http://hum.sagepub.com Human Relations DOI: 10.1177/0018726706064179 2006; 59; 351 Human RelationsOla Bergström and David Knights processes of recruitment Organizational discourse and subjectivity: Subjectification during http://hum.sagepub.com/cgi/content/abstract/59/3/351 The online version of this article can be found at: Published by: http://www.sagepublications.com On behalf of: The Tavistock Institute can be found at: Human Relations Additional services and information for http://hum.sagepub.com/cgi/alerts Email Alerts: http://hum.sagepub.com/subscriptions Subscriptions: http://www.sagepub.com/journalsReprints.nav Reprints: http://www.sagepub.com/journalsPermissions.nav Permissions: http://hum.sagepub.com/cgi/content/refs/59/3/351 SAGE Journals Online and HighWire Press platform s): (this article cites 24 articles hosted on the Citations
29
Embed
Organizational discourse and subjectivity: Subjectification during processes of recruitment
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/26/2019 Organizational discourse and subjectivity: Subjectification during processes of recruitment
of subjects occurs, Foucault (1975: 30) concentrates primarily on the
materiality of the body and how it serves as a target for a ‘whole technology
of power’. He does not focus his attention on the detail of the social inter-
actions and how they produce and reproduce subjects. In addition, the
absence of a detailed analysis of how subjectification occurs makes an
analysis of resistance difficult.1 Only by examining organizational discourses
in action so to speak; that is, in the immediate contexts of their production
and reproduction can we see how human agency and organizational
discourses interact. An analysis of situated organizational interactions may
help us avoid the trap of determinism-voluntarism (Conrad, 2004) and
provide a better understanding of the role of discourse and agency in consti-
tuting organizations (Fairhurst, 2004).The term discourse generally refers to practices of talking and writing,
but is also used in social theory and analysis to refer to different ways of
structuring areas of knowledge and social practice (Fairclough, 1992). There
are several approaches to discourse, which have been used to investigate a
variety of social phenomena and contexts, including organizations (Oswick
et al., 2000). As Chia (2000) has argued, organizational discourse must be
understood in its wider ontological sense as the bringing into existence of an
‘organized’ or stabilized state. According to Fairclough (2005) analysis of
organizational discourse should include detailed analysis of texts in a broadsense, both written texts and spoken interaction.2 For the purposes of our
study, it is important to make a distinction between discourse as an analyti-
cal concept and discourse as an empirical phenomenon. It is primarily this
empirical usage that is drawn upon in our study of recruitment although we
trust that our discussion of organizational discourse and subjectivity
contributes also to their analytical meaning. In this article, we define organiz-
ational discourses as those instances of talk, text and conversations that take
place within organizational ‘boundaries’. Thus, subjectification may bedefined as the process of interaction contributing to the production of a
subject. Hollway (1991), with reference to Foucault, defines subjectification
as follows:
How do you ensure change without imposing it? You convince the indi-
vidual who is the object of change that they are choosing it. This is
what I mean by subjectification.
(Hollway, 1991: 95)
In order to illustrate how processes of subjectification take place in practice,
and therefore also how organizational discourses relate to subjectivity, we
draw upon ethnographic data of the recruitment practices collected by the
Bergström & Knights Organizational discourse and subjectivity 3 5 5
are hard to take into account. They therefore appear deterministic concern-
ing the capacity of recruitment practices and organizational discourses to
produce subjectivity.
However, the construction of subjectivity may also depend on the kind
of practices that are implemented. Indeed, most recruitment practices imply
processes of objectification and examination to evaluate whether candidates
should be offered a post or not. At the same time, as suggested by researchers
proposing a different approach to recruitment and selection (Wanous, 1991),
candidates are seeking to find a job that suits their personal needs and inter-
ests, assuming that recruitment is a mutual matching process between
organization and the individual recruit. Such practices, it has been argued,
would provide both the individual and the organization better opportunitiesto find a match between individual values and the corporate culture, greater
productivity (Wanous, 1991) and also to allow for the integrity of the subject
by recognizing the importance of the ‘other’ as an equivalent to the self in
terms of status and respect (Townley, 1994). This aspect of recruitment fits
well with the ambitions of the company that forms our case study, as the
recruitment manager expresses it:
The most important thing is . . . we have such a strong culture in
this company that there must be a match – a cultural match – in orderto be able to work here. It is as simple as that. Otherwise you are
pushed out.
(Recruitment Manager)
Cultural match signifies a merger between organizational discourses and
individual subjectivity. Thus, the recruitment practice of Amcon seems
appropriate for a study of subjectification. The next two sections will discuss
the characteristics of the firm’s recruitment practices and outline thediscourse analytical methods that were used.
Setting
At the time of the study Amcon was going through a period of substantial
expansion. It was also one of the most popular firms among Swedish under-
graduate students. Amcon is a career-oriented company with around 300
consultants, based on the notion of ‘up-or-out’, which means that initialadvancement is fast and dramatic for the individual. There are four basic
levels: assistant, senior, manager and partner. New entrants typically start as
assistants.
Bergström & Knights Organizational discourse and subjectivity 3 5 7
The recruitment process consists of a number of activities designed to
attract and sift candidates to achieve 70 new recruits per year. The nature of
the business and organizational structure of Amcon puts particular demands
on the recruitment activities. First, the positions vacated due to personnel
turnover need to be filled. The so-called ‘up-or-out’-process creates a continu-
ous need for new entrants. Second, there is a need to hire new workers to
fill the demand from new consultancy projects with clients. Due to the labour
intensity of consultancy work, expansion of the business is dependent on the
supply of new recruits. Third, there is a need for continuity and long-term
planning in order to be able to plan for introductory training for new
employees. The popularity of the firm means that they receive on average 85
applications every month. Applications are examined, ranked and gradedaccording to a preset grid. About a third of them are invited to a first screen-
ing interview. This first selection is based strictly on the principle of qualifi-
cations and relevant experience. Since there are no personality tests or
assessment centres, the selection interview is the event where recruiters decide
about whether or not to offer the candidate a job.
The recruitment practices at Amcon are characterized by a chain of
interviews beginning with a one-hour screening, which is held by a Manager.
The candidate may progress to another round of interviews on condition that
the result of the first interview is favourable, and this process selects outabout half of the candidates. The second stage involves a chain of three inter-
views, one hour each, where the candidate talks to Amcon staff at higher
levels of the corporate hierarchy (Senior, Manager and Partner). The process
from screening interviews to a job offer takes from three to five weeks.
After each interview the recruiter takes notes of his/her observations
of the candidate on a form. The form is constructed as a grid ranking the
candidate in terms of ‘outstanding’, ‘above average’, ‘satisfactory’ or ‘does
not meet requirements’ in terms of personal skills (e.g. personality, verbalcommunication, self-confidence) and professional skills (e.g. leadership
potential, understanding of the company; and interest in career with our
firm). The form is concluded by an overall assessment of the ‘probability of
the applicant accepting an offer’ and the ‘probability of the applicant staying
for more than 2 years’. A central objective of the recruitment interviews,
however, was to assess whether a candidate would fit the ‘organizational
culture’ – or what was termed achieving a cultural match.
The recruitment manager emphasized the importance of providing
applicants with ‘honest’ and ‘correct’ information, in order for them to makean independent, well-grounded choice to work for the company. The essence
of such processes is to deflate any unrealistic expectations of potential
employees (Townley, 1994). It is premised on the view that individuals have
fewer regrets making a decision if they can anticipate its probable negative
consequences (Townley, 1994). Amcon representatives explicitly claim to do
this by establishing a friendly atmosphere and a ‘mutual dialogue’ during the
recruitment interviews.
To establish a mutual dialogue in this chain of interviews, however, is
more complicated than might be expected. This ordinarily only occurs in
conversations between close friends. It presumes that both participants have
equal opportunity to influence the topic of speech and engage in what is
described as ‘symmetrical and co-operative’ interaction where ‘speakers both
respond to what their partners have just said and introduce something new
for them to respond to’ (Linell, 1990: 169). Mutual dialogues also presume
that there is no judgement or control of the utterances of the other (Linell,1990). Most conversations held in institutional contexts diverge from this
ideal since they are most often task oriented, that is, they have more or less
predetermined objectives. In recruitment interviews, the recruiter controls the
interaction (Linell & Gustavsson, 1987) and, in so doing, clearly exercises
power. By analysing the interaction that takes place during recruitment inter-
views, our concern is to reveal how candidates are made into subjects of
organizational discourses, that is, how subjectification takes place in practice.
We are not suggesting that by being transformed into subjects, their agency
is denied them because much of our argument is concerned with how theyparticipate in the construction of their own subjectivity and sometimes this
involves refusing to be constituted in this way.
Method
The article reports on an in-depth case study of the recruitment practices in
the Stockholm subsidiary of a large American consulting firm operating inSweden. The data for this in-depth case study of Amcon were collected
through various fieldwork techniques. These included participant observa-
tions of job fairs at universities, textual analyses of recruitment brochures,
and observations of the day-to-day work of recruitment managers and
recruitment assistants, who administer applications and handle all communi-
cation with candidates. Above all, participant observations of the personal
encounters of candidates and organizational members in recruitment inter-
views were conducted. In total 21 recruitment interviews (six screening inter-
views and 15 second interviews) were tape-recorded and transcribed. Theinterviews usually began with a promise of anonymity and an assurance by
the interviewer that the research would not affect the outcome of the recruit-
ment process.
Bergström & Knights Organizational discourse and subjectivity 3 5 9
In addition, 36 in-depth and unstructured tape-recorded interviews
were held with both recruiters and candidates after the recruitment inter-
views had taken place. These were aimed at recording the experience of both
sides in the recruitment interviews prior to job offers being made. The candi-
dates were both men and women graduates between 23 and 26 years old,
without significant work experience. By the end of the research process a
large dataset had been accumulated. The analysis began by studying the
interviews conducted with candidates, after they had gone through recruit-
ment interviews. The analysis revealed that candidates were divided into two
subgroups – those who expressed negative and critical views of the company,
questioning the professionality of the interviewers, and those who were
highly sympathetic, describing the interviewers as friendly, giving them anopportunity to ask questions, while acknowledging the status differences
among the different interviewers. The second group also described the inter-
view situation and the company as in line with their own preferences: ‘I feel
at home’, ‘it feels completely right for me’; ‘it feels like a step further’. Maybe,
most importantly, they described the interview situation as informative,
enabling them to make a well-informed choice to work for the company.
Thus, the interviews with candidates showed examples of both active resist-
ance and active subjection to the company and its recruitment practices. The
last group were those who were offered a job at the company; however, itshould be noted that candidates were not given any notification of a job offer
during the interviews.
The analysis continued by identifying the recruitment interviews in
further detail, aiming at identifying the pattern of interaction between candi-
dates and interviewers. The analysis of recruitment interviews consisted of
three main stages. The first stage involved transcription of interviews, coding
and close reading of the material. The first impression was that interviewers
dominated the conversation by talking more than interviewees, as if they hadswitched roles. In order to check this observation the number of words
spoken by interviewers and interviewees respectively was calculated. It
turned out that the interviewer dominated the conversation in 16 of the 21
interviews. The relevant question was then to analyse how interviewers
created and retained this dominance during the course of the conversation.
All interviews were then again carefully coded and analysed according to the
scheme suggested by Adelswärd (1988), elucidating the structure and phases
of the conversation: how agendas were set, turn-taking, change of topic and
how interviews were ended (Fairclough, 1992). This analysis revealed threetypes of interaction – discursive moves – that seemed to characterize all the
interviews with some variation. These were then interpreted through the
conceptual framework of Linell and Gustavsson (1987) in terms of how they
contributed to controlling, supporting and/or circumventing candidate
responses as a part of the social practice in which recruitment interviews took
place. Finally, the interpretations were validated, discarded, or modified
through repeated rereading of the material in search of examples, counter-
examples, evidence and exceptions.
Findings
The analysis of recruitment interviews at Amcon provided evidence that
subjectivity was constructed in the interaction between candidates and inter-
viewers through three subsequent and systematically recurrent discursivemoves. These were, first, response-control (where interviewers set the agenda
to have a ‘mutual’ dialogue). Second, the enunciation of organizational
discourses (where interviewers describe the organization as being honest, fair
and realistic), and, third, various housekeeping moves (to control and
confirm that the candidates’ subjective expression of organizational
discourses were ‘genuine’). Each of these were found in all the interviews
analysed and all three of them supported the impression for candidates that
they could actively participate in, and influence, the decision to secure a job
offer. Thus, these discursive moves may be regarded as examples of howsubjectification was constructed – the process contributing to the production
of a subject. The following three subsections will map out the characteristics
and dynamics of these discursive moves in further detail.
Invitation to a dialogue
The way that recruitment interviews are initiated at Amcon follows a similar
pattern. The interviewer informs the candidate that this is an occasion foran ‘open and mutual dialogue’, where the participating parties should ‘get
to know each other’. In particular they emphasize the importance of provid-
ing the candidate with: ‘an honest and realistic image of the company’,
‘personal information from people from all hierarchical levels’, ‘a possibility
to ask questions’, ‘to feel completely confident about what you buy yourself
into’ and ‘a job that you really want’. Thus, this setting of the agenda could
be seen as a ‘distribution of communicative responsibility’ (Linell & Gustavs-
son, 1987: 27). When a participant takes the initiative to communicate, a
pattern of expectations, rights and obligations are generated. The initiationof the interview includes the expression of a number of expectations (that
the candidates should actively engage in seeking information, to form an
opinion and an understanding of the job). Furthermore, it includes the
Bergström & Knights Organizational discourse and subjectivity 3 6 1
establishment of rights (to ask questions and to obtain all the necessary infor-
mation to make a decision) and obligations (that the interviewers should
contribute with all the required information and be honest). Thus, the distri-
bution of responsibility implies that the candidate is made responsible for
gathering information, evaluation and assessment and decision-making,
while the interviewers are expected to be passive transmitters of the infor-
mation required by the candidate to make his or her decision.
However, the distribution of responsibility between participants in the
conversation actually takes the form of, what Linell and Gustavsson (1987:
37) call, response control – ‘an expression of initiative that limits the freedom
of the respondent to vary his/her answer’. At the outset, the interviewers
generally present their version of what the conversation is about, and therebyset the agenda without explicitly requesting any response from the candidates.
Manager: OK, Christian, I thought we should take up a few things. I
can start by describing a little about myself, then I want you to tell me
as much as you know about our company, so that I can get a picture
of what you know about us. People’s understanding about us varies
considerably. It may be dependent on who you know or don’t know,
if you have been to information events and so on. Then, when you have
said that, I can go in and tell you more about us, to complement yourimage. I want you to get as good an image as you could possibly get,
or as realistic an image as possible. It is good with a fair and realistic
image, and then we talk more about you.
As this example shows, the interviewer provides a rather detailed descrip-
tion of what will happen during the interview. He gives clear instructions
of who should speak, what should be talked about and in what order this
should take place. The interviewer addresses the candidate with his firstname, thereby indicating a personal tone for the conversation. Two aspects
are emphasized. First, the interviewer seeks to discover the candidate’s
knowledge about the company. Second, the interviewer provides infor-
mation complementary to what the candidate already knows about the
company – so they have a ‘fair’ and ‘realistic image’. The interviewer’s
interest in obtaining information about candidates is limited to their knowl-
edge of the company and is mentioned in a subordinate clause. Thus, the
interview is not first of all structured as a way to extract information about
the candidate. Instead, it is focused on providing candidates with a soundbasis for their eventual decision as to whether to accept the job if offered
construct the conditions for the applicant to arrive at a decision whether he
or she wants to work there.
The organizational discourse at Amcon is structured in terms of what
may be regarded as advantages to the candidate (e.g. development, continu-
ous change, variation, career, possibility to work in different industries,
learning and training) and disadvantages (programming, heavy workload,
travel requirements, overtime and difficulties to combine the work with
private and family life). According to Amcon representatives, the presen-
tation of disadvantages was regarded as a way to provide a realistic image.
‘They should know what they embark upon! They must like all aspects of
it!’ – one of the managers, who claims to have a clear understanding of what
the interviews are all about, said. It was argued that the reason for provid-ing the applicants with an abundance of information about the company
was the experience of having ‘newcomers leaving the company because they
lacked an understanding of what the company stands for, and such mistakes
are expensive’. Thus, from the company’s perspective it is important that
the applicant not only has a positive image of the company. They should
also be aware of the dark side of consulting. The interviewer needs to make
sure that the applicant does not have any delusions about what it means to
be a consultant.
The aspects the applicant is informed about are the travelling require-ments, overtime, that the job is difficult to combine with having children and
that the job, to a large extent, is about programming and coding. The inter-
viewers also inform candidates about the assessment programs, internal
training, business plans, the competitive situation in the industry, oppor-
tunities for advancement in the firm and what it means to work in a project
team. It is also emphasized that working as a consultant means you can work
in different industries and build up experience over time. Instead of describ-
ing the working conditions in further detail, interviewers talk about thepossibility of making a career. The consultancy career is described as exciting,
with opportunities to move and change jobs on condition that you are
‘happy’. This message is conveyed both explicitly and by the interviewers
talking about their own career as a living evidence of what is said.
To a large extent the stories told by interviewers are repetitions of the
same stories that the candidates could obtain in previous contacts with the
company or through brochures. It may also be argued that the discourses
candidates are exposed to are similar to those analysed by Grey (1994).
However, the analysis of the situated interaction in which the discourses arepresented reveal important differences. During the interviews, a more
detailed image of the working conditions is presented. As an example, one
interviewer specifies what the travel requirements mean:
and industries). As isolated phenomena, each of these is relatively simple to
take a stand in favour of or against. By directing attention to separated
decisions and a given number of alternatives, attention is diverted from the
more general choice that is made at the point of each single decision. This
means that it is difficult for the candidate to foresee the consequences of the
decision on the basis of the information that is provided by the interviewers.
Second, the description of opposites is subdivided in a time dimension.
Disadvantages are tied to the present (‘the first two years’; ‘the first time’; ‘a
certain time’ or ‘concern the newly employed’), and advantages are pushed
forward to the future (‘increases the longer you work’; ‘happens when you
are Senior’; ‘small chances during the first year’). The disadvantages then are
presented as if they are merely of a temporary and superficial nature, whereasthe positive aspects of the job are what candidates can look forward to and
enjoy later, assuming they commit themselves to the company. This way of
describing the job has the effect of making the choice to work at the company
comparatively simple. However, the descriptions do not include any guar-
antees that the advantages will be realized nor that additional disadvantages
might not appear. This means that the future consequences of a choice to join
the organization, based on the information provided, are uncertain despite
the clarity, transparency and ‘realistic’ image provided by the interviewers.
Third, the descriptions are expressions of a division of responsibilitybetween employer and employee. The positive aspects are those that the indi-
vidual employee has responsibility to take advantage of (‘it’s up to you’;
things that are ‘individual’) or dependent on factors that are beyond the
influence of the employer (‘clients needs and demands’; ‘what kind of clients
we have at the moment’). The negative aspects are, however, what the candi-
date ‘should count on’, ‘be aware of’, ‘need to accept’, ‘important that you
approve of’, ‘to be in agreement with’ or ‘be clear of’. Thus, the individual
is held responsible for any possible consequences of a decision. The responsi-bility of the employer is less clearly defined. This means that any decision
based on this information only has consequences (both negative and positive)
for the individual (the one who decides) and nobody else.
In sum, despite the interviewers’ ambition to provide candidates with
fair, honest and realistic information, the way it is presented implies a
particular truth effect, providing an impression of having access to infor-
mation that would facilitate a well-informed choice of working for the
company. The interviewer conducts the discourse as if the images of the
working conditions presented were perfectly transparent and fair, but theywere really prescriptions or imperatives regarding how the working
conditions should be understood and interpreted. In the next section we will
take a closer look at how the interviewers make sure that the candidates have
of the interviewers, but has some difficulties in accepting working overtime.
In the third move the interviewer confirms the candidate’s views and signals
that his understanding is correct.
In the recruitment interviews, the follow up of the candidates’
responses fill several functions: (1) They signal to the candidate what is the
right kind of answer, give a hint of how the next initiative should be answered
and support the decision-making of the candidate: ‘I think that’s best, really’;
‘I believe that if you should start working for us, that’s probably what is least
in line with what you’ve been doing so far.’ (2) They evaluate and deepen
the standpoints made by the candidate in order to guarantee that it is the
‘right’ motives that drive him or her: ‘that’s rather broad’; ‘that’s a little un-
focused’; ‘But do you buy into this? Because that’s a really difficult question.’(3) The third move remarks also fill the function of denying, correcting or
softening understandings that the candidates have achieved during previous
interviews, as in the following case when the interviewer asks whether there
is anything else the candidate wonders about:
Candidate: Eh, not really, I guess I had some doubts when I came here.
But I think that I have received many good answers. What I have tried
to figure out so far has been more about, if I now get the chance to
work with you, what kind of people I will meet and what kind of people I will work together with? How large groups for example? How
does this . . . work? The organization is quite permeated by American
thinking and I have had several hierarchical pictures drawn for me,
huh, and then you would like to know how cooperation works in this
hierarchy. You may look at it and get scared to death and think – this
is how it is going to be! Do you have to make an appointment to talk
to your closest boss? Are you sure about having another five, six, ten
other people as well? But now I have been told that it doesn’t reallywork like that, rather . . .
Partner: No, we often draw these hierarchies. But it doesn’t really work
like that, so to say. It is rather so that it is a little messy and it might
as well be a way to get some structure in our existence. It is very mixed
really. It is important to look at how both formal and informal
communication channels work out here at the company. It is very cosy
to work here.
[Interview 3:4]
When the candidate asks questions about things that worry him or her,
the interviewer plays down or corrects the misunderstandings. What the
Bergström & Knights Organizational discourse and subjectivity 3 6 9
candidate regards as worrying is turned into something that may be regarded
as an advantage. For example, the hierarchy is described as a way ‘to get
some structure in our existence’, which otherwise is quite messy. In other
words, the interviewers make sure that it is the company’s version of reality
that forms the basis for the decision of the candidate: ‘No, we often draw
these hierarchies, but it does not really work in that way’; ‘You could get a
feeling that there is a one-to-one competition between individuals, but it’s
not really like that.’
The result of the interviewers’ third move remarks on the responses of
the candidates is that those who enter the organization do not have any
doubts about the advantages of the company. It is rather the other way
around, as one Partner expressed it; ‘all question marks are made into excla-mation marks’. Thus, the discursive moves of the interviewers contribute to
what Deetz (1992) calls discursive closure – the establishment of normalized,
conflict-free experiences and social relationships. The company is guaranteed
that those who are recruited do not have any ‘incorrect understandings of
how it is. We simply do not want any wrongful recruitments’, as one of the
Managers put it. The organization is, so to say, vaccinated from doubts.
Conclusion and implications
This article has sought to illustrate and explore some of the ways that subjec-
tivity is constructed in the context of a recruitment process where there was
an ambition to recruit candidates that match the culture of the firm. Drawing
upon recruitment interview data collected in a Swedish subsidiary of an
American consultancy firm, it challenges widespread assumptions about the
relationship between organizational discourse and subjectivity by revealing
some of the ways that individual subjectivity was constructed through socialinteraction. The story of Amcon, we argue, suggests that the relationship
between organizational discourses and subjectivity cannot be captured
adequately either through determinist or voluntarist theoretical approaches.
In short, subjectivity is neither wholly determined by organizational
discourses nor simply a product of human agency. Rather, the research indi-
cates that subjectification is a complex condition and consequence of the
mutually interdependent relations of agency and discourse, not a determi-
nant of either.
More specifically, the analysis of data collected at Amcon provided threeexamples of how human agency and organizational discourse interacted in
the construction of subjectivity in this specific context. First, subjectification
revealed itself through the process of candidates’ expressing their acceptance
of the working conditions presented by the interviewers, spontaneously or
through their independent careful consideration. Second, subjectification was
illustrated through the candidates’ active subjection to organizational
discourses, describing the interview situation and the company as in line with
their own preferences: ‘I feel at home’, ‘it feels completely right for me’. Third,
it indicated that those candidates who displayed resistance or were reluctant
to accept the organizational discourse failed to secure a job offer, while the
successful were made to believe that they were making active autonomous
choices. This is a major aspect of their subjectification since it appeals to more
universal meta-narratives of human autonomy that are a legacy of the
Enlightenment (Knights & Willmott, 2002). Subjectification, in sum, included
both elements of active participation and resistance from the point of view of the candidates.
Subjectification, it was further argued, was shaped and fixed through
the interaction between recruiter and candidate but this was largely inspired
by the concern to provide ‘honest’, ‘realistic’ and ‘fair’ information and to
engage in an open dialogue. Since it seems plausible to assume that
power/knowledge relations are prevalent in any recruitment process, the
uniqueness of this particular setting lies not in the effect on subjectivity so
much as in how it is accomplished. Interviewers sought to influence not only
what information candidates’ use as a basis for their decision, but also howthis should be evaluated and judged in relation to other information. This
was achieved through a number of discursive moves such as distributing
communicative responsibility, controlling candidate responses, and through
various housekeeping moves. According to researchers of institutionalized
conversations (Linell & Gustavsson, 1987), such discursive moves are readily
available in recruitment interviews and are not so much outcomes of asym-
metrical power relations as examples of how power relations are constructed.
Thus, subjectification was constructed through systematic control of candi-date expressions but with their full support given that candidates were led
to believe they were making autonomous choices. Butler (2004) has referred
to this as the paradox of seeking recognition while feeling impelled to reject
the social norms through which that recognition is conferred. As a conse-
quence the ‘I’ that bears an identity may be undone precisely by that identity
but resistance will depend on the personal or social costs of the constraints.
In the sphere of gender and sexual subjectivity, these constraints may be a
price too high for the recognition that is accorded and may indeed produce
a life that is ‘unlivable’3 (Butler, 2004: 4). Our research subjects, by contrast,rarely found the constraints too extensive but where this was so, they simply
remained outside of the norms that could ensure their being recruited to the
job for which they had applied.
Bergström & Knights Organizational discourse and subjectivity 3 7 1
For the assumption of some of these Foucauldians is that management and
organizational discourses constitute subjectivity directly as if subjects were
‘cultural dopes’ (Garfinkel, 1967) rather than active participants in the
construction of their own self-identity. Candidates themselves reproduced
organizational discourses as a part of their own self-understanding, continu-
ally recreating rather than being passively determined by the discourse. But
they were also indirectly influenced to take on these descriptions through the
interviewers’ control of the agenda, presentation of organizational discoursesand the way that candidates’ responses were managed.
The notion of interaction does not only raise questions about certain
Foucauldian analyses of work organizations, but equally also some of the
critics of these approaches. For example, it is argued that some Foucauldians
underestimate the power of human agency to avoid or resist the impact of
organizational discourses and, therefore, overestimate the power of discourse
to determine individual subjectivity (see, for example, Sosteric, 1996;
Newton, 1998). This critique has been very important in identifying the limi-
tations of applying a Foucauldian analysis to organizational contexts, andwe have sought through detailed empirical analysis to avoid those limitations
and show how subjectivity is a complex outcome of the co-related practices
of self-managed agency and discourses of power/knowledge. The tendency
for Foucault to come across as a determinist and for some theorists (e.g.
Sewell & Wilkinson, 1992; Fernie & Metcalf, 1998) to use him in this way
is partly because he failed to spell out the social processes of subjectification
– how we come to be constituted through power – through our own agency
in the context of discourses and their interdependent co-production of subjectivity and power/knowledge relations. Our study points to an alterna-
tive that privileges neither discourse nor agency but sees them as in a complex
intermediation. As Newton (1998) suggests, to emphasize such agency is not
to posit some essential subject, but rather to argue that understanding how
the subject is constituted in discourse requires attention to the social
processes through which people actively manoeuvre in relation to discursive
practices (p. 426). This we believe provides a stronger and more elaborated
critique of power/knowledge relations in organizations.
It should be noted, however, that while the end of exposingpower/knowledge relations in organizational contexts is ‘relevant’, it should
not be confused with its means. The critique of industrial or commercial
organizational discourses and practices with reference to their parallel with
those practices of hospital and prison discipline and surveillance analysed by
Foucault (1967, 1975) may well have critical potential, but it does so merely
by analogy. The empirical detail still needs to be documented in order to
show how subjects are vulnerable to the power effects of management and
organizational discourse and practice. One problem with Foucauldians is
that they often assume that the disciplinary technologies within organiz-
ational practices (or discourses) simply constitute subjectivity. In principle,
they claim that organizational discourses and practices are problematic
because of their disciplinary effects on subjectivity, but seldom are they clear
about how this occurs. In the absence of a disclosure of how the processes
of subjectification occur, we cannot further critical analysis of organizational
discourses and delimit when, where and how they have their effects.Insofar as critical organizational analysis concentrates so much on the
object of its critique as to neglect the means it uses, there is a risk that it
undermines its own criticality. Not all organizational discourses have the
effect of subjectifying individuals. Equally, far from all instances of subjectifi-
cation are to be seen as negative or problematic. As Foucault (1980) argues,
power/knowledge relations also have a positive and productive side. It would
be rare, for example, to question the subjection of medical doctors to
discourses of medicine in their practice to save human lives, but sometimes
medical discipline can be counterproductive in rendering the professionimpotent to respond adequately to patients (Gleeson & Knights, 2006). It
should also be noted that we are not simply encouraging more scientific
rigour and empiricism in critical research, but in order for critical research
to realize its potential, it should be more explicit about the conditions under
which its critique applies. By identifying when, where and how discourses
have their effects on individuals, for example, a more constructive critique
of managerial and organizational discourses may be upheld. This study has
sought to illustrate the claim that a particular organizational discourseconstitutes the subjectivity of candidates through providing a detailed
analysis of the particular social interactions of the recruitment processes
where subjectification took place.
Since our case study is limited to one particular kind of interaction,
that is, recruitment practices, more research is clearly in order. As a means
of realizing more fully how and when organizational discourses constitute
subjectivity, there is a need to complement this study with studies of other
kinds of interaction and in other contexts. More specifically, since recruit-
ment is an occasion where subjectification takes place, there is clearly a needto study how subjectivities are reproduced when subjects have entered the
organization. Furthermore, since subjects are continuously exposed to
organizational discourses, there also seems to be a need to examine how
Bergström & Knights Organizational discourse and subjectivity 3 7 3
subjectivity is constituted in less formal interactions of everyday life, before
candidates have entered a particular organizational context, where inter-
action is more distant and less direct. In examining the micro-practices of
communication between organizational members and subjects, such studies
may help organizational theorists further the understanding of the relation-
ship between organizational discourses and subjectivity in contemporary
organizational life. This study, we trust, is a case in point.
Acknowledgements
We acknowledge the helpful critical comments of the anonymous reviewers and
the associate editor, Barbara Townley.
Notes
1 We are aware that Foucault sought to avoid theorizing resistance on the basis that
he believed it only provided those at whom it was targeted with the information
that could be used to deflect or incorporate such resistance (Foucault, 1980, 1982).2 As one of the reviewers pointed out to us, since writing our first draft of this article,
Fairclough (2005) has declared himself a critical realist. We resist following his
critical realist ontology and his support for Reed’s claim that critics of critical realismcollapse ontology into epistemology. Unless we have God-like metaphysical powers,
how else would we know anything about ontology except through our epistemo-logical reasoning or sensemaking? In our article we are trying to avoid the meta-
physics of ontological debate since this would be difficult to resolve even in an articlethat was entirely theoretical. While aware of the tensions between Foucault and
critical realism, we avoid following the extreme position of some constructivists (e.g.
Grint & Woolgar, 1992) or that of critical realists (e.g. Reed, 2000b). We thereforestay with epistemology – what it is possible to know even though such knowing is
invariably transient, transitory and precarious.
3 When gays or lesbians demand the same kinship rights as heterosexuals through
marriage, for example, they reproduce social norms through which they havetraditionally been oppressed (see Butler, 2004).
References
Adelswärd, V. Styles of success: On impression management as collaborative action in job
interviews. Linköping: University of Linköping, 1988.
Alvesson, M. & Kärreman, D. Varieties of discourse: On the study of organizations through
discourse analysis. Human Relations, 2000, 53(9), 1125–49.Anderson-Gough, F., Grey, C. & Robson, K. In the name of the client: The service ethic in
two professional services firms. Human Relations, 2000, 53(9), 1151–74.Barratt, E. Foucault, HRM and the ethos of the critical management scholar. Journal of
Management Studies, 2003a, 40(5), 1069–87.
Barratt, E. Representing enterprise: The texts of recruitment and change in the UK bankingsector. Culture and Organization, 2003b, 9(3), 145–60.
Burrell, G. Modernism, post modernism and organizational analysis 2: The contribution
of Michel Foucault. Organization Studies, 1988, 9(2), 221–35.Butler, J. Undoing gender. New York: Routledge, 2004.
Chia, R. Discourse analysis as organizational analysis. Organization, 2000, 7 (3), 513–18.Collinson, D.L. Identities and insecurities: Selves at work. Organization, 2003, 10(3),
527–47.
Collinson, D., Knights, D. & Collinson, M. Managing to discriminate. London: Routledge,1990.
Conrad, C. Organizational discourse analysis: Avoiding the determinism-voluntarism trap.Organization, 2004, 11(3), 427–39.
Cooper, R. & Burrell, G. Modernism, postmodernism and organisational analysis: An
introduction. Organization Studies, 1988, 9, 91–112.Dean, M. Critical and effective histories, Foucault’s methods and historical sociology.
London: Routledge, 1994.Deetz, S. Democracy in an age of corporate colonization. Albany: State University of New
York Press, 1992.Fairclough, N. Discourse and social change. Cambridge: Polity Press, 1992.Fairclough, N. Discourse analysis in organization studies: The case for critical realism.
Organization Studies, 2005, 26(6), 915–39.
Fairhurst, G.T. Textuality and agency in interaction analysis. Organization, 2004, 11(3),335–53.
Fernie, S. & Metcalf, D. Not hanging on the telephone: Payment systems in the new sweat
shops. London: Centre for Economic Performance, LSE, 1998.
Foucault, M. Madness and civilization. London: Tavistock, 1967.
Foucault, M. Discipline and punish. Harmondsworth: Penguin, 1975.Foucault, M. Power/knowledge: Selected interviews and other writings 1972–77 . Trans.,
Ed. C. Gordon. London: Tavistock, 1980.Foucault, M. The subject and power. In H.L. Dreyfus & P. Rabinow (Eds), Beyond struc-
turalism and hermeneutics. Brighton: Harvester Press, 1982, pp. 208–26.
Gabriel, Y. The unmanaged organization: Stories, fantasies and subjectivity. Organization
Studies, 1995, 16(3), 477–501.
Garfinkel, H. Studies in ethnomethodology. Upper Saddle River, NJ: Prentice Hall, 1967.Gergen, K. The saturated self: Dilemmas of identity in contemporary life. New York: Basic
Books, 1991.
Giddens, A. Central problems in social theory. London: Macmillan, 1979.Giddens, A. Modernity and self identity: Self and society in the late modern age. Cambridge:
Polity Press, 1991.
Gleeson, D. & Knights, D. Challenging dualism: Public professionalism in ‘troubled’ times.Sociology, 2006, 40(2).
Grey, C. Career as a project of the self and labour process discipline. Sociology, 1994,28(2), 479–97.
Grint, K. & Woolgar, S. Computers, guns and roses: What’s social about being shot?Science, Technology and Human Values, 1992, 17 , 366–80.
Hardy, C. Researching organizational discourse. International Studies of Management and Organization, 2001, 31(3), 25–47.
Hardy, C., Palmer, I. & Phillips, N. Discourse as a strategic resource. Human Relations,
2000, 53(9), 1227–48.
Hetherington, K. Expressions of identity: Space, performance, politics. London: Sage,
1998.Hodgson, D. Discourse, discipline and the subject: A Foucauldian analysis of the UK
Knights, D. Organization theory in the age of deconstruction: Dualism, gender and post-
modernism revisited. Organization Studies, 1997, 18(1), 1–19.Knights, D. & Collinson, D. Disciplining the shopfloor: A comparison of the disciplinary
effects of managerial psychology and financial accounting. Accounting, Organisationsand Society, 1987, 12(5), 457–77.
Knights, D. & Morgan, G. Corporate strategy, subjectivity and organisations. Organiz-ation Studies, 1991, 12(2), 251–73.
Knights, D. & Willmott, H. Power and subjectivity at work: From degradation to subju-
gation in social relations. Sociology, 1989, 23(4), 535–58.Knights, D. & Willmott, H.C. Virtue or vice? Reflecting on competing discourses of
autonomy. In M. Parker (Ed.), Utopia and organization. London: Sage, 2002, pp. 59–81.
Linell, P. The power of dialogue dynamics. In I. Markova & K. Foppa (Eds), Dynamics of
dialogue. New York: Harvester Wheatsheaf, 1990.
Linell, P. & Gustavsson, L. Initiativ och respons, om dialogens dynamik, dominans ochkoherens (Initiative and response, about the dynamics of dialogue, dominance and
coherence). Linköping: Universitetet i Linköping, 1987.McKinlay, A. & Starkey, K. Foucault, management and organization theory. London: Sage,
1998.
McNay, L. Foucault and feminism. Oxford: Polity Press, 1992.
Miller, P. & Rose, N. The Tavistock programme: The government of subjectivity and sociallife. Sociology, 1988, 22(2), 171–92.
Newton, T.J. Discourse and agency: The example of personnel psychology and ‘assessmentcentres’. Organization Studies, 1994, 15(6), 879–902.
Newton, T.J. Theorizing subjectivity in organizations: The failure of Foucauldian studies.Organization Studies, 1998, 19, 415–47.
Oswick, C., Keenoy, W.T. & Grant, D. Discourse, organizations and organizing: Concepts,
objects and subjects. Human Relations, 2000, 53(9), 1115–23.Peltonen, T. Finnish engineers becoming expatriates: Biographical narratives and subjec-
tivity, studies in cultures. Organizations and Society, 1999, 5, 265–95.
Phillips, N. & Hardy, C. Discourse analysis: Investigating processes of social construction.
Thousand Oaks: Sage, 2002.
Potter, J. Representing reality: Discourse, rhetoric and social construction. London: Sage,1996.
Reed, M. The limits of discourse analysis in organizational analysis. Organization, 2000a,7 (3), 524–30.
Reed, M. In praise of duality and dualism: Rethinking agency and structure in organiz-
ational analysis. In S. Ackroyd & S. Fleetwood (Eds), Realist perspectives on manage-
ment and organizations. London: Routledge, 2000b, pp. 45–65.Rose, N. Governing the soul: The shaping of the private self . London: Routledge, 1989.
Sewell, G. & Wilkinson, B. Someone to watch over me: Surveillance, discipline and the JITLabour Process. Sociology, 1992, 26(2), 271–89.
Sosteric, M. Subjectivity and the labour process: A case study in the restaurant industry.Work, Employment and Society, 1996, 10(2), 297–318.
Sveningsson, S. & Alvesson, M. Managing managerial identities: Organizational fragmen-
tation, discourse and identity struggle. Human Relations, 2003, 56(10), 1163–93.Ten Bos, R. & Rhodes, C. The game of exemplarity: Subjectivity, work and the impossible
politics of purity. Scandinavian Journal of Management , 2003, 19, 403–23.
Thompson, P. & Ackroyd, S. All quiet on the workplace front? A critique of recent trends
in British Industrial Sociology. Sociology, 1995, 29(4), 615–33.Townley, B. Foucault, power/knowledge, and its relevance for human resource manage-
ment. Academy of Management Review, 1993, 18(3), 518–45.
Townley, B. Reframing human resource management. Power ethics and the subject at work.