Organization Studies

Controls of Knowledge 32(4) 489­–508

© The Author(s) 2011
Reprints and permission:
Production, Sharing and sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0170840611400284
Use in Bureaucratized www.egosnet.org/os

Professional Service Firms

Marion Brivot
Concordia University, Canada

Abstract
One of the main obstacles to the current bureaucratization trend in large professional service firms (PSFs)
is the organic nature of professional knowledge production, sharing and use. Centralized knowledge
management (KM) systems aimed at codifying ‘best practice’ solutions to recurrent client questions
for large-scale reuse are a common strategy increasingly employed to overcome this obstacle. Using a
socio-ethnographic case study of a business law firm in Paris, this research examines whether the use
of centralized KM systems in bureaucratized PSFs contributes to a shift in power from professionals
to managers. More specifically, are administrative controls over knowledge resources increasing, or do
professionals retain power (i.e. some level of social and self-control) over knowledge production, sharing
and use? The results of this study indicate that, far from losing ground, professionals’ social and self-controls
have been reinvented and reformed in a bureaucratized context.

Keywords
bureaucratization, knowledge management (KM) systems, organizational archetypes, organizational controls,
professional service firms (PSFs)

Introduction
After the early debates amongst sociologists about the real or purported distinctiveness of profes-
sional work (e.g. Freidson 1986), contemporary views have converged upon a recognition that all
professions deal with human-related issues that are unique, complex and have uncertain outcomes
(Champy 2009). That being so, it has been suggested that the systematic application of standard-
ized knowledge to solve unique client problems (e.g. legal or medical issues) could be catastrophic.
For example, in a healthcare setting, professionals need to take account of specificities such as a
patient’s medical history, their current medications and allergies, and the existence of alternative

Corresponding author:
Marion Brivot, John Molson School of Business, Concordia University, 1455 de Maisonneuve Blvd. West,
Montreal, Quebec, Canada H3G 1M8
Email: [email protected]
490 Organization Studies 32(4)

pathologies, before rendering a professional judgement; even the diagnosis of a well-known

condition does not automatically lead to the prescription of a single ‘best practice’ treatment. Such
decisions necessarily involve uncertainty, risk-taking and the adaptation of theoretical knowledge
to a particular case, and it is this non-algorithmic quality of professional judgements which has
been used to justify professionals’ claims for autonomy at work and the appropriateness of self/
peer controls rather than enforced regulation.
Established forms of professional governance and practice are, however, being undermined. As
Olgiati (2008: 557) notes, recent economic trends have emphasized an alternative approach: ‘the
knowledge-based economic model is rooted in rationalistic and universalistic principles about
the existence of a best way. This implies a top down enforcement of best practices as formally,
technically, standardized, specialized and certified according to the one-dimensional logic of mar-
ket imperatives’. This philosophy has fuelled the rise of first-generation knowledge management
(KM) systems1 which, from the late 1990s onwards, have made the centralization and control of
processes of knowledge production, sharing and use both easy and cheap (Hansen et al. 1999).
Professional service firms (PSFs) have been enthusiastic adopters of such systems, not least
because ‘successful professionalisation embodies a system of codification and standardisation of
knowledge, which at the same time can be used as a strategy to control professionals ... . It is in this
respect that professionalism and the entrenched expert knowledge system may act as a host for
governance practices’ (Kuhlmann and Burau 2008: 624).
Yet the knowledge-as-a-commodity vision advocated by the knowledge-based economic model
and the first generation of KM systems challenges professionals’ long-standing claims that their
expertise is largely tacit and that quality professional judgement requires experience and acumen
rather than the application of standardized solutions to known problems.
Although the use (or non-use) of KM systems as decision support tools by professionals has
been studied extensively, their use as managerial control devices is not well understood. Given the
mood of growing distrust in professional judgements (Evetts 2008) and an increasing demand for
transparent, accountable, and cost-efficient professional services (Kuhlmann and Burau 2008), this
paper asks whether KM systems are contributing to the breakdown of the traditional ‘organic
professional-knowledge nexus’, to borrow Olgiati’s (2006: 533) expression. More specifically,
do KM systems shift the balance of organizational power in PSFs towards administrators, thus
reducing professionals’ social and self-control over the production, sharing and use of codified
knowledge? This paper seeks to answer this question by reporting an exploratory analysis of a
case study in the Parisian office of a French business law firm, a member of an international multi-
disciplinary practice providing legal, accounting, assurance and tax services.

A Recent Shift towards More Administrative Controls in Large PSFs

Market and institutional pressures towards rationalized professional services
The market for professional services is becoming more competitive and deregulated, notably in
Europe where, in 2004, the European Commission initiated a dialogue with the professional bodies
of lawyers, notaries, engineers, architects, accountants, tax advisors and pharmacists, and with
national regulatory authorities, to eliminate some of the cross-border restrictions on professional
practice (European Commission 2005):

The European Parliament ... welcomes the dialogue between the Commission, the Member States and the
professional bodies of professional services providers aimed at dismantling barriers to competition which
are unjustified or harmful to the pursuit of the general interest and rules which are against the interests of
consumers. (European Parliament 2006)
Brivot 491

At the same time, recent financial scandals have challenged established notions of professional
responsibility. As Evetts (2006: 516) notes, ‘doctors, lawyers, scientists and many others are treated
with suspicion. … An increasingly litigious culture, fuelled by knowledge of large financial gains
from negligence cases in the USA, is further undermining trust and professionalism’. This climate
of mistrust has created new discourses that emphasize transparency and objectivity, which are
gradually replacing discourses of trust. Professionals are now required to be both transparent in
their dealings with clients and accountable for their performance (Kuhlmann and Burau 2008;
Evetts 2008; Champy 2006). In particular, there is a greater need to make explicit the ways in
which they use evidence and existing knowledge. At the same time, professions are facing intense
economic and institutional pressure to deregulate, rationalize and ‘managerialize’ their activities.
Such demand are not new (see, for instance, Porter 1995), but they have recently gained momentum,
resulting in the application of new managerial regimes within a number of PSFs, and the integration
of more explicit governance regimes into their organizational structure.
In the legal field, Dezalay (1992: 18) describes large, bureaucratized law firms as ‘law factories’,
which are manifestations of the supremacy of an American model over European legal craftsman-
ship: ‘because of the concentration of resources that these factories can mobilise, small practitioners
have no choice but merge or disappear, as the international competition [in the market for legal
services] intensifies’. Similarly, firms in the accountancy field have been moving away from the
‘Professional Partnership’ (P2) organizational model (Greenwood et al. 1990) – which is character-
ized by informal controls – in favour of the ‘Managed Professional Business’ (MPB) model (Cooper
et al. 1996), where administrative controls are systematized (Greenwood et al. 1998). In the latter,
controls are applied using management by objectives (MBO) techniques and explicit performance
indicators (Dirsmith et al. 1997), such as partners’ annual fee generation, the reporting of billable
hours, and the types of clients and engagements that should be given priority.
The professional healthcare services field has been subject to a similar phenomenon. Studies
by Ford and Angermeier (2008), Harrison and McDonald (2008), and Hunter (2006) document
the standardization of care, the rise of evidence-based medicine, and stricter clinical controls in
hospitals: ‘[t]he new health policies of managerialism, markets and consumer “choice” change
the substance of governance in different ways. ... One characteristic of the changes in governance
is the increasing turn towards managerial steering and performance that in turn shifts power to
the meso level of organisations and professions’ (Kuhlmann and Burau 2008: 620, 624).
In the following section, I describe the evolution of organizational controls in PSFs in the
context of extant market and institutional pressures for transparency, accountability and cost-
efficiency.

Organizational Controls in PSFs

Definition of ‘control’
Control can be defined as a form of influence which is intentionally or unintentionally exerted by
some individuals on themselves or on other individuals within a social group. This influence effec-
tively limits the freedom of those upon which it is exerted (Lebas 1980). In an organizational set-
ting, intentional control is exerted to ensure that members of an organization act in a coordinated
and cooperative fashion. However, as Lebas and Weigenstein (1986: 261) point out, ‘while the
necessity of control may be readily apparent, the route to control is less obvious. Every organisation
elects some combination of input control coupled with output control, and defines a general
approach to control which may be labelled “market,” “rules” or “culture” (although the approach is
rarely a pure version, but is rather adulterated by elements of the above three means of control)’.
492 Organization Studies 32(4)

Relevance of Hopwood’s (1974) typology of controls: social,

administrative and self-controls
Other descriptions of control within organizations have focused on the influence of ‘markets’,
‘bureaucracies’ or ‘clans’ (Ouchi 1980), or have variously classified internal controls as ‘cybernetic’
or ‘non-cybernetic’ (Hofstede 1981); ‘cultural’ or ‘bureaucratic’ (Child 1984); ‘actions’, ‘results’,
or ‘personnel’-oriented (Merchant 1982); or as based on ‘mutual adjustment’ (Perrow 1967). A
literature review undertaken by Chiapello (1996) identified six dimensions of organizational
control: (1) the means of control (market, organization, culture, or social relations); (2) control
processes (cybernetic or non-cybernetic); (3) the time at which controls are performed (before,
after or during action); (4) the controller’s identity (organization, person, group, or self); (5) the
subject of control (e.g. actions, results, individuals’ characteristics, the organization’s objectives
and strategy, organizational culture and context), and (6) the attitude of the person being controlled
(moral commitment, ‘working the system’, or alienation). Dambrin (2005) later proposed two
additional dimensions: the direction of control (lateral, ascendant or descendant) and the instru-
ments of control (observation and analysis, or orientation).
Notwithstanding this plethora of dimensions and constructs, there is no complete or defini-
tive typology of organizational controls. In this paper I adopt Hopwood’s (1974) model, which
proposes that organizational control is the unpredictable result of three competing spheres of
influence: (1) ‘self-controls’ – those informally exerted by each individual over his own behaviour;
(2) ‘social controls’ – those mutually and informally exerted by group members upon other
members’ behaviour; and (3) ‘administrative controls’ – the intentional and formally exerted
managerial power over subordinates’ behaviour. Although they often conflict, these three realms
of influence are not mutually exclusive and can co-exist within an organization, forming a
dynamic, unstable equilibria of ‘contradictory tensions’.
Hopwood’s (1974) typology of self, social and administrative controls is consistent with extant
research examining the form and management of PSFs (e.g. Brock 2007; Cooper et al. 1996;
Greenwood et al. 1990; Greenwood and Empson 2003; Hinings et al. 1999; Malhotra and Morris
2009; Pinnington and Morris 2002, Von Nordenflycht 2007),2 as well as sociological studies of
organizations employing professionals (Scott 2004). These studies frequently examine the tension
between formal control structures within PSFs and informal structures such as social obligations
and professional self-discipline. In addition, various organizational archetypes have been proposed
to characterize the distinctive organizational and governance arrangements of PSFs which empha-
size social and self-governance rather than formal rules and other kinds of administrative control
(see for example Scott’s (1965) ‘autonomous and heretonomous professional organizations’;
Bucher and Stelling’s (1969) ‘professional organizations’; Litwak’s (1961) and Mintzberg’s (1979)
‘professional bureaucracy’; Mills et al.’s (1983) ‘flexiform’ or Greenwood, Hinings and Brown’s
(1990) ‘professional partnership’). In the following section I set out the reasons why social and
self-controls have traditionally been seen as appropriate in professional work settings.

The growing emphasis on administrative controls in large PSFs

Professional work is frequently defined as ‘special’, with distinctive outputs and throughput
processes, a view which has contributed to the legitimation of professionals’ demands for self- and
social regulation, and their resistance to administrative controls (see, for instance, Wallace 1995).
For example, one of the defining features of professional work is that the quality of ‘outputs’
(Lebas and Weigenstein 1986) is difficult to define and to measure (Malhotra et al. 2006), whether
Brivot 493

by clients, regulators, or even professionals themselves. Audit work, for example, is traditionally
judged according to the independence and competence of the professionals who perform it (De
Angelo 1981), qualities which can only be inferred but not measured. Given the intangibility of
professional service quality, clients resort to imperfect assessment techniques, such as word-of-
mouth recommendations based on organizations’ reputations (Greenwood et al. 2005), media rank-
ings, and positive press coverage (Karpik 1989). Commentators such as Child (1984), Hofstede
(1981), Ouchi (1977) and Merchant (1982) have suggested that when the quality of outputs is not
measurable, administrative controls over such outputs are counter-productive, or at best useless.
Professional work is also characterized by an element of ‘co-production’, where clients work
with professionals to varying extents in order to create outputs (Mills et al. 1983). Some organiza-
tion theorists have argued that because professionals are dependent upon a client’s willingness or
ability to share key information relevant to the production process, they face a high degree of task
uncertainty and that administrative controls over the ‘throughput’ process are therefore irrelevant
(Chapman 1997).
Professionals’ production ‘technology’ (Perrow 1967)3 is also incompatible with administrative
controls. Professional work is frequently non-repetitive because many client questions are unique
(Pinnington and Morris 2002), and is traditionally seen as depending more upon practitioners’
personal attributes such as their intuition, aptitude, experience and discretion (Abbott 1988;
Freidson 1986) than on their compliance with predefined production standards. Professional work
would thus qualify as ‘non-routine work’ in Perrow’s view, to which administrative processual
controls are ill-suited.
Nonetheless, there is a growing emphasis upon administrative controls within professional
organizations, particularly within the largest PSFs which emerged after several decades of internal
growth and a succession of mega-mergers (Cooper et al. 2007). As discussed previously, this trend
for the direct regulation of professionals has also been encouraged by an institutional and regula-
tory environment that promotes bureaucratic control as a solution to the current confidence crisis
in professional self-governance.

Organic Knowledge Production, Sharing and Use in PSFs

Another distinctive characteristic of professional work is the (partially) organic nature of its knowl-
edge creation, sharing and use. Professionals cultivate a body of abstract and codified knowledge
as a resource for addressing client problems, and apply that knowledge using discretionary
judgement (Freidson 1986). Abbott (1988: 40) refers to this significant component of professional
work as ‘inference’,4 whereby abstract knowledge is linked with the particulars of a client’s
problem or situation.5 For example, doctors may allocate some aspects of a patient’s diagnosis
and treatment to non-professionals, but the interpretation of diagnostic results and the treatment
strategy – their inference – cannot be delegated to subordinates. Thus, although professional
knowledge is often codified, its abstract features mean that its adaptation to specific questions is
difficult to standardize and control, in opposition to the current bureaucratization trend in PSFs.
However, that is not to say that professional knowledge creation, sharing and use processes are
identical across professions and PSFs. For example, in some small, collegial-style PSFs there is
little or no attempt to ‘recycle’ existing knowledge from one client assignment to another, and
constant ‘reinvention of the wheel’ is a source of professional pride (Robertson et al. 2003). In
these firms, knowledge sharing is territorialized and restricted (Lazega 2001). In contrast, large
PSFs such as the ‘Big Four’ accounting firms may engage in ‘knowledge commodification’6 as a
strategy to colonize new jurisdictions and practice areas (Suddaby and Greenwood 2001: 933).
494 Organization Studies 32(4)

This relatively new approach to knowledge production, sharing and use is indicative of a shift
towards a more inclusionary intra-organizational perspective, which contrasts with the territorial,
exclusionary knowledge regimes of the past (Knorr Cetina 2006).
When the first KM systems became available in the early 1990s, organizational knowledge was
seen as any practice, experience or other know-how that had proved to be valuable or effective, and
might be applicable to other organizations (e.g. Nonaka 1994). Viewed as a commodity, knowledge
was a ‘strategic intangible asset’ which was sticky (Szulanski 2000), codifiable (Kogut and Zander
1992), and transferable (Grant 1996). Other commentators described it in terms of, inter alia, the
level at which it was held (individual or collective) (Malhotra 2003); whether it existed in a tacit or
explicit state (Polanyi 1969; Nonaka 1994); or the extent to which it was retained consciously or
unconsciously (Polanyi 1967).7 Specific systems – also known as ‘people-to-document knowledge
management systems’ (Hansen et al. 1999) – emerged to manage the process of its creation, dis-
semination and use. Since then, this idea of knowledge as an asset has been challenged by a more
existential view (Maturana and Verela 1998), whereby knowledge is indissociable from the
‘knower’. So-called ‘second generation KM systems’ are arguably more consistent with this
alternative view, and are designed to encourage the development of communities of practice and
to stimulate ‘knowing’ experiences, rather than merely facilitating transfers of knowledge.
The commoditization of knowledge so that it can be managed and controlled embodies one of
the many forms of bureaucracy; KM systems, however, are appealing to PSFs because they appear
to offer a measure of operational transparency, particularly in light of the fact that these organiza-
tions are under increasing regulatory pressure to improve their accountability (Kuhlman and Burau
2008; Olgiati 2008). Before KM systems were implemented ‘no one best way [was] predictable
from the formal body of knowledge itself’ (Freidson 1986: 217) because practitioners employed it
inconsistently. New methods of knowledge dissemination, coordination, translation and reuse have
since become commonplace, with numerous studies describing professionals’ acceptance or rejec-
tion of people-to-document KM systems as decision aids in the provision of services (e.g. Hsiao
et al. 2006; Werr and Stjernberg 2003). Whilst such systems are a clear attack upon professionals’
long-established autonomy and discretion, their use as bureaucratic control devices intended to
increase output standardization, reduce litigation risks and improve cost efficiency has received
scant academic attention. The purpose of this research is to find out whether the use of codified
people-to-document KM systems in bureaucratized PSFs contributes to a shift in power8 away
from professionals (i.e. reduced exercise of social and self-controls over knowledge production,
sharing and use) and towards managers (i.e. increased formal controls over knowledge production,
sharing and use).

Methodology
A number of prominent researchers have argued that qualitative research methodologies are best
suited to investigations of phenomena or topics for which little or no previous theory exists (e.g.
Barley 1990; Bouchard 1976; Eisenhardt 1989). For example, a grounded theory approach to
connect data to existing and suggestive new theory (Glaser and Strauss 1967) is appropriate for
exploratory research questions where data is collected and analysed ahead of the formation of
hypotheses. Such an approach advocates the simultaneous gathering of rich and detailed data
alongside its analysis – a process known as ‘abduction’ – such that ‘data analyses often alternate
and iterate with data collection processes’ (Edmondson and McManus 2007: 1163). The analysis
of material such as exploratory interviews, direct observations, archival data, and longitudinal
observations is a pertinent research strategy because it allows a detached investigation of a topic,
as well as providing the opportunity for triangulation.
Brivot 495

Given the lack of extant literature on the mutual influence of KM and organizational controls
in PSFs (Ditillo 2004), an exploratory socio-ethnographic approach was adopted and an abductive
analysis undertaken. Organizational socio-ethnographies permit full immersion in the research
field and involve the use of data collection and analysis procedures which follow grounded theory
principles. In addition, the adoption of such an approach (see Beaud and Weber 1997) implies
the existence of the following: (1) intimate personal knowledge of the subjects of the study;
(2) longitudinal rather than cross-sectional observation; and (3) sensitivity to the constructed
nature of the research results. Although the research strategy excludes the possibility of general-
izing the findings to other types of organization or contexts, it is believed that this limitation is
offset by the study’s ability to provide insights into an organizational process which is not well
understood, thereby providing directions for future research.

The Research Field

This study investigates the behaviour of business lawyers when creating, sharing and using
professional knowledge. According to Robertson et al. (2003: 835), professionals operating in the
legal field resort to methods ‘characterised by deduction from previous cases and precedents and
reinterpretation of existing judgments’. This reliance on documentation from previous cases in
forming judgements on new cases draws attention to the role of people-to-document KM systems
in the delivery of professional legal advice. Such systems seem compatible with lawyers’ ‘epis-
temic culture’ (Knorr Cetina 1999): ‘where knowledge is highly interpretable and contested,
knowledge creation is grounded within, and relies far more [than in other professional fields] upon
an explicit knowledge base articulated in text-based forms’ (Robertson et al. 2003: 852–3).
Data collection and analysis was carried out from 1999 to 2008 in the Parisian office of a large
PSF which was a member of an international network of accountants, lawyers and consultants. In
order to protect the anonymity of the subject organization, I refer to the office as ‘JurisParis’, and
its network member organization as ‘JurisFrance’.
JurisParis employs 250 lawyers (and close to 50 people in support functions such as IT, secre-
tarial services, and knowledge management), and is one of the principal offices of JurisFrance,
which employs some 2000 lawyers. Clients of JurisParis consist mainly of large French compa-
nies, many with foreign operations; often, these clients are referred to the office by colleagues in
correspondent firms within the global network in which JurisFrance is a member.
Legal work in JurisParis falls mainly into three categories: (a) answering questions from clients
(ranging from one-off engagements to long-term, ongoing client assistance over several years); (b)
the administrative management of client files (e.g. statements of time spent, production of invoices,
collection of payment, etc.); and (c) ‘development’, an activity aimed at improving the firm’s
infrastructure. Given the international orientation of its client list, many of JurisParis’ lawyers
specialize in legal and tax advice relevant to cross-border operations. This study examines (a), with
a particular emphasis upon the role (if any) of the firm’s KM system in the provision of tax and/or
legal opinion letters.
JurisParis has made a significant investment in knowledge management resources. Two people
are employed full-time to ensure the continuing maintenance, development, user training and
technical support of a range of information and communication technologies, including the firm’s
own KM system. Known in-house as ‘knowledge base’, the system was developed in 1999 using
Lotus Domino software and contains more than 20,000 documents. Using a search engine, users
can access legal documents from previous client engagements, such as contracts, opinion letters,
memoranda, presentations and reports. A number of these have been classified by an internal
standards committee as representing ‘best practice’.
496 Organization Studies 32(4)

Initial contact with the managing partner of JurisParis indicated that knowledge management is
considered to be an essential aspect of the firm’s survival, with the primary purpose of the KM
system to improve the technical quality of work. However, an analysis of the firm’s annual
business plans, and the firm’s annual reviews of its KM function for the period 1999–2008, indicates
that the purpose of the KM system has changed over time. Its initial objective has gradually been
supplanted by an emphasis upon productivity gains and cost efficiency.
The significance of the KM system within JurisParis is demonstrated by the surveillance
processes employed by the firm to ensure that it is used. Document downloads are monitored by
the firm’s Knowledge Manager and reported to the managing partner several times each month.
Use of the system is compulsory: firm policy dictates that if a lawyer does not use appropriate
templates and other relevant resources from the database they may be subject to disciplinary
proceedings and would not be covered by the firm’s liability insurance if a client were to sue for
negligent advice. Although client-initiated litigation is rare, its adverse consequences are a signifi-
cant motivation for compliance. In addition, lawyers’ annual assessments (which have a direct
impact on their career progression and salary) include criteria which take account of time allocated
to KM activities although, as the managing partner conceded, these criteria have only a negligible
effect on lawyers’ salaries.

Study Data
A number of data sources were subject to analysis:

• Interviews. I identified 12 tax and consultancy engagements which were representative of

the issues typically addressed by the firm. For each of these, I spoke with the lead partner,
principal manager, and the junior lawyer charged with researching the issues pertinent to the
eventual opinion letter or report. Other individuals were interviewed if they had a significant
role in the engagement. A total of 51 interviews with 41 employees were undertaken between
2005 and 2008. In addition, the managing partner of JurisParis was interviewed three times
during this period.
• Personal reflections. My own experience as the firm’s Knowledge Manager between 1999
and 2005.
• Electronic and archival documents. I examined the document downloads from the KM
system over a three-month period (February to April 2005), totalling 18,336 observations.
In addition, I obtained access to 30 archival documents from the period 1999–2008, including
model lawyer objective sheets for each hierarchical level of the firm, and knowledge
management activity reports prepared by the partners responsible for KM supervision.
• Non-participant observation. I attended seven meetings, during which discussions were
held regarding the selection of knowledge resources to be added to the KM system.

Analysis and Coding

Data sources were analysed and coded with NVivo 7. Consistent with an abductive analysis, codes
emerged and were revised as data was gathered, or were drawn from conceptual frameworks derived
from a literature review conducted both before and during the data collection. In particular, the
analysis was mindful of Orlikowski’s (1992) assertion that technology cannot have a social effect
unless it is utilized. She argues that users are creative and do not always follow a technology’s
prescribed uses. Rather, they are frequently agents of change who circumvent its prescribed uses or
Brivot 497

apply a technology in imaginative ways to unanticipated situations. As I describe below, the

purpose of the KM system as defined by the firm’s management committee was not always
consistent with its actual use by JurisParis’ lawyers.

Results
Uses of the KM system by lawyers in JurisParis
Data analysis revealed a typology of uses and non-uses of the KM system at JurisParis.

1.  To find ready-made answers to known questions of law.  Lawyers often search the KM database to
find documents which have been used in similar situations to the one they face, or to find out if a
specific legal question has previously been addressed by another lawyer. If a document or opinion
letter is relevant and appropriate to their client’s case, most of the lawyers interviewed for this
study said that they would reuse existing templates or answers rather than formulating solutions
from scratch

2.  To identify leads for developing answers to new questions of law.  Users who cannot find an appropri-
ate document or solution often decide to proceed on the basis that their query is a novel one, which
requires the preparation of an original answer. In such a situation, lawyers indicated that they fre-
quently turn to their personal network of legal contacts in order to generate leads for developing
their answers. In addition, they may use the KM system as a resource to identify those individuals
within the firm who may have the expertise to suggest possible routes to a solution.

3.  For self-training or future reference.  A third type of KM system use is self-training. Whilst search-
ing for ready-made answers or leads for new analyses, users often identify documents of interest
which, although not immediately relevant to their current search, may be of use in the future or as
a source of information about a point of law or practice. Users download these to their personal
libraries for later use or reference.9

4.  To observe and monitor peers’ work.  The author’s prior employment in the firm, and her close
relationship with some of its lawyers, resulted in confessions of a clandestine practice: scrutinizing
what colleagues were working on. More specifically, a number of lawyers admitted that they would
examine the documents entered into the KM system by a colleague in order to evaluate their
technical quality and whether they contained inaccuracies. This type of use is far removed from the
uses recommended or anticipated by the system designers or the managers of JurisParis. For some
lawyers, this practice was a way of monitoring the activities of their rivals, or a means of gathering
evidence to undermine them. Some users claimed that checking up on others was a way of policing
whether they were ‘poaching’ on technical fields outside of their recognized area of expertise or
declared specialization in the firm’s matrix structure.

5.  Use of the KM system – or deliberate non-use – to territorialize knowledge.  JurisParis maintains a
list of lawyers’ technical and sector-specific specializations, which is published in its client
brochure. However, the list is not static: a number of internal working parties meet regularly to
discuss the creation and colonization of new practice areas. When these groups agree that a corpus
of knowledge has been developed which is substantial enough to support the creation of a new
practice group, the group tends to claim exclusive ownership and use of the knowledge resources
498 Organization Studies 32(4)

associated with their sphere of expertise. This research reveals that, in terms of the use of the KM
system, members of newly-formed practice groups demonstrate one of two types of response. The
first is to put their specialist resources ‘on display’, with the aim of attracting recognition from
other members of the firm and securing their rights of ownership and exclusive use over this
knowledge. The second strategy is to boycott the KM system, i.e. restrict the use of their accumu-
lated knowledge resources only to those who claim expertise in that area. To prevent knowledge
from leaving the group’s ‘territory’, members do not contribute to the central KM system. This
results in the creation of a ‘black market’ in specialist knowledge, which passes between a select
group of lawyers but is outside the boundaries of the official knowledge market of the KM system.
Thus, whilst one type of behaviour is a territorialization strategy designed to protect knowledge
within a social niche, the other is a form of deterritorialization/reterritorialization whereby knowl-
edge is made accessible to the entire firm in order to demonstrate expertise of the author and to
establish his legitimacy.
Lawyers were also categorized by the frequency of their use of the KM system (moderate or
intensive user, or non-user), and on their expressed enthusiasm or hostility to it. Frequency of
individual use was assessed according to the number of documents downloaded by each lawyer,
as compared with the average number of downloads undertaken by their peers. Enthusiasm or
hostility towards the KM system was determined from analysis of interview data. Enthusiasts were
those who made no (or minor) criticism of the database (e.g. poor design, lack of functionality),
whereas hostile users were those who expressed more severe disparagement, such as questioning
the rationale for the system, or vilification of its use and effects.
In the following section I set out some of the implications of these results; in particular, I
consider how the various uses of the KM system, and individual variations in use and enthusiasm,
may affect the firm’s control over the production, sharing and use of knowledge.

The Evolving Nature of Knowledge Control

Control over the knowledge production process
Use of the KM system to find existing answers to common questions, or to identify a course for
tackling new ones, has resulted in the development of an unforeseen application: the ability of
users to check what others within the firm may have written on a particular question of law. The
systematic downloading and appraisal of documents created by colleagues encourages individual
professionals to align their legal opinions and advisory style with that of peers and seniors. Despite
the fact that some legal topics may be subject to varying and even contradictory interpretations,
widespread observation of what others have said results in a standardization of individual profes-
sional judgements across the firm. Such a result is congruent with the bureaucratic objective origi-
nally envisaged for the KM system: that professional knowledge production and delivery should
be standardized in order to improve cost-efficiency and productivity. However, this alignment of
legal opinion was not obtained by administrative controls but emerged as an unforeseen result of
individual lawyers’ use of the KM system.
Although administrative, coercive controls exist to compel the use of the KM system to produce
consistent legal advice, these have little or no force because JurisParis has no way of knowing
whether or not a lawyer has produced an original document or opinion letter, or has reused and
adapted an existing document available in the KM system. Lawyers themselves have internalized
the bureaucratic need to standardize their opinion letters and other documents, and to check what
others have written before formulating their advice. Their self-regulation has thus not lost ground
to the bureaucracy of the KM system. Indeed, lawyers’ control over the quality of their work has
Brivot 499

Table 1.  User profiles

Frequency Non-use Moderate use Intensive use

Attitude

Enthusiasm Prescriber Satisfied user Unconditional user

(2 interviewees) (14 interviewees) (12 interviewees)
Illustration with (Pierre, Partner): ‘Me? (Hélène, Junior): ‘Well, if (Frank, Senior Manager):
selected interview No. I never use it it’s something repetitive, ‘I generally find what I
excerpts directly. I ask [name well known and if there am looking for [in the
of a junior lawyer] to are models available, I KMS]. “Boiler-plate”
go and find stuff in it am not going to recreate clauses, for instance,
for me. Because, you things from scratch. That which we find in every
know, I always forget would be totally stupid, contract ... .You read
my password.’ and a waste of time … . them, and if they look all
The knowledge base right, all you have to do
is very important, you is copy and paste.’
know. When I did my job
interviews, it’s something
that I considered. Not
something fundamental
but important. Because
there are law firms where
you get nothing like this.’
Hostility Boycotter Frustrated user Schizophrenic user
(1 interviewee) (7 interviewees) (5 interviewees)
Illustration with (Bertrand, Tax (Henri, Director): ‘Am I (Sabine, Senior Manager):
selected interview Partner): ‘I have a user of the knowledge ‘I use certain things in
excerpts never ever opened base? Only under duress the knowledge base,
the database. You can … My feeling is that especially when I work
check it out if you this database is useless. with [name of a partner
want.’ … The whole thing is whom we did not
(Researcher): ‘But do messed up. … It was interview], but only
you prescribe its use supposed to be for things written by people
to the juniors with knowledge sharing but I know and trust. …
whom you work?’ he [the firm’s managing Because the database is
(Bertrand): ‘No. It’s partner] uses it for loaded with errors.You
their problem if they policing people.’ find lots of things that are
want to use it.’ not technically sound.’

become more sophisticated: it is no longer solely concerned with the self-discipline to behave
‘professionally’, but has been extended to include a concern not to engage in the unnecessary
duplication of extant knowledge. By not ‘reinventing the wheel’ for every client engagement,
professionals act to minimize the risk of technical errors appearing in work outputs, and ensure that
their time is utilized in the most cost-efficient way.
I also propose that lawyers’ creative use of the KM system for self-training or to ‘territorialize’/
‘deterritorialize’ their expertise has influenced how knowledge is produced and used. Individuals
no longer operate as ‘general practitioners’ but are increasingly creating specialized knowledge
spaces which become incorporated into the search algorithms and taxonomy of the KM system.
The bureaucratic thesis advocates that increasing specialization of a workforce creates productivity
improvements, yet administrative controls such as formal rewards and sanctions were not respon-
sible for the increase in lawyers’ specialization at JurisParis. Rather, it was their self-imposed
500 Organization Studies 32(4)

social controls and norms, played out in the arena of the KM system, which created social pressure
to maintain their specialist fields and not stray into those of their colleagues. This new form of
social control was revealed by lawyers’ tendency to observe and monitor peers’ work, and to check
that their contributions to the KM system were solely within a specific individual’s recognized area
of expertise. In this way, lawyers could police their own knowledge jurisdictions and deter others
from encroaching on them.
The KM system can also be used to assess the technical capabilities of others’ work, which adds
a layer of quality control to the firm’s professional output. Because the KM system makes errors
visible to the entire firm, lawyers attempt to produce flawless work so that their reputations are
not tarnished publicly. Again, the KM system operates indirectly to improve the quality and
consistency of the firm’s client service by encouraging professional self-monitoring without the
need to impose formal administrative controls.
Thus, although the KM system was implemented in order to increase bureaucratic controls over
professional work by improving lawyers’ productivity and cost effectiveness, these outcomes have
been achieved without applying administrative controls. Work standardization, specialization, and
technical reliability were achieved because new forms of social and self-regulation emerged as the
system was used, as demonstrated by lawyers’ anticipated and unanticipated uses of the KM
system. Professional social and self-controls have not been superseded by bureaucratic manage-
rial controls but have been reinvented and reformed in a new context.

Control of the knowledge sharing process

Before the implementation of the KM system, knowledge sharing occurred informally between
individuals and was determined to a large extent by their personal network ties and position with
the firm’s self-constituted social niches. Sharing now occurs (although not exclusively) via the
centralized system. Although the new technology has not eradicated the traditional distribution
channels, it has created new ones. Lawyers are now able to access knowledge despite a lack of
social capital. Maxine, for example, a newly recruited lawyer, used the KM system because she
was unable to consult personal contacts within the firm to advise her on matters of law:

Maxine: You think and think, you can’t see any answers so then you look into the base
[KM] to see whether someone has already written something that could help you
move ahead in your analysis. …
Interviewer: When you get stuck on a technical point, what’s your procedure, to look in the
base [KMS] first or to ask colleagues for assistance?
Maxine: I do sometimes look in the base before going to see the others. Yes, of course I
do. Asking the others is really … well I think the idea is stupid, but there you
go … it’s really the last resort.

Other lawyers are keen to demonstrate their personal resourcefulness and research abilities,
without the aid of colleagues. As Matthieu explained: ‘My job, you know, my job is to use my
brains to find solutions to problems. It’s not to ask someone else or the knowledge base to do the
job for me!’
Use of the KM system has allowed lawyers to bypass established and exclusive social niches in
their search for expert knowledge. Yet the findings of this study indicate that the creation of a
centralized knowledge-sharing system need not be accompanied by a greater emphasis on admin-
istrative controls. I find that professionals are inclined to regulate themselves, and to apply social
Brivot 501

sanction to those who do not agree to share their knowledge resources. James, the firm’s managing
partner, explained that he was confident that a ‘natural’ peer pressure mechanism would eventually
isolate individuals who did not conform to the new regime. He also acknowledged that self-
monitoring and regulation of the system by professionals had drawbacks, notably some feelings of
inequity about the relative contributions to the database by some of its contributors and users:

The types of people who are most involved in the commercial side and not much in research always
complain that this new mutualization works against them. It’s only logical after all, because they don’t
have great awareness of the resources they use and the investment of capital needed to produce these
resources. You see the same thing in other professions ... . In contrast, specialists are very sensitive to the
looting of their work. They consider that when they’ve developed a product, it should be their job to sell
it. … I think they are against reuse by anyone other than themselves. If I carry out a study and no one uses
it, I grumble about it. If I carry out a study and I’m told it has to be used by everyone and I grumble about
that too, then we’re at a dead end!

In summary, I find that the knowledge-sharing process at JurisParis is now partially bureaucra-
tized, in that it is centralized, searchable and subject to scrutiny by committees of experts who
decide on what is ‘best practice’. Although ‘black markets’ for knowledge exist, they do not
prevent the majority of tax and legal opinion letters from being shared via the central KM system.
This new, bureaucratized sharing process did not require the imposition of increased administra-
tive controls; rather, professionals have remained in direct control of their knowledge resources,
and continue to decide what they share or do not share. I suggest that most lawyers use the KM
system – even if they fundamentally disagree with the idea of centralized knowledge sharing
(see Table 1, ‘Hostile Users’) – because new forms of social and self-regulation have evolved to
produce a set of institutionalized behavioural norms. If a professional does not play by the new
rules and avoids posting their outputs to the central KM system, he is perceived by his peers as
being a mere consumer of others’ ideas. To prove their competence, lawyers are expected to
display their work; to refuse to do so is suspicious and attracts disapproval. My analysis found
only one lawyer who refused to use the KM system at all (see Table 1, ‘Boycotter’), and even
those who were most hostile to the system were occasional users (see Table 1, ‘Frustrated’ and
‘Schizophrenic’ profiles). This suggests that resistance to knowledge sharing is isolated.

Control of the knowledge use process

The reuse of personal documents and opinion letters has always happened within JurisParis, but the
new KM system means that lawyers can reuse resources produced by others, without having been
involved in their production or knowing the authors. The firm’s managing partner explained that
he wanted work outputs to be consistent and of high technical quality, and for lawyers to avoid
duplicating existing documents and research efforts; however, these largely bureaucratic objectives
were achieved without applying administrative controls over the use of the KM system. Although
there appeared to be a type of formal, bureaucratic sanction in place, i.e. the threat of personal legal
liability for lawyers who did not use the firm’s authorized templates and other knowledge resources,
this threat did not materialize during the observation period of this study (almost ten years), and
may not have exerted any influence upon the behaviour of lawyers at JurisParis. In effect, individu-
als could decide for themselves if and how they wanted to use the KM system. Nevertheless, most
indicated that they felt compelled to reuse existing documents and advisory opinions, even if they
disagreed with the very idea of doing ‘copy-and-paste’ jobs. A tax partner, for example, explained
that reusing materials from past engagements was one of the least satisfying aspects of his work.
502 Organization Studies 32(4)

At the same time, he felt that the firm was making progress in its approach to producing ‘standard
types of engagements’:

Hyppolite: My job is to analyse a given situation, and to be able to say, ‘here is what you
could do, here are the possibilities that you have’. Given the constraints of a
particular client question, we have to find the solution that would be most suit-
able. That’s what I like about my job. That’s what I find interesting. Currently, in
our profession, law firms like ours … you know … we are transforming
ourselves into mere suppliers of products, of pre-packaged solutions, and we
increasingly use ordinary commercial techniques.
Interviewer: Since when did you notice this evolution?
Hyppolite: About five years ago. And it is an irreversible phenomenon. In fact, in France we
are even lagging behind on this.
Interviewer: Was this evolution imposed by the market, do you think, or was it the result of
some sort of internal strategy, at JurisParis?
Hyppolite: Well, both. Clients find their interest in all this. They get more for less. We have
no choice. What is sad is that, when you write, you sharpen your thoughts; the
drafting phase is of primary importance because it is by writing your conclusions
that you realize what the weak points are. If all you do is just copy-and-pastes,
you miss that. … Personally, I rarely use the knowledge base. I never find any-
thing in it. I ask the young ones and they find things, most of the time. Including
my own work, things that I had forgotten! … On the standard types of engage-
ments, we have made lots of progress.

The rationale behind reusing existing knowledge includes improving cost-efficiency, produc-
tivity and, ultimately, profits. However, client invoices and calculations of an engagement’s
profitability (‘realization rate’) rarely took account of the costs of the initial production of
resources downloaded and reused in the production of an opinion letter or project document.
Indeed, productivity gains achieved via the large-scale reuse of documents contained in the KM
system are essentially redistributed to clients free of charge, since most clients are invoiced on
the basis of ‘billable hours’ spent by the lawyers involved on that particular engagement.
The other rationale for reusing existing knowledge and standard form templates is that this
reduces the risk that technical errors are introduced into documents as they are produced afresh.
However, some users were sceptical about the technical reliability of some of the documents and
solutions in the database. A typical concern shared by the majority of lawyers was doubt as to
whether the tax and legal advice found in the KM system had been proven. These misgivings call
into question the ability of the database to provide reliable, high quality knowledge solutions.
My analysis of the usage statistics for the KM system, and of the comments of interviewees,
suggests that knowledge reuse has become more systematized since the system was introduced,
particularly amongst lawyers with less than five years of experience.10 However, the efficient and
methodical reuse of best practice solutions did not result from lawyers’ fear of administrative sanc-
tions. Most had internalized the need to be more productive and cost-efficient, and were using the
KM system regardless of their level of enthusiasm or hostility towards it (see Table 1).

Conclusion and Discussion

This study has examined whether the use of codified people-to-document KM systems in bureau-
cratized PSFs has caused a shift in power away from professional self-regulation and monitoring
Brivot 503

in favour of increased administrative controls over the creation, sharing and use of knowledge. A
socio-ethnographic study of a large Parisian law firm indicated that using a KM system contributes
to a bureaucratization of knowledge production, sharing and use processes, but is not accompanied
by a concomitant shift in the balance of controls within the firm. Tighter administrative controls
were not necessary. Far from losing ground to bureaucratic rules, professionals adopted self-
imposed controls to regulate the use of knowledge within the firm. They were able to align their
behaviour with managerial goals intended to enhance transparency, accountability and cost-
efficiency, but at the same time retain their independence within a bureaucratized setting.
This research makes three important contributions. First, it adds to the professional-bureaucratic
conflict debate by suggesting that professionals, who have historically resisted any bureaucratic
constraint (Raelin 1985), are now, in certain settings, actively participating in the bureaucratization
of their own knowledge production, sharing and use. Given that the success of people-to-document
KM projects depends upon professionals’ willingness to codify and share their work outputs widely –
which may diminish their professional power – why do they actively embrace such systems? Why
are they promoting the very mechanisms that might limit their autonomy?
One plausible explanation could be that the bureaucratization of knowledge production, sharing
and use actually facilitates professional power. In particular, transparency might assist in address-
ing increasing public mistrust of professional judgements, which have been accused of being
subjective, arbitrary and untrustworthy. By adopting a more systematic, scientific approach to
problems, and producing more consistent solutions to the same problem, professionals’ work
becomes more reliable and auditable, which is likely to increase public confidence in profession-
als’ work. This suggestion concurs with the views of Bastard et al. (2005) and Castel and Merle
(2002), who claim that standards, norms and other administrative constraints, which initially
appear to be rationalizations or bureaucratic emanations limiting professional discretion and auton-
omy, are actually resources that can be used to improve the reputation of professionals. In addition,
if KM systems could be used to find solutions which could be sold to clients, then professionals’
powers are in fact increased rather than diluted. If so, one could argue that bureaucracy does not
always crush individual agents’ powers and creativity, but rather it redefines where their powers
and creativity can be applied.
Further, I note that a shift away from the old ‘exclusionary regime of knowledge’ (Knorr
Cetina 2006) toward a more inclusive, accessible system is of benefit to those previously excluded
from accessing knowledge resources. Individuals formerly disenfranchised by a lack of social
capital or network associations can now use KM systems to access powerful knowledge resources
and to demonstrate their own abilities to create and colonize knowledge. Given the significance
of this revolution for both the old guard and the new generation, it is not surprising that this power
reconfiguration finds both supporters and opponents.
Another possible explanation for professionals’ active participation in the bureaucratization of
knowledge production, sharing and use is that their acceptance of such projects might not trigger
the classic professional-bureaucratic conflict that the extant literature proposes. It is possible that
KM codification projects can be implemented in such a way as to satisfy both organizational and
individual goals, as suggested by Morris (2001), or that professionals are (perhaps dangerously)
unaware of conflicts between professional and bureaucratic objectives and values (Suddaby et al.
2009). The extent to which self- and peer-imposed control mechanisms are resisted or not, and
the extent to which professionals unwittingly or consciously accept the principles of bureaucratic
control, remain to be determined in future studies.
A second contribution from the results of this research is to expand upon current conceptualiza-
tions of organizational control archetypes. Extant literature suggests that administrative controls
are necessarily formal and cybernetic, and always use clear reward and sanction procedures. In this
504 Organization Studies 32(4)

study, however, an increase in the appearance of administrative control (e.g. through repeated
verbal injunctions, by the firm’s managing partner, to re-use best practice solutions to known
questions of law) was observed, but no formal sanctions or clear reward system was in place.
Despite their reported personal preference for autonomy and discretion in the performance of their
work, most professionals felt compelled to change their behaviour in relation to the production,
sharing and use of knowledge. Should one infer that a façade of administrative control can be
effective in encouraging certain desired behaviours? Further investigations are needed. The results
of this study suggest that the bureaucratization of PSFs influences not only their internal mix of
self, social and administrative controls, but also the content of each of these three modes of control.
Studying how each control vector changes over time opens up new research possibilities. More
work is also necessary to understand how and why the targets of organizational supervision become
either consciously or inadvertently involved in the structuring of their own surveillance.
Thirdly, the results of this study contribute to the organizational archetype literature by suggest-
ing that the bureaucratization of professional firms and the imposition of administrative controls are
not necessarily linked. This research found that the control of knowledge production, sharing and
use at JurisParis remained largely informal, despite the application of bureaucratic processes. This
may imply that PSFs might not need more explicit regimes of governance and control, even if they
are attempting to increase their productivity and cost efficiency and to become more ‘managerial’.
More research is needed to answer this question.
This study suggests that new regimes of professional knowledge production, sharing, use and
control are emerging in some PSFs. These regimes are facilitated by technology, but professionals
still play the most significant role in creating, applying and distributing knowledge resources.
Even though work environments are becoming more rationalized and mechanized, the essence of
professional work is still the largely intangible application of individual creativity, experience
and judgement. Because professionals retain their autonomy and discretion, they have remained
free to devise novel and unanticipated uses for the technology. Moreover, management instru-
ments such as KM systems are arguably facilitating professional work rather than directing it,
which has produced benefits in terms of quality, productivity and efficiency for both individual
professionals and PSF managers. This study seems to indicate that professional work is trans-
forming itself and that ‘decustomization’ – in the sense of recycling existing knowledge by cre-
atively adapting it to novel contexts – is a form of behaviour which has previously been overlooked
in PSF research. But given that decustomization is now so prevalent in large PSFs, can they still
be defined as organizations that ‘apply complex knowledge to non-routine problems’ (Morris and
Empson 1998: 610)? More broadly, will such decustomization lead to ‘fissuring the professional-
knowledge-power knot’ (Kuhlmann and Burau 2008) in society? The bureaucratization of
PSFs presents a number of opportunities for scholarly research and debate regarding the defini-
tion of professional work, and it remains to be seen how well professionals maintain their
autonomy and power whilst dealing with bureaucratic demands to produce better, cheaper and
more consistent services.

Acknowledgements
I thank Editor-in-Chief David Courpasson, senior editor Catherine Paradeise, and three anonymous OS
reviewers, who helped me to significantly improve this paper. I also thank Samantha Fairclough for her
invaluable assistance in clarifying my arguments and writing this paper in proper English. In addition, I am
very grateful to Royston Greenwood, Tim Morris, Laura Empson, and Christopher McKenna for their useful
comments on earlier versions of this paper.
Brivot 505

Notes
  1. First generation KM systems are also referred to as ‘people-to-document’ KM systems in the literature
(Hansen et al. 1999).
  2. Other researchers who have analysed the evolution of control types in the context of professional
services have used similar typologies. For example, Kuhlmann and Burau (2008) used the classifica-
tion ‘hierarchy, network and self-regulation’, which is very similar to Hopwood’s (1974) scheme.
  3. Production technology refers to ‘the actions that an individual performs on an object, with or without the
aid of tools or mechanical devices, in order to make some change in that object. The object ... may be a
living being, human or otherwise, a symbol or an inanimate object’ (Perrow 1967: 195–6).
  4. Abbott (1988: 40) explains that professional work always consists of three phases: ‘diagnosis, inference
and treatment.’
  5. Abbott (1988) prefers to use the expression ‘professional knowing’, rather than ‘professional knowl-
edge’, since the latter implies that knowledge is reified but the former better suggests its transitory nature
and its continual enactment and re-enactment each time it is applied. However, for the sake of readability
and simplicity, this paper uses the term ‘professional knowledge’ consistent with its common-sense
meaning.
  6. Note that Suddaby and Greenwood (2001) refer to ‘management knowledge’ commodification in their
paper, rather than ‘professional knowledge’ commodification.
  7. An ongoing debate in the knowledge management literature concerns the definition of organizational
knowledge (Hsiao et al. 2006). Is organizational knowledge a commodity, a mode of cognition or a
capability? This paper does not attempt to resolve this epistemological dispute. Rather, it analyses the
use of first-generation KM systems, which view organizational knowledge as a commodity. Consist-
ently, ‘knowledge’ is referred to as the codifiable output of a learning process. The analytical dichotomy
between knowing as a process and knowledge as an output of that process does not necessarily imply an
ontological separation between the two.
  8. Freidson (1986: 213), citing Wilding (1982), identifies various areas in which power can be exercised
by professionals: ‘[t]here is power in policy making and administration, power to define public needs
and problems, power in resource-allocation, power over clients, and power to control work’. Because
ordinary members of the professions – such as the lawyers interviewed for this study – are not typically
involved in policy-making, administration or defining public needs, this study focuses on professionals’
power to exclusively control their work outputs.
  9. Constructing a personal database from the KM system distorts system use statistics, because once a
downloaded document is stored on a local hard disk, it can be reused any number of times without that
use being acknowledged and measured.
10. In 2005, the average number of documents downloaded per lawyer per month was 38 for juniors, 77 for
seniors, 50 for supervisors, 20 for managers, 7 for senior managers, 4 for directors and 5 for partners.

References
Abbott, A. (1988). The system of professions: An essay on the division of expert labor. Chicago: University
of Chicago Press.
Barley, S. R. (1990). Images of imaging: Notes on doing longitudinal field work. Organization Science, 1,
220–247.
Bastard, B., Mouhanna, C., & Ackermann, W. (2005). Une justice dans l’urgence. Le traitement en temps reel
des affaires pénales. Paris: Ministère de la justice, Mission Droit et recherché.
Beaud, S., & Weber, F. (1997). Guide de l’enquête de terrain. Paris: La Découverte.
Bouchard, T. J. (1976). Field research methods: Interviewing, questionnaires, participant observation,
systematic observation, unobtrusive measures. In M. D. Dunnette (Ed.), Handbook of industrial and
organisational psychology (pp. 363–413). Chicago: Rand McNally.
Brock, D. (2007) The reconstructed professional firm: A reappraisal of Ackroyd and Muzio. Organization
Studies, 29 (1), 145–149.
506 Organization Studies 32(4)

Bucher, R., & Stelling, J. (1969). Characteristics of professional organisations. Journal of Health and Social
Behaviour, 10, 3–15.
Castel, P., & Merle, I. (2002). Quand les normes de pratiques deviennent une ressource pour les médecins.
Sociologie du Travail, 44 (3), 337–355.
Champy, F. (2006). Professional discourses under the pressure of economic values: The case of French
architects, landscape and industrial designers. Current Sociology, 54, 649–661.
Champy, F. (2009). La sociologie des professions. Paris: Presses Universitaires de France.
Chapman, C. (1997). Reflections on a contingent view of accounting. Accounting, Organisations and Society,
22 (2), 189–205.
Chiapello, E. (1996). Les typologies des modes de contrôle et leurs facteurs de contingence: un essai
d’organisation de la littérature. Compabilité, Contrôle, Audit, 2, 51–74.
Child, J. (1984). Organisation: A guide to problems and practice. London: Paul Chapman.
Cooper, D., Hinings, R., Greenwood, R., & Brown, J. (1996). Sedimentation and transformation in organi-
zational change: The case of Canadian law firms. Organization Studies, 17, 623–647.
Cooper, D., Greenwood, R., & Hinings, R. (2007). Knowledge management in global accounting firms. In
M. Grandlund (Ed.), Total quality in academic accounting: Essays in honour of Kari Lukka. Turku:
Publications of Turku School of Economics and Business Administration.
Dambrin, C. (2005). Le controle a distance ou l’autocontrole par les technologies: Le cas des commerciaux.
Paris, Universite Paris Dauphine, Doctorate Dissertation in Administration Sciences.
De Angelo, L. E. (1981). Auditor size and audit quality. Journal of Accounting and Economics, 3, 183–199.
Dezalay, Y. (1992). Marchands de droit. Paris: Fayard.
Dirsmith, M., Heian, J., & Covaleski, M. (1997). Structure and agency in an institutionalised setting: The
application and social transformation of control in the Big Six. Accounting, Organisations and Society,
22 (1), 1–27.
Ditillo, A. (2004). Dealing with uncertainty in knowledge-intensive firms: The role of management control
systems as knowledge integration mechanisms. Accounting, Organisations and Society, 29, 401–421.
Edmonson, A., & McManus, S. (2007). Methodological fit in management field research. Academy of
Management Review, 32 (4), 1155–1179.
Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review,
14, 532–550.
European Commission (2005). Communication from the Commission to the Council, the European Parliament,
the European Economic and Social Committee and the Committee of the Regions – Professional services –
Scope for more reform, Brussels, 5 September 2005 COM(2005) 405 final: 1–11.
European Parliament (2006). European Parliament resolution on follow-up to the report on Competition in
Professional Services – Procedure number: 2006/2137(INI) – OJ C 308E, 16 December 2006: 178–181.
Evetts J. (2006). Introduction: Trust and professionalism: Challenges and occupational changes. Current
Sociology, 54 (4), 515–531.
Evetts, J. (2008). Introduction: Professional work in Europe. European Societies, 10 (4), 525–544.
Ford, R., & Angermeier, I. (2008). Creating a learning health care organisation for participatory management:
A case analysis. Journal of Health Organisation and Management, 22 (3), 269–293.
Freidson, E. (1986). Professional powers: A study of the institutionalization of formal knowledge. Chicago:
University of Chicago Press.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research.
New York: Aldine.
Grant, R. (1996). Toward a knowledge-based theory of the firm. Strategic Management Journal, 17, 109–122.
Greenwood, R., & Empson, L. (2003). The professional partnership: Relic or exemplary form of governance?
Organization Studies, 24 (6), 909–933.
Greenwood, R., Hinings, R., & Brown, J. (1990). The P2 form of strategic management: Corporate practices
in the professional partnership. Academy of Management Journal, 33, 725–755.
Greenwood, R., Rose, T., Cooper, D., Hinings, R., & Brown, J. (1998). The global management of profes-
sional services: The example of accounting. In S. R. Clegg, E. Ibarra, & L. Bueno (Eds.), Theories of the
management process: Making sense through difference. Thousand Oaks: SAGE.
Brivot 507

Greenwood, R., Li, S., Prakash, R., & Deephouse, D. (2005). Reputation, diversification, and organisational
explanations of performance in professional service firms. Organization Science, 16 (6), 661–673.
Hansen, M., Nohria, N., & Tierney, T. (1999). What’s your strategy for managing knowledge? Harvard
Business Review, 77 (2), 106–116.
Harrison, S., & McDonald, R. (2008). The politics of healthcare in Britain. London: SAGE.
Hinings, C., Greenwood, R., & Cooper, D. (1999). The dynamics of change in large accounting firms. In
D. Brock, C. Hinings, & M. Powells (Eds.), Restructuring the professional organisation (pp. 131–153).
London: Routledge.
Hofstede, G. (1981). Management control of public and not for profit activities. Accounting, Organisations
and Society, 6 (3), 193–211.
Hopwood, A. (1974). Accounting and human behaviour. London: Haymarket.
Hsiao, R., Tsai, S., & Lee, C. (2006). The problems of embeddedness: Knowledge transfer, coordination and
reuse in information systems. Organization Studies, 27 (9), 1289–s1317.
Hunter, D. J. (2006). From tribalism to corporatism: The continuing managerial challenge to medical
dominance. In: D. Kelleher, J. Gabe, & G. Williams (Eds.), Challenging medicine. London: Routledge.
Karpik, L. (1989). L’économie de la qualité. Revue Française de Sociologie, XXX, 187–210.
Karpik, L. (1995). Les avocats. Entre l’état, le public et le marché. Paris: Gallimard.
Knorr Cetina, K. (1999). Epistemic cultures: How the sciences make knowledge. Cambridge, MA: Harvard
University Press.
Knorr Cetina, K. (2006). Knowledge in a knowledge society: Five transitions. Knowledge, Work and Society,
4 (3), 23–41.
Kogut, B., & Zander, U. (1992). Knowledge of the firm, combinative capabilities and the replication of
technology. Organisation Science, 3 (3), 383–397.
Kuhlmann, E., & Burau, V. (2008). The healthcare state in transition. European Societies, 10 (4), 619–633.
Lazega, E. (2001). The collegial phenomenon: The social mechanisms of cooperation among peers in a
corporate law partnership. Oxford: Oxford University Press.
Lebas, M. (1980). Toward a theory of management control: Organisational process, information economics,
and behavioral approaches. Jouy-en-Josas: Cahiers de Recherche HEC.
Lebas, M., & Weigenstein, M. (1986). Management control: The roles of rules, markets and culture. Journal
of Management Studies, 23 (3), 259–272.
Litwak, E. (1961). Models of bureaucracy which permit conflict. American Journal of Sociology, 67,
177–184.
Malhotra, N. (2003). The nature of knowledge and the entry mode decision. Organization Studies, 24,
935–959.
Malhotra, N., Morris, T. & Hinings, R. (2006). Variation in organisational form among professional service
organisations. Research in the Sociology of Organisations, 24, 171–202.
Malhotra, N., & Morris, T. (2009). Heterogeneity in professional service firms. Journal of Management
Studies, 46, 895–922.
Maturana, H., & Varela, F. (1998). The tree of knowledge: The biological roots of human understanding.
Boston: Shambhala.
Merchant, K. (1982). The control function of management. Sloan Management Review (summer), 43–55.
Mills, P., Hall, J. Leidecker, J., & Margulies, N. (1983). Flexiform: A model for professional service organisa-
tions. Academy of Management Review, 8 (1), 118–131.
Mintzberg, H. (1979). The structuring of organizations. Englewood Cliffs: Prentice Hall.
Mitchell, J. C. (1983). Case and situation analysis. Sociological Review, 31, 186–211.
Morris, T. (2001). Asserting property rights: Knowledge codification in the professional service firm. Human
Relations, 54 (7), 819–838.
Morris, T., & Empson, L. (1998). Organisations and expertise: An exploration of knowledge bases and the
management of accounting and consulting firms. Accounting, Organisations and Society, 23, 609–624.
Nonaka, I. (1994). A dynamic theory of organisational knowledge creation. Organization Science, 5, 14–37.
Olgiati, V. (2006). Shifting heuristics in the sociological approach to professional trustworthiness: The
sociology of science. Current Sociology, 54 (4), 533–547.
508 Organization Studies 32(4)

Olgiati, V. (2008). The European learned professions and the EU higher education project. European Societies,
10 (4), 545–565.
Orlikowski, W. (1992). The duality of technology: Rethinking the concept of technology in organisations.
Organization Science, 3 (3), 398–427.
Ouchi, W. G. (1977). The relationship between organisational structure and organisational control.
Administrative Science Quarterly, 22 (1), 95–113.
Ouchi, W. G. (1980). Markets, bureaucracies and clans. Administrative Science Quarterly, 25, 129–141.
Perrow, C. (1967). A framework for the comparative analysis of organisations. American Sociological Review,
32, 194–208.
Pinnington, A., & Morris, T. (2002). Transforming the architect: Ownership form and archetype change.
Organization Studies, 23, 189–210.
Polanyi, M. (1967). The tacit dimension. London: Routledge.
Polanyi, M. (1969). Knowing and being. Chicago: University Press of Chicago.
Porter, T. (1995). Trust in numbers: The pursuit of objectivity in science and public life. Princeton: Princeton
University Press.
Rabardel, P. (1995). Les hommes et les technologies. Approche cognitive des instruments contemporains.
Paris: Armand Colin.
Raelin J. (1985). The basis for the professional’s resistance to managerial control. Human Resource
Management, 24 (2), 147–175.
Robertson, M., Scarbrough, H., & Swan, J. (2003). Knowledge creation in professional service firms:
Institutional effects. Organization Studies, 24, 831–857.
Scott, R. (1965). Reactions to supervision in a heteronomous professional organisation. Administrative
Science Quarterly, 10, 65–81.
Scott, R. (2004). Reflexions on a half-century of organisational sociology. Annual Review of Sociology, 30,
1–21.
Suddaby, R., & Greenwood, R. (2001). Colonizing knowledge: Commodification as a dynamic of jurisdic-
tional expansion in professional service firms. Human Relations, 54, 933–953.
Suddaby, R., Gendron, Y., & Lam, H. (2009). The organisational context of professionalism in accounting.
Accounting Organisations and Society, 34 (3–4), 409–427.
Szulanski, G. (2000). The process of knowledge transfer: A diachronic analysis of stickiness. Organisational
Behavior and Human Decision Processes, 82 (1), 9–27.
Von Nordenflycht, A. (2007). Is public ownership bad for professional service firms? Ad agency ownership,
performance and creativity. Academy of Management Journal, 50 (2), 429–445.
Von Nordenflycht, A. (2010). What is a professional service firm?: Toward a theory and taxonomy of
knowledge-intensive firms. Academy of Management Review, 35 (1), 155–174.
Wallace, J. (1995). Organisational and professional commitment in professional and nonprofessional
organisations. Administrative Science Quarterly, 40, 228–255.
Werr, A., & Stjernberg, T. (2003). Exploring management consulting firms as knowledge systems.
Organization Studies, 24, 881–959.
Wilding, P. (1982). Professional power and social welfare. London: Routledge.
Wilensky, H. (1964). The professionalization of everyone? The American Journal of Sociology, 70 (2),
137–158.

Author Biography
Marion Brivot is assistant professor of accountancy at the John Molson School of Business of Concordia
University in Montreal. Her research interests include social and behavioural aspects of accounting and
management control systems in professional service firms. This paper is based on her thesis, which studies the
implications of using knowledge management systems for the control of professional work in law firms.