[Mick] Yeh. You will find that on all construction projects, you go to a
32
site meeting, there are only a relatively small group of people who have
33
authority to do things, yeah? And it is those people we are concerned
34
with, so project architect, er, clients.. quantity surveyor, engineer,
35
whether.. engineers, I should say, because you’ve got civils, structural,
36
mechanical, electrical and services, however, well y’know, however they
37
pan out, but those people. And then the contractor’s representatives,
38
which is likely to be possibly the contract manager, and or the quantity
39
surveyor.
40
[David,S] Notice that, are we addressing the roles that are provided for
41
within the NEC? In which case
42
[David,R] Well, I know there’s a specific term of project manager in that,
43
er
44
[Mick] I’m not, addressing that, I haven’t read it yet.
45
[David,R] I’m not particularly bothered what [] you give to them, but you
46
know the guy, the main guy, the contractor, who is actually [], then you’ve
47
got the QS whose runs the financial side, whether or not [] someone like a
48
planning engineer [].
ject architect whose the main focus of the
105
49
[Mick] But actually identifying who is important in a project is a very
50
simple thing to do, it varies from project to project
51
[David,S] Well exactly, we’re looking to what we can say at the end of the
52
day about what? Will we be saying that what our survey has established is
53
that.. somebody, some category of person thinks like this, as opposed to
54
this other category of person who thinks like this.
55
[Mick] Yes.
56
[David,S] So what category of person are we fingering, to tell our readers?
57
Are we saying that what, we claim this to be true of architects, or of...
58
[Mick] Well we’re taking the whole range aren’t we?
59
[David,S] Well OK, but lets have the list. Y’know, what are we
60
[David,R] Basically what we are doing is identifying each.. occupational
61
group like architects, like contractor’s surveyors, like client’s surveyors
62
and so on, and you want to find out what their relationship is with the
63
project to a certain extent.
64
[David,S] So you have sub categories?...
65
[David,R] er
66
[David,S] Within the categories?...
67
[Mick] A-ah, I see what, I think I see what you mean
68
[David,S] Look, its a simple question. When we’ve got our findings in
69
we’ve got to be able to say, ‘look, this substantive finding relates to those
70
people, I said what people are they?
71
[Mick] I think there are several categories.
72
[David,S] Yeah, right?
106
73
[Mick] There’s the category of.. there’s the basic them and us, right? The
74
client’s representatives and the contractor’s representatives, right? That’s
75
one sort of basic division. Then within each of those there are further sub-
76
divisions, right? So under clients representatives, you might have
77
architect, quantity surveyor
78
[David,S] But you recognise what this means in terms of a survey, that if
79
you’ve got [], the more and more categories you have, the bigger and
80
bigger the sample you’ve got to have, in order to have enough data in each
81
box to say something about that particular category.
107
Appendix 3: Interview with Client's Quantity Surveyor Establishing Who is on Site. [QS] - A client's quantity surveyor [I] - interviewer 1
[QS] Sorry, where were we?
2
[I] Er.. right.. what else? Whose the project manager?
3
[QS] The project manager being the actual project manager?
4
[I] Well yeh, under the contract.
5
[QS] [architect’s name], architects up at [location].
6
[I] And is he on site all the time?
7
[QS] No he’s just a site visitor, he comes from [location].
8
[I] Who is on site?
9
[QS] Who is on site is the three clerk of works, that’s electrical,
10
mechanical and er, building. Or, at least that’s what they’re called under a
11
normal contract, obviously it’s supervisor’s representatives, or project
12
manager’s representatives on site. [name], who’s the building clerk of
13
works, I think he’s actually called the supervisor, or the supervising
14
officer. And they’re, I wouldn’t say they’re full time, as you can see
15
they’re not here at the minute, but they have been here this morning and
16
they’re probably half a day every day. They do have other [client’s name]
17
jobs they have to er go to sites, but I don’t think they have other site
18
offices, to me they’re the only offices they have, then they’re just visiting
19
other sites to er see what []. All their administrative stuff is down here, so
20
they obviously work from here [] electrical guy ‘cos there’s not a great deal
21
occurring there, but they’re obviously down here. So they’re the three
22
guys that are on site permanently, plus myself, I’m always, or should
23
always be here, if I’m not
24
[I] If you’re not working on a different job.
25
[QS]Er yeah I should always be down here full time, strictly speaking, if I
26
do ever get any other work I should bring it here to er, to work. Just in
27
case anything does crop up. The contractor, he has er ... two QSs, er
28
obviously, my opposite, plus er [] I think he’s only about 20, 21 years old,
29
but he knows his stuff, y’know, he’s not too bad. And er they’re, should I
30
say the lad is on site permanently and the QS is on site 80% of the time.
31
Most the time, it’s not right often that we don’t see each other once in a
32
day, y’know [] every day. An I’m sure they’ve got their own site set up.
33
[I] I’d like to talk to somebody from the contractor’s side, if possible.
34
[QS] Well, [name] is the contractors QS, the boy is [name], the agent is
35
[name]. I’ve got a contracts guy, but I think he works from their main
36
offices, I could probably get you a name if you like. He’s the guy whose
37
dealt with it contractually.
38
[I] What’s the contractor’s name again?
39
[QS] [contractor’s name]. They’re not an overly big contractor.
40
[I] Whereabouts are they based?
41
[QS] [location]. I can give you the number if you like.
42
[I] That’d be good yeh.
43
[QS] Their telephone number is [] The other people involved are
44
obviously the structural engineer, and the mechanical and electrical
45
engineer. The structural engineer is [name] and the guy we have been
46
involved with is [name], the number is [telephone number] and he’ll pass
109
47
you on to [name], because the design element of the work has been done in
48
their [location] office is it? Or, down south somewhere. But he’s our
49
first contact, I should imagine he’ll just put you on to this guy. Contact
50
him first and he’ll say oh you’ll have to contact him. They’re very poor at
51
attending site, they don’t attend very much at all, I don’t know why, they
52
should do, but they wait till a problem occurs and then they come out,
53
rather than coming down every, I mean I would expect a structural
54
engineer to come down at least every two days. That’s my experience to
55
date, but they don’t,. they don’t even come once a week, to be honest.
56
[I] They won’t know a lot about what’s going on then.
57
[QS] Hence the scenario with the [] y’know, they were [] at the wrong end
58
of the building, but er what was [] on site. Obviously, er, that’s it in a
59
nutshell, obviously, things that can occur. Who else did I say?
60
Mechanical and electrical guys, that’s [company name], and that’s [].
61
And the guy you want to speak to is either [name] and [name]. I’ll tell
62
you which is which, one of them’s mechanical, and one of them’s
63
electrical.
64
[I] And they have someone on site?
65
[QS] No, again they’re just visiting.
66
[I] But is it too early for them to be?
67
[QS] It’s really too early for them to be coming down. [] they basically
68
just attend the site meetings at the minute, they have been involved in er,
69
the mechanical and electrical elements are domestic sub-contracts, rather
70
than nominated sub-contracts, so er, they’ve been involved with a
71
negotiation on there. [] they’ve been negotiating directly with the subbies.
110
Appendix 4: Interview with Two Contractor's Engineers - Doing Social Differentiation.
Extract 1. 1
[manager A] And hand on heart, y'know, if you get to the end of a job and
2
you've made a loss, you look at bloody everything, to see if there are any
3
commercial opportunities you've missed. Now that that's fair enough, I
4
mean what's traditionally happened is the client has not wanted to make a
5
decision, the contractors have wanted to keep their powder dry and see
6
what they needed to ask, so that's why. Y'know, it's kept till the end of the
7
end. It's all a damned good job creation scheme by quantity surveyors,
8
erm, who are a growing infestation in the industry and have been for ten
9
years or more. It's a great job creation scheme and we've had experience,
10
we think we've got one surveyor in our organisation who's bloody good at
11
this. An erm, no names, no pack drill, but his style appears to be just to
12
put in any old interim account during the period of the job and
13
[I] yeh
14
[manager A] the job loses money and er he pulls it round at the end.
15
Y'know, he should have been fucking doing it all the way through [laugh]
16
[I] right [laugh]
17
[manager A] but, but we're both engineers, so we, so we've got a bit of a
18
hang up about surveyors
19
[manager B] let's face it, quantity surveyors were only engineers clerks
20
and we as engineers have got ourselves to blame
21
[manger A] yeah
22
[manager B] if we've managed to make [indistinct]
23
[manager A] mm, we didn't like dealing with things like this and we've
24
we've left it to them, as the accountants. Erm, but I mean that, y'know
25
that's all engendered by the fifth edition competitive attitude. Which
26
again, started less than twenty years ago, I mean, when you and I were first
27
agents, I presume there was, a bit less competition, a bit more money
28
around, a bit more give and take
29
[manager B] A lot more give and take
30
[manager A] Things were settled. I don't think things cost more in the
31
end, actually, but things were settled easily between the clients
32
representative on site and the site agent, QSs weren't invented then. It all
33
happened in the mid-seventies didn't it, with people like [contractor's
34
name] and [contractor's name], who first got the, virtually y'know
35
commercial boys in, I suppose and screwed the client and clients employed
36
them. Then in anticipation of being able to make 5% on claims, the
37
[contractor's name]s of this world bid everything 5% below cost and that,
38
to compete, everybody did the same thing then then then it just escalated.
39
And you get companies still doing this, y'know, the [contractor's name]s of
40
this world still operating on that sort of, that sort of line. But it's a very,
41
it's a bit like farming, y'know very high input, hopefully very high return
42
business, you've got to put in an awful lot of QS time to produce these
43
savings and overall, are they real savings? I mean the department of, the
44
Highway Agency was saying, wasn't it, last year, over the past year or two
45
all their outturn costs have finished up 30%, on average their outturn costs,
46
their final account values are something like, yes it was 30%, wasn't it?
47
[manager B] must have
48
[manager A] higher than average tender values. Erm, and people are
49
saying, tut, tut, tut, the avaricious contractors and their QSs, but what that 112
50
meant was in fact that the tender prices were 30% too low. Because there
51
are not excessive profits swilling around the industry. Not on average.
52
Occasional jobs make a lot, occasional jobs don't, go bad, so
53
[manager B] Either that or it was a direct result of er multiplicity of
54
changes, as the work
55
[manager A] Oh yes, very much so yes
56
[manager B] I mean that is the largest single source of increased cost
57
[manager A] yeh
58
[manager B] whether it be on a design and build, where the client hasn't
59
done the brief properly. Details, spent insufficient time on his brief and
60
basic specification. Or, on conventional forms of contract where they just
61
change their mind as time progresses
62
[I] right
63
[manager B] There is a ratio, I'm sure somebody would work it out. I'm
64
sure somebody somewhere has done the equation, but I mean the outturn
65
cost is certainly in some way proportional to the number of variations
66
[manager A] um um mm
67
[manager B] there's no two ways about it
68
[manager A] probably the biggest single
69
[manager B] clients generally think that it is the contractors taking
70
advantage, err picking up the money on the variations, whereas in reality, I
71
think we would prefer a contract with no variations. Because that enables
72
you to get stuck in, build it, work efficiently without stop, start
73
[I] right
113
74
[manager B] to manage change, which has a delay and disruption effect on
75
your works, which is very very difficult to quantify.
Extract 2. 1
[manager A] Under the fifth edition, the contractor builds bloody
2
everything, I mean, y'know, if he's got a profit in it. If there's some profit
3
built into one particular bit of work, he'll do the bit of work
4
[I] right
5
[manager A] We've both had cases where we have gone to a client and
6
said hey you can save money if you do it differently, he's said, oh, thanks
7
very much and issued the instruction to do it differently and kept the
8
saving himself. That's happened to me
9
[I] right
10
[manager A] Usually you propose it on the basis of let's share the saving
11
and some honourable clients do that. But these days I would tend to
12
expect most clients to say that's a good idea, thank you very much and
13
keep the savings themselves. In which case, sod that, you know, if you've
14
got some of your prelims built into that item, then you've lost out haven't
15
you, straight away?
Extract 3. 1
[manager A] Well consultants that have designed it you see, especially if
2
they're on a percentage of the fee, don't always take kindly to being told
3
that y'know, Fred Bloggs Contractor Ltd. can think of a design which is
4
cheaper which then reduces the client's fee. And in fact there was a job
5
which I bid a couple of years ago, which we came up with quite a dramatic
6
alternative, which saved sort of 20%. It was quite profound, I'm not sure 114
7
what happened in the end, but we bid it, we bid this as an alternative and
8
when we rang up the consultant to say how how'd we go on? - you
9
bastards. This was his reaction, because em, not only had we embarrassed
10
him by coming up with a design 20% cheaper, his fees had gone down by
11
20%, because he's on scale fees you see, the consultant's usually paid a
12
percentage of the costs of the job.
13
[I] mm
14
[manager A] So he's none too pleased, because his fee expectation's just
15
gone down by 20% you see
16
[I] [laugh] yeh
17
Er, erm, so but he's professional, you see, he's paid per hour, cos he's like
18
the lawyer, he can be trusted you see. Whereas contractors, well nobody
19
likes paying us like that, we can't be trusted, emm... sssssshhhhhhh, what
20
else does NEC do? I don't know. It's good training for our own guys I
21
suppose
22
[I] In the sense that?
23
[manager A] Well they, y'know, there's the incentive there to, to be co-
24
operative and to look, in a pure engineering sense, y'know pure
25
engineering is solving a problem at minimum cost. Erm, so it, y'know,
26
this incentivise people to do this on both sides, which is good training for
27
the guys coming up. Rather than the bitter and twisted claim training
28
which they might have on a fifth edition Highways Agency job where the
29
QSs are running the job. I mean it happens on big road jobs, the QSs look
30
at the bill an look at the bill and drawings and seek to arrange the
31
programme and arrange the method of working to exploit the client's
32
problems. Y'know, they look at the way that the quantities of excavation
115
33
and fill are em, are billed, by the client, if they can spot mistakes there they
34
will angle their programme to exploit those.
116
