Knowledge, Knowledge Work and

Organizations: An Overview and

Interpretation*
Frank Blackler

Abstract

Frank Blackler There is current interest in the competitive advantage that knowledge may
The Management provide for organizations and in the significance of knowledge workers, organ-
School, Lancaster izational competencies and knowledge-intensive firms. Yet the concept of
University, U.K. knowledge is complex and its relevance to organization theory has been insuf-
ficiently developed. The paper offers a review and critique of current
approaches, and outlines an alternative. First, common images of knowledge
in the organizational literature as embodied, embedded, embrained, encultured
and encoded are identified and, to summarize popular writings on knowledge
work, a typology of organizations and knowledge types is constructed. How-
ever, traditional assumptions about knowledge, upon which most current
speculation about organizational knowledge is based, offer a compartmental-
ized and static approach to the subject. Drawing from recent studies of the
impact of new technologies and from debates in philosophy, linguistics, social
theory and cognitive science, the second part of the paper introduces an altern-
ative. Knowledge (or, more appropriately, knowing) is analyzed as an active
process that is mediated, situated, provisional, pragmatic and contested. Rather
than documenting the types of knowledge that capitalism currently demands
the approach suggests that attention should be focused on the (culturally
located) systems through which people achieve their knowing, on the changes
that are occurring within such systems, and on the processes through which
new knowledge may be generated.

Descriptors: activity theory, knowledge, knowledge-intensive firms,

knowledge work, organizational competencies, organizational learning

Introduction

Ever since Galbraith (1967) suggested that a powerful new class of

technical-scientific experts was emerging, and Bell (1973) proposed that
knowledge is a central feature of post-industrial societies, the signific-
ance of experts in contemporary society has attracted much comment

(see Reed 1991 for a discussion of contemporary trends). Indeed, in

recent years, the importance of expertise for competitive advantage has
been emphasized again by economists and business strategists who have
suggested that wealth creation is less dependent on the bureaucratic
1022

control of resources than it once was, and more dependent on the exer-
cise of specialist knowledge and competencies, or the management of
I
organizational competencies (e.g. Prahaled and Hamel 1990; Hague
1991; Reich 1991; Drucker 1993; Florida and Kenny 1993). This debate =
has found echoes in discussion about ‘knowledge-intensive firms’, that
is, organizations staffed by a high proportion of highly qualified staff
who trade in knowledge itself (Starbuck 1992, 1993; Alvesson 1993a),
[I
in the suggestion that organizational competencies can be nurtured by =

the development of inter-organizational links (Kanter 1989; Badaracco

1991; Wikstrom and Normann 1994), and in the proposal that, because
of technological changes, team organization is becoming of crucial
importance and employees generally should be managed as ’knowledge
workers’ (Zuboff 1988).
Within the literature on the established professions the privilege sug-
gested by the term ’knowledge’ and the opportunities it offers occupa-
tional groups to protect their positions and ’black box’ their skills (for
example, by claiming the authority of medicine, law, or other complex
bodies of knowledge) have been well documented (e.g. Baer 1987;
Abbott 1988). Writing in a special edition of the Journal of Manage-
ment Studies on knowledge work Alvesson ( 1993a) notes how special-
ists in the new generation of knowledge firms are, in exactly the same
way, attracted to the mystique associated with the terms such as know-
ledge and knowledge worker; knowledge-intensive firms are, above all
else he suggests, systems of persuasion. Developing a similar point
Knights, Murray and Willmott (1993) suggest that the growing use of
such terms may be regarded as as normalizing discourse which, as it
legitimates a particular division of labour, distracts attention from the
knowledge that is an essential characteristic of all forms of
activity.
This paper explores the relevance of the terms knowledge, knowledge
work and knowledge-intensive firms for organization studies, by devel-
oping an approach which seeks neither to perpetuate the mystique often
associated with abstract, codified knowledge nor to present claims to
knowledge merely as normalizing discourse. Conventional images of
knowledge within the literature on organizational learning are first iden-
tified and are distinguished by the assumptions they make about the
location of knowledge, i.e. in bodies, routines, brains, dialogue or sym-
bols. Recent commentary on the emerging significance of knowledge
work amounts to the suggestion that, in place of a strong reliance on
knowledge located in bodies and routines (in the terminology of this
paper, in place of knowledge which is ’embodied’ and ’embedded’),
emphasis is increasingly falling on the knowledge that is located in
brains, dialogue and symbols (i.e. knowledge which is ’embrained’,
’encultured’ and ’encoded’). Conventional assumptions about the nature
of knowledge are not without their difficulties, however; a point which
has emerged strongly from studies of the impact of new information
and communication technologies. Inspired by such studies, and drawing
 1023

from recent debates in philosophy, linguistics, social theory and cognit-

ive science, an alternative approach is outlined. Rather than regarding
knowledge as something that people have, it is suggested that knowing
is better regarded as something that they do. Such an approach draws
attention to the need to research ways in which the systems which medi-
ate knowledge and action are changing and might be managed. The
conclusion of the paper is that debate about the growing importance of
esoteric experts and flexible organizations should be located within a
broader debate about the nature of expertise and of the changing sys-
tems through which activities are enacted.

Images of Knowledge within Organization Studies

Within the organization studies literature a variety of approaches to

knowledge can be identified. One obvious place to begin exploring
these is the literature on organizational learning. The metaphor or
organizational learning is not new, it has attracted attention at least since
Chandler (1962). Interest in the United States has been consistently high
(see, e.g., Argyris and Schon 1978; Duncan and Weiss 1979; Nelson
and Winter 1982; Daft and Weick 1984; Fiol and Lyles 1985; Nonaka
and Johansson 1985; Levitt and March 1988; Zuboff 1988; Henderson
and Clark 1990; Senge 1990; Brown 1991; Kochan and Useem 1992;
Dixon 1994; and special editions of Organization Science 1991 and of
Organization Dynamics 1993) although, especially in recent years, a
strong interest has been developing in the United Kingdom and Europe
also (Hedberg 1981; Garratt 1987; Pedler et al. 1991; Swieringa and
Wierdsma 1992; Dodgson 1993, and see also Douglas 1987).
At least five images of knowledge can be identified in this literature.
Adapting and extending a categorization of knowledge types suggested
by Colilns (1993) these are knowledge that is embrained, embodied,
encultured, embedded and encoded.
Embrained knowledge: is knowledge that is dependent on conceptual
skills and cognitive abilities (what Ryles 1949, called ’knowledge that’
and James 1950, termed ’knowledge about’). As discussed further
below, within Western culture abstract knowledge has enjoyed a
privileged status, and in the organizational learning literature a number
of commentators have emphasized its importance. Fiol and Lyles
(1985), for example, reflect the predominant view of the distinctive
status of abstract knowledge when they contrast ’routine’ behavioural
adjustments with what they term ’higher level’ abilities to develop com-
plex rules and to understand complex causations. Perhaps the best
known theorist of organization learning who has featured embrained
knowledge is Argyris, whose theory of ’double-loop’ learning (e.g.
Argyris and Schon 1978) encourages an explicit recognition and
reworking of taken-for-granted objectives. A recent account in this
tradition is Senge (1990) who synthesizes personal insights, models,
1024

systems thinking and shared visions in a general account of organization I

learning.
Embodied knowledge: is action oriented and is likely to be only partly
explicit (what Ryles 1949, called ’knowledge how’, and James 1950,
’knowledge of acquaintance’). A contemporary account of embodied I

knowledge is included in Zuboff (1988): such knowledge, she says,

depends on peoples’ physical presence, on sentient and sensory
information, physical cues and face-to-face discussions, is acquired by
I
doing, and is rooted in specific contexts. Other accounts include Scrib-
ner’s (1986) description of ’practical thinking’, i.e. problem-solving
techniques which depend on an intimate knowledge of a situation rather
than abstract rules, Hirschhorn’s (1984) analysis of mechanization and
his conclusion that operators’ tacit understandings of machine systems
are more important than their general knowledge, and Suchman’s I

(1987) studies of how people spontaneously construct interpretations of

technologies as they interact with them.
Encultured knowledge: refers to the process of achieving shared
understandings. Cultural meaning systems are intimately related to the
processes of socialization and acculturation; such understandings are
likely to depend heavily on language, and hence to be socially con-
structed and open to negotiation. As Swidler (1986) indicated, in
periods of social transformation explicitly formulated ideologies
become the main vehicle for promoting new recipes for action. Follow-
ing Pettigrew (1979) and Ouchi’s ( 1980) discussions of organizational
culture there has, of course, been considerable interest in the relevance
to organizations of such processes. Within the literature on organiza-
tional learning, Srivastva and Barrett (1988) demonstrated how the
imagery in the language of a group can change over time: as people
grasp for new insights, they experiment with new metaphors into their
talk which others may take up and develop; and Czamiawska-Joerges
(1990) illustrated how consultants explicitly endeavour to manage this
process. Other important contributions include Orr’s (1990) account of
stories shared by maintenance technicians about complex mechanical
problems, and Nonaka’s (1991, 1994) discussions of ’knowledge-
creating’ organizations (these are discussed further below).
Embedded knowledge: is knowledge which resides in systemic rou-
tines. The notion of ’embeddedness’ was introduced by Granovetter
(1985), who proposed a theory of economic action that, he intended,
would neither be heavily dependent on the notion of culture (i.e. be
‘oversocialized’) nor heavily dependent on theories of the market (i.e.
be ‘under-socialized’): his idea was that economic behaviour is intim-
ately related to social and institutional arrangements. Following Badar-
acco (1991), the notion of embedded knowledge explores the signific-
ance of relationships and material resources. Embedded knowledge is

analyzable in systems terms, in the relationships between, for example,

technologies, roles, formal procedures, and emergent routines. This is
how, for example, Nelson and Winter (1982) analyzed an organization’s
 1025

capabilities. They noted that an individual’s skills are composed of sub-

elements which become co-ordinated in a smooth execution of the over-
all performance, impressive in its speed and accuracy with conscious
deliberation being confined to matters of overall importance; thus, they
maintained, may an organization’s skills be analyzed. In addition to the
physical and mental factors that comprise individual skills however,
organizational skills are made up of a complex mix of interpersonal,
technological and socio-structural factors. Similar approaches include
Levitt and March’s (1988) development of the notion of organizational
routines (which, they suggest, make the lessons of history accessible
to subsequent organizational members) while other writers refer to
’organizational competencies’ (Prahaled and Hamel 1990). A related
orientation has been proposed by Henderson and Clark (1990) who
distinguish between the knowledge of specialist elements in an organ-
ization (’component knowledge’) and knowledge about how such ele-
ments interact ( ‘architectural knowledge’); architectural knowledge is
often submerged within an organization’s taken-for-granted routines
and interactions, yet is central to an understanding of its strengths and
weaknesses.
Encoded knowledge: is information conveyed by signs and symbols.
To the traditional forms of encoded knowledge, such as books, manuals
and codes of practice, has been added information encoded and trans-
mitted electronically. Zuboff’s (1988) analysis of the ’informating’
power of information technologies explores the significance of this
point for organizations: information encoded by decontextualized,
abstract symbols is inevitably highly selective in the representations it
can convey. Poster’s (1990) thesis on how the new information techno-

logies may be ’culturally alien’ and Cooper’s (1992) analysis of the

significance of technologies of representation for the theory of organiza-
tion are amongst the writings which have complemented such lines of
analysis.
Brown’s ( 1991 ) account of efforts to develop Xerox as a learning organ-
ization provides an example of how the development of each of these
different forms of knowledge may contribute to organizational learning.
Brown pointed to the advantages for a company like Xerox of undertak-
ing new product development in close association with potential cus-
tomers (i.e. in the terminology of this paper, he identified the relevance
of the embedded knowledge of Xerox’s customers for an understanding
of their reactions to new office machinery). He illustrated how design
engineers at Xerox learned from ethnographic studies of how people
interact with machines (i.e. from studies of the ways in which encoded
knowledge interacts with, and may disrupt, embodied knowledge) and
he emphasized too how studies of communications between engineers
in Xerox have revealed how essential dialogue is between them (i.e.
encultured knowledge) to increase their effectiveness in solving prob-
lems. Finally, Brown emphasized the importance of encouraging senior
managers to develop new appreciations of their company’s established
1026

practices (i.e. he pointed to the importance of developing embrained

and encultured knowledge at senior management levels).
Derived as it is from the literature on organizational learning, the five
types of knowledge identified here do not focus on the commodification
of knowledge into products, systems, or services. (Thus, economists’
interests in the immediate competitive potential of industrial secrets,
patents, etc.-

see e.g. Winter 1988 -

or with the cumulative advant-

ages that such knowledge may provide, Arthur 1990, are not included
within this typology). What the variety of images of knowledge identi-
fied here serves to emphasize is the complexity of issues that any dis-
cussion of knowledge within organizations must address. For example,
it indicates that all individuals and all organizations, not just so-called
’knowledge workers’ or ’knowledge organizations’, are knowledgeable.
As is discussed in the following sections, the typology can also be used
to review claims that significant changes are presently taking place in
the relationship between knowledge and economic success, and to intro-
duce a critique of conventional approaches to analyzing such
developments.

Organizations and Different Types of Knowledge

Drucker ( 1993) has offered an historical interpretation of the suggestion
that, within the demands of contemporary capitalism, a shift is occurring
in the relationship between knowledge and wealth creation. In the eight-
eenth century, he suggests, the basis for an economic system based on
machines and factories was laid with the development of ’technologies’.
These he describes as ’knowledge applied to tools, processes and prod-
ucts’ (in the terminology introduced above, this involved the develop-
ment of new approaches to the study of embodied knowledge, i.e. craft
skills, supported by the granting of patents to inventors and
entrepreneurs). Later, in the early years of this century, F. W. Taylor’s
development of a technology of work analysis provided the basis for
a further impetus to productivity. Drucker describes this as ’knowledge

applied to human work’ (in the terminology used above this involved
the systematic development of systems of embedded knowledge). Now,
Drucker maintains, a society is emerging that is dependent upon the
development and application of new knowledges. ’Knowledge is being
applied to knowledge itself.’ In the terminology of this paper, Drucker’ss
thesis can be taken to imply that embrained and encultured knowledge
are beginning to assume predominant importance.
Both the practical and the theoretical implications of Drucker’s thesis
are significant. Just as the nature of organization and management

changed dramatically at the time of industrial revolution and later as a

result of Taylorism, Drucker maintains that new approaches are now
becoming necessary. Productivity is becoming dependent on the
application and development of new knowledges, and on the contribu-
 1027

tions of specialist knowledge workers. Drucker’s thesis is that know-

ledge workers are unlike previous generations of workers, not only in
the high levels of education they have obtained, but principally because,
in knowledge-based organizations, they own the organization’s means
of production (i.e. knowledge). Drucker suggests that, in these circum-
stances, familiar images of organizations as hierarchical, decentralized
or as a matrix should be discarded. Alternative models can be developed
from examples of organizations based on key specialist experts, such
as hospitals, symphony orchestras or the British Colonial administration
in India.
In recent years, other American commentators have presented related
ideas. Shortly before his appointment as U.S. Secretary of State for
Labour the political economist Reich ( 1991 ) suggested that the globaliz-
ation of the world’s economy is creating a split between the production
of standardized products in low-wage economies, and high value-added
problem solving which may be undertaken wherever useful insights can
be found. Accordingly, the maxim that a nation’s chief economic asset
are the skills and insights of its citizens assumes new significance. From
his discussions with a range of senior executives in major American
corporations, Reich believes that the strategies of big businesses no
longer focus on products as such, rather, they are endeavouring expli-
citly to exploit the competitive advantage that specialized knowledge
can provide. High value-added depends on problem solving; in the
international economy, value-added accrues anywhere around the world
where useful insights can be channelled to respond to the particular
needs of individual customers. The tendency for manufacturing and
service companies to concentrate on the provision of speciality services
has become so advanced, Reich believes, that the traditional distinction
in economics between goods and services has broken down. Moreover,
he emphasizes the undesirable social consequences that are likely to
result from the dependency of low pay, low status workers in service
industries or routine production, on highly paid, high status ’knowledge
workers’ .
The skills of what Reich calls ’symbolic analytic’ workers are varied.
They command high rewards, he believes, because they are difficult to
duplicate. Such skills include problem solving (research, product
design, fabrication), problem identification (marketing, advertising, cus-
tomer consulting), and brokerage (financing, searching, contracting).
When combined, Reich observes, these skills allow technical insights
to be linked both to marketing knowhow and to strategic and financial
acumen. In the terminology suggested in this paper, Reich is high-

lighting the contemporary significance of embrained knowledge. Such

knowledge can be used to support new forms of organization based on
networks, partnerships or contractual arrangements.
Both Drucker and Reich attribute particular significance to knowledge
workers. While Drucker’s thesis is clearly influenced by Bell’s theory of
post-industrialism, Reich’s analysis is strongly influenced by the current
1028

difficulties of the American economy, especially the reduced interna-

tional dominance of American conglomerates, changes in American
manufacturing industry, an influx of foreign capital, and acute social
inequalities. Yet both claim their approaches reach beyond the Amer-
ican experience, and locate their interpretations in a world perspective,
suggesting that as national economies are integrated into the global
economy, similar developments are occurring in other countries as
well.
A less sophisticated account than either Drucker’s or Reich’s, but one
which anticipated a number of their points had been presented by a
Swedish businessman and a British journalist in the mid-1980s. Sveiby
and Lloyd (1987) developed an account for general managers not of
knowledge workers but of knowledge organizations. Defining know-
how as ’value added information’ they suggested that ’know-how com-
panies’ provide a non-standard, creative, problem-solving service. To
be successful, know-how companies must, Sveiby and Lloyd suggested,
be high on what they called professional skills, yet in itself this would
be insufficient. The new breed of know-how organizations also need a
high level of ’managerial skills’ (defined as ’the ability to preserve
added value’). Examples of professional know-how organizations that
they provided included highly entrepreneurial (and very profitable) mer-
chant banks, advertising agencies, software firms, and management,
architectural and engineering consultancies.
Sveiby and Lloyd’s account is not without its problems but their ana-
lysis of knowledge-intensive organizations (rather than capital, techno-
logy, or labour intensive) was unusual. Anticipating some of Drucker’s
observations they noted how such firms present particular problems of
organization and management (for example, power in know-how com-
panies stems primarily from ability and reputation; new forms of
employment relationship may be demanded by know-how workers;
high short-term profit is likely to be a mistaken goal for know-how
companies, what matters is the company’s ability to convince clients
of the value of a long-term relationship). Rather than the specifics of
their observations, it was the powerful image of the ’professional know-
how organization’ that attracted the attention of academics. Starbuck
in the United States and Alvesson in Sweden both preferred a different
terminology, ’knowledge-intensive firm’, but their interests covered
similar ground to Svieby’s. Thus, in presenting his comments on know-
ledge-intensive firms Starbuck (1992) emphasized the economic signi-
ficance of esoteric knowledge over common knowledge and pointed to
the potential distinctions between specialist expertise and the skills of
the established professions. He emphasized the importance of social
skills and client relationships to the activities of knowledge workers
and the success of their companies, and explored the difficulties that
knowledge-intensive firms may have in developing their own learning
(for example, experts may not be receptive to new ideas). In a sub-
sequent paper (Starbuck 1993) he further explores the distinctive iden-
 1029

tities of knowledge-intensive firms, and the need to analyze them within

their particular market situations.
Alvesson (1993b), on the other hand, has reported how the managers
of knowledge-intensive firms may cope with their dependency on their
specialist workers. In an analysis of a computer consultancy, he identi-
fies the ideological controls that management used, striving to create a
’strong’ culture, by manufacturing a sense of community, using per-
formance related rewards, cultivating a positively buoyant outlook, and
systematically intervening in an attempt to influence the ways in which
employees thought of themselves and the company.
Recent commentaries on knowledge-intensive firms in the popular man-
agement literature have concentrated less on knowledge workers as the
recipients of cultural manipulation and more on their active participa-
tion within their organization’s dialogue. Daft and Weick’s (1984)
notion of organizations as systems of interpretation anticipated many
of the issues that are now being raised: to survive, they argued, organ-
izations must find ways to interpret events. Indeed, the processes of
’sensemaking’ that Daft and Weick highlighted are likely to be espe-
cially important for firms that concentrate on the solution of unfamiliar
problems; thus, Peter’s (1992) discussion of the consultancy firm
McKinsey’s points to the central role that communication plays in that
organization where energetic efforts are made to share key reports, a
data bank of project lessons is maintained to create ‘an internal mar-
ketplace of readily accessible ideas’, and experienced consultants rou-
tinely make themselves available to other staff for comments or guid-
ance. The conclusion that Webber (1993) takes from Peter’s description
is that, in a sense, conversations are McKinsey’s.
Drawing from four of the knowledge types identified at the start of this
paper, an overview of the knowledge work literature reviewed in this sec-
tion is offered in Figure 1. Organizations which depend differentially on
knowledge that is embodied, embedded, embrained, and encultured are
distinguished in a two-by-two matix. This is developed by distinguishing
between organizations which first, focus on problems of a routine kind
versus those that are preoccupied with unfamiliar issues and second,

depend heavily upon the contributions of key individuals versus those

who are more obviously dependent upon collective effort. Four kinds of
organization are thus differentiated in the Figure: (i) expert-dependent
organizations, which depend heavily on embodied knowledge; (ii) know-
ledge-routinized organizations, which depend heavily on embedded
knowledge; (iii) symbolic-analyst dependent, which depend heavily on
embrained knowledge; and (iv) communication-intensive organizations,
which depend heavily on encultured knowledge. This classification pro-
vides a way of summarizing key suggestions in the knowledge work liter-
ature. The arrows depicted in Figure 1 highlight the trends many of the
commentators reviewed in this section purport to have identified: that a
shift is occurring away from dependence on the embodied and embedded
knowledge towards embrained and encultured knowledge.
1030

Figure 1
Organizations
and Knowledge
Types (Arrows
summarize trends
suggested in the
knowledge work
literature)

Encoded Knowledge, and Criticisms of Conventional Approaches

to Knowledge
The accuracy of these very general claims can only, of course, be estab-
lished by empirical investigation and it may be that current develop-
ments are not all one way (for example there would appear to be a trend
in the United Kingdom to organize certain professional bureaucracies in
the public sector, not as symbolic-analyst-dependent or communication-
intensive organizations, but as machine bureaucracies). However there
remains a more basic problem. In recent years, taken-for-granted
assumptions about the nature of knowledge, which underpin the distinc-
tions presented on Figure 1, have been exposed as problematic. Studies
 1031

of the ways in which new forms of encoded knowledge have affected

organizations have played a major part in this reassessment.
It would be a mistake to regard the new generation of information and
communication technologies as neutral tools that can merely be grafted
onto existing work systems. Of particular interest to the present discus-
sion is the way such technologies have been found to disrupt conven-
tional practices, as Hirschhorn (1984) has noted of automated work
systems, Zuboff (1988) of informated work systems, and Pava (1986)
of how such technologies demand new approaches to socio-technical
systems design. The way the technologies intimately interlace with the
minutiae of everyday practices is exposing processes which, previously,
were taken for granted, ignored or misunderstood.
Zuboff’s studies, for example, document in detail how action oriented
skills (in the terms used here, embodied knowledge) are being displaced
by computer technologies (encoded knowledge). The new technologies
bypass the use of immediate, physical responses to situated cues;
instead they require operators to interpret the selective, decontextualized
and abstract symbols that machines present to them. Computers require
sophisticated cognitive abilities; the skills of deduction and a knowledge
of systems and procedures are essential for their satisfactory operation.
Zuboff (1988) and Weick (1985) have suggested that it is foolish to
believe that high-technology work systems can be managed as if con-
ventional processes of sense making are outmoded. Talk about com-
puter-mediated information and the transformation of isolated problem-
solving attempts into a shared activity are crucial to the effective
operation of the ‘informated’ organization. It is only through such pro-
cesses that the process of collective interpretation can be recreated. The

point may be summarized in the suggestion that managers in informated

organizations must contrive to develop the skills which are referred to
here as encultured knowledge.
It is not only through their ’infonnating’ effects, however, that techno-
logies based on micro-electronics are transforming organizations. Such
technologies are also associated with the changes precipitated by eco-
nomic globalization. This is not to say that the transformations associ-
ated with contemporary capitalism are technologically determined. As
Castells (1989) documents in his account of changes occurring in con-
temporary capitalism, governmental enactment of post-Keynesian pol-
icies in, for example, weakening trade unions, developing fiscally aus-
tere policies, retreating from policies of wealth redistribution, and
reducing the size of the public sector are fundamental to an understand-
ing of developments. By the operations they support, the new technolo-
gies are playing a vital role in facilitating the internationalization of
capital, production and labour processes. Castells summarizes how
modem information technologies have transformed money markets and
eroded the distinctions between mass- and customized-markets, and at
the same time they make it possible for organizations to develop flexible
methods of production, to disperse their operations, and to compete in
1032

alliances. The communication and control operations they support

within organizations are also facilitating the demise of bureaucratic
approaches of organization, promising vigorous internal networks, col-
laborative work relations and significantly reduced hierarchical struc-
tures of control (developments explored in detail by Malone and Roch-
art 1991, and Sproull and Kiesler 1991).
Thus, just as familiar working patterns are being transformed by the
encodification of knowledge, at the same time, the new technologies
are making it possible for organizations to operate relatively independ-

ently from geographical location, thereby blurring the boundaries

between one organization and another, and freeing internal communica-
tions within organizations. During the 1980s, much social science com-
mentary on the relationship between information technologies and
organizations emphasized how technologies are not deterministic in
their effects (Buchanan and Boddy 1983). Instead, it was maintained,
they open a range of options from centralization to decentralization,
from automation to work enrichment. Current developments suggest,
however, that rather than thinking of the new technologies as flexible
tools for organizations to use as they believe is appropriate, it may be
better to consider the technologies as the medium for organizing itself.
Organizations that are heavily dependent on the new technologies are,
simultaneously, being imploded into electronic codes and exploded into
(global) information networks.
Some of the tensions involved in these processes are not new. The
uneasy relationship that encoded knowledge may have to other forms
of knowledge has been documented before. For example, Zuboff quotes
the mediaeval historian Clanchy, (1979) who recorded the slow and
largely reluctant acceptance in eleventh and twelfth century England of
written documentaion: ’both to ignorant illiterates and to sophisticated
Platonists, a written record was a dubious gift, because it seemed to
kill living eloquence and trust and substitute for them a mummified
semblance in the form of a piece of parchment’. Yet the extent of the
disruption to knowledge bases associated with electronically encoded
information is new. As summarized in Figure 1, the new media of
encoded knowledge not only affect embodied knowledge, but may also
affect the nature and significance of embrained knowledge (as informa-
tion becomes ever more accessible and expert computer systems are
developed), encultured knowledge (as new communication systems are
introduced to support group working between individuals who are sep-
arated in time and space), and embedded knowledge (through, for
example, the development of integrated manufacturing systems).
The close relationship between encoded knowledge and the other
images of knowledge highlighted in this discussion illustrate the point
that it is a mistake to assume that embodied, embedded, embrained,
encultured and encoded knowledge can sensibly be conceived as separ-
ate one from the other. Knowledge is multi-faceted and complex, being
both situated and abstract, implicit and explicit, distributed and indi-
 1033

vidual, physical and mental, developing and static, verbal and encoded.
Analysis of the relationships between different manifestations of know-
ledge identified in this paper is at least as important as any delineation
of their differences.

From Theories of Knowledge to Theories of Knowing

Nonaka’s ( 1991, 1994) descriptions of the ’knowledge-creating’ organ-
ization provides a useful starting point for theorizing about the links
between the different forms of knowledge identified in this paper.
Nonaka is concerned with the management of innovation. This he
regards as an ongoing process in which organizations create problems,
define them, then develop new knowledge for their solution. He
develops the idea that knowledge is created out of a dialogue between
peoples’ tacit and explicit knowledge. Knowledge may move from tacit
to tacit (e.g. in a craft apprenticeship), from explicit to explicit (e.g.
when hitherto distinct but related bodies of information are brought
together), from tacit to explicit (e.g. the study of craft skills), and from
explicit to tacit (e.g. the internalization of new knowledge). Nonaka
maintains that all four of these patterns exist in dynamic interaction in
‘knowledge-creating’ companies. He does not wish to suggest, however,
that the processes involved here are merely a recycling of knowledge.
Knowledge creation, he believes, is closely associated with language
and communications, requiring the creative use of metaphors, analogies
and models, and a resolution of the conflicts and disagreements that
new approaches may provoke. In the terminology of this paper, Nonaka
is suggesting that encultured knowledge is intimately related to the
development of embodied, embrained and embedded knowledge. His
approach traces the link between different forms of knowledge to the
processes through which they are created.
In other respects, however, Nonaka’s approach is rather traditional. He
insists that knowledge is a specific entity, formed in the minds of indi-
viduals (albeit generated in interactions with others), and conceptually
distinct from the material technologies around which organizations are
structured (see Nonaka 1994). Similarly, while his concept of ’know-
ledge-creation’ pushes the distinction between knowledge and learning
to its limits, he wishes to maintain a distinction between them.
To develop the analysis of the interrelations between different types of
knowledge further, it is necessary to address the basic question: what
is knowledge? (or perhaps as Pear 1972, asks, what is not knowledge?).
In recent years there has been considerable debate about this issue.
Postmodernists, for example, have challenged the idea of fundamental
truth by suggesting that truth is a story (see, e.g., Lawson 1989); cognit-
ive anthropologists, ethnomethodologists and symbolic interactionists
have queried the value of abstract plans and the notion of social struc-
ture and have demonstrated the significance of situated skills and prag-
1034

matic knowledge (e.g. Suchman 1987); and sociologists of science have

challenged deep-rooted assumptions about the privileged status of expli-
cit abstract knowledge by studying knowledge creation as a cultural
process and by de-emphasizing conventional distinctions between
people and technology (e.g. Latour 1987; Law 1992).
The various implications of such approaches remain to be fully
described. Yet it is becoming clear that traditional conceptions of know-
ledge as abstract, disembodied, individual, and formal are unrealistic.
Polkinghome ( 1992), for example, reviews the implications of postmod-
ernism for the theory of practice. Practical knowledge, he suggests, is
foundationless, partial, constructed and pragmatic. A similar outlook is
presented by Lave ( 1993 ), who reviews points of agreement between
cognitive anthropologists, ethnomethodologists and activity theorists.
Such theorists agree, she says, that major difficulties occur when educa-
tionalists assume that knowledge can be divorced from context and
transmitted either as abstract data or as universally applicable
approaches to problem solving; learning is not a passive process, she
argues, but an active one. Defining learning as a creative (and
collective) interpretation of past experiences she summarizes the emer-
ging consensus between educational researchers as agreement
that:

‘ 1. Knowledge always undergoes construction and transformation in use.

2. Learning is an integral aspect of activity in and with the world at all times.
That learning occurs is not problematic.
3. What is learning is always complexly problematic.
4. Acquisition of knowledge is not a simple matter of taking in knowledge;
rather, things assumed to be natural categories, such as ’bodies of know-
ledge’, ’learners’, and ’cultural transmission’, require reconceptualization
as cultural, social products.’ (Lave 1993: 8)

Star (1992) has also presented a summary of patterns of agreement

between contemporary social theorists, drawing on a similar literature
to Lave’s although placing a heavier emphasis on research studies in
the actor-network tradition. The emerging consensus that conventional
views of knowledge are unacceptable is so widespread, Star believes,
that she refers to it as ’an invisible college’ and ’an intellectual move-
ment that as yet has no name’. Reviewing detailed studies of technology
and work, she echoes Lave’s points in her observation that the boundar-
ies of knowledge in complex organizations are fluid and overlapping.
Star explains this by reviewing studies which suggest that cognitions are
situated (as the circumstances of action shape even the most abstractly
represented tasks); cognitions are collective (as practices are distributed
socially and technologically); and that, rather than being a mere internal
manipulation of ideas, cognitions are also forms of material practice
(i.e. cognitions not only involve an internal manipulation of ideas but
they involve physical, manual and interactional actions as well).
Accounts such as these (see also Brown and Rogers 1991, for a similar
 1035

approach developed for the theory of communciation) provide a useful

>

starting point for the development of a unifying theory of organizational

:
knowledge. First, rather than talking of knowledge, with its connotations
of abstraction, progress, permanency and mentalism, it is more helpful
to talk about the process of knoH’ing. Second, to avoid segregating the
forms of knowing identified in this paper, old concepts (such as the
split between the abstract and the specific, individuals and communities,
and the social and the technical) need to be abandoned and new
approaches to COIIC’PptllClll~lil~~ the multi-dimensional processes of
knowing and doing need to he created. One approach to this task could
be to develop from the insights that knowing is situated, distributed
and material.
.
.

, , .. , ,~ -, ,
.

:; I.
’’. &dquo;tt :,
Activity Theory, Knowing and Doing
Out of the range of theoretical approaches that both Lave (1993) and
Star (1992) include in their reviews which might be of value in this
project, activity theory offers particular promise. Activity theory has its
.

origins in the ideas of the Russian psychologist Vygotsky who, working

.
in the 1920s, endeavoured to develop an understanding of mind and
society which did not depend upon the dichotomies (e.g. mind versus
body, thought versus action, individual versus society, etc.) that have
characterized mainstream Western thought (and which lend credence
to the clear distinctions assumed between embodied, embedded,
embrained and encultured knowledge). Basic to the Vygotsky approach
is the Marxist idea that it is not the consciousness of humans that deter-
mines their social being, but social experiences which shape their con-
sciousness : psychological processes can only be understood by an
appreciation of the, culturally provided, factors that mediate them.
(Vygotsky thought, for example, that it would be a mistake to think
.

that children pass through a stage of egocentric speech before they use
language socially, his view was the opposite, i.e. that children learn to
internalize speech which is, from the start, oriented to their external
,
social environment, see Kozulin 1990).
’
Contemporary versions of activity theory take a variety of forms. How-
ever, all are explicit in their attempts to develop a unified account of
.

knowing and doing, and all emphasize the collective, situated and tent-
ative nature of knowing. Some (e.g. Brown, Collins and Duguid 1989:
,
and Lave and Wenger 1991) concentrate on the processes through
which people develop shared conceptions of their activities. Others,
(Hutchins 1983; Engestrom 1987, 1993) model the relationships that
; exist between a community’s conceptions of its activities and the mat-
erial, mental and social resources through which it enacts them. While
the former approach develops a model of learning as socialization, the
latter explores the circumstance in which communities may enact new
conceptions of their activities. . - .
1036

Orr’s ( 1990) analysis of Xerox maintenance technicians is in the Brown/ I

Lave tradition of activity theory. He describes how the stories shared
by maintenance personnel about complex technical problems is an
essential part of their activities. In the first place the stories they tell
each other serve a key informational function, preserving and circulat,
ing essential news about particular problems. Second, the storytelling
has an educational function: not only do the technicians learn about
particular faults on the machines, they also help the participants develop
their diagnostic and trouble-shooting skills. Finally, the stories provide
an opportunity for the technicians to establish their identity within the

community of technicians itself; as newcomers contribute to the

storytelling process they begin both to demonstrate their identity as
professionals and to contribute to the collective wisdom of their group.
In their discussion of the wider implications of this study Brown and
Duguid (1989) emphasize the general significance for organizations of
such processes. Learning is a socially constructed understanding, they
argue, that emerges from practical collaboration. Collective wisdom
depends upon communal narratives. Note that this analysis suggests
that it is not just esoteric consultancies like McKinsey’s (see Peters
1993; and Webber 1993, discussed earlier) which benefit from lively
internal communications; collective dialogue is also an essential aspect
of life in other, less glamorous, organizations, developing skills and
abilities which are distributed (often unnoticed) amongst the employees
within them.
Engestrom’s ( 1989, 1991) study of a medical practice in Finland illus-
trates a second version of activity theory. Partly through discourse ana-
lysis, partly from observation, and partly from accounts of the history
of medicine he was able to distinguish the variety of conceptions that
doctors may have of their activity. Doctors may conceive of their work
in biomedical, administrative-economic, psychiatric, socio-medical, or
in system-interactive terms. Doctors in the same medical practice may,
perhaps unknowingly, be enacting different conceptions of health care,
yet attempts to refocus priorities may not be easy to achieve. In Enges-
trom’s study, attempts to reorientate priorities towards psycho-somatic
and socio-medical priorities were hampered by the resource system
within which the doctors operated: ( i ) the division of labour between
doctors and other health-care professionals proved inflexible; (ii) the
way patients are randomly allocated to doctors in the Finnish health-
care system created problems of continuity of care; and (iii) the biomed-
ical concepts and techniques that the doctors had become accustomed
to using encouraged them to continue treating health-care problems as
biomedical problems.
The analysis Engestrom offers of this, and other, work settings is expli-
citly intended to avoid separating the individual from the collective, or
the social from the technical. Fundamental to his approach is the unit
of analysis he adopts, namely, the socially-distributed activity system.
The general model he offers of such systems is shown in Figure 2.
 1037

Essential to such systems are the relations between agents, the commun-
ity of which they are members, and the conception(s) people have of
their activities (the inner triangle of relations in Figure 2). Such relations
are mediated by a further series of factors,
including the language and
technologies used by participants within the system, the implicit and
explicit social rules that link them to their broader communities, and the
role system and division of labour adopted by the community.

1’B ,
,- .

Figure 2
A General Model
of Socially-
Distributed
Activity Systems
(Based on
Engestrom 1987)

. ’ .=&dquo; .
1 1

A summary model of Engestrom’s analysis of the dynamics of the med-

ical practice he studied is shown on Figure 3. Note that the relations
depicted in this figure are neither static nor are they necessarily harmo-
nious. The three points of tension noted in Engestrom’s fieldwork,
detailed above, are featured on the model [see points (i), (ii) and (iii)].
Indeed Engestrom’s approach suggests that, far from being unusual,
tensions such as these are commonplace within distributed work sys-
tems. His analysis of the dynamics of activity systems is reminiscent
.

of Perrow’s (1984) suggestion that accidents are a normal feature of

life in complex industrial work systems. Likewise, for the most part,
everyday interruptions and breakdowns in the workings of activity sys-
tems are skillfully, regularly and normally repaired (although system
breakdown may sometimes occur). It is through their collective deter-
mination and skill, both in their actions and their language, that particip-
ants enact particular frames (i.e. impose conceptions of their activities
: on situations they believe appropriate) and maintain a (seemingly)
smooth flow of events.
Note that the incoherencies, paradoxes and conflicts that feature within
activity systems provide a potential driving force for change. Enges-
trom’s analysis suggests that organizations and institutions are a lot less
stable and rational than is usually recognized. The incoherencies and
contradictions that feature within them are obscured, however, partly
no doubt by conventional imagery of the organization as a rational

machine, but also by the skills of participants who learn to work within
the situation in which they find themselves. New ways of knowing and
doing can emerge if communities begin to rethink what, in a different
1038

Figure 3
The Tensions
within a Medical
Practice (adapted
from Engestrom
1991. Points (i).
(ii) and (iii) mark
points of tension
within the
activity system)

context, Unger (1987) has called the ’false necessity’ of everyday life,
and to engage with the tensions in their activity systems. The complexi-
ties of socially distributed activity systems suggest that incoherencies
and tensions are inevitable: the issue is not how can they be eradicated
but how they should be treated.

Rethinking Knowledge and Organizations I

As the review presented in the first part of this paper indicated, current
interest in knowledge and knowledge work marks a change of emphasis I
within contemporary capitalism away from knowledge that is embodied
and embedded, to knowledge that is embrained, encultured and
encoded. The approach presented in the second part of this paper offers
a different orientation. To summarize:
l. Lave suggests that knowledge should not be conceived as a timeless
body of truth that experts have internalized and which organizations
may harness. She suggests that the notion of ’bodies of knowledge’
(with its connotations of universal truth) is a problem in its own right.
General abstractions are no more than resources to be used in specific
circumstances where (in actions, improvization and dialogue) creativity
is ubiquitous. By focusing on knowing rather than knowledge, the dis-
tinction that is conventionally assumed between knowledge and learn-
ing is avoided.
2. Engestrom’s interpretation of the dynamic relationships between
individuals, their communities and the objects of their activities pro-
vides a clear alternative to approaches which attempt to study such
entities, or the factors which mediate the relationships between them,
in isolation one from the other. His suggestion is that the appropriate
unit of analysis is neither individuals nor organizations, but socially-
distributed activity systems. People act on the world, with others, utiliz-
ing (and contributing to the development of) the linguistic, material and
social resources currently available. Knowledge does not appear as a
separate category in Engestrom’s model, rather, it permeates the rela-
tions he depicts. His approach models the dynamics of knowing: each
 1039

moment is a compromise, the balance within an activity system changes

constantly. Participants employ their situated knowledge in a situation
which is itself constantly developing. In response to this changing situ-
ation participants’ knowledge and behaviour will also inevitably
develop.
3. Activity theories in general argue that knowledge is constantly evol-
ving. Analysis of the tensions that inevitably develop within socially-
distributed activity systems points to the opportunities for system devel-
opment that (routinely) arise. Orr, Brown and Duguid demonstrate how
.

essential language is to this process. Talk enables collective interpreta-

tions, negotiates behavioural priorities, signals group membership, and
helps to create a community. Language is an archetypal communal
activity, integral to the enactment of practical actions.
, Thus, helpful though it can be to characterize knowledge as embodied,
,
embedded, embrained, encultured and encoded, the concept of know-
ledge is problematic. Rather than studying knowledge as something
individuals or organizations supposedly have, activity theory studies
knowing as something that they do and analyzes the dynamics of the
.

systems through which knowing is accomplished. Recaste in this way,

~
knowing in all its forms is analyzed as a phenomenon which is: (a)
manifest in systems of language, technology, collaboration and control
.

(i.e. it is mediated); (b) located in time and space and specific to particu-
<
lar contexts (i.e. it is situated); (c) constructed and constantly develop-
ing (i.e. it is provisional): and (d) purposive and object-oriented (i.e. it
is pragmatic).
Before considering how these conclusions can be used to inform debate
i about knowledge and knowledge work, however, it should be noted
that, in at least one respect, an extension of activity theory is required.
Activity theory is not alone in its attempts to draw attention to the need
to rethink supposed distinctions between events and contexts, language
: and action, the social and the technical, etc.; as noted above, similar
>
suggestions have also been made by anthropologists, social theorists,
and linguists and others. Of the comparisons that might be made
between these various approaches, one point stands out: activity theory
is weak in the analysis it offers of the relationship between knowledge
and power. This is not to say that power as an issue does not occur at
all in the writings of activity theorists. (For example, in her criticism
of the term ’bodies of knowledge’, highlighted above, Lave 1993,
adopts Latour and Woolgar’s 1979, terminology to suggest that claims
to the possession of decontextualized knowledge are frequently no more
than examples of erasure, collusion or domination). However, analysis
of power in everyday life has featured far less in the writings of activity
theorists than it has in the work of others who are theorizing practice
from different traditions. This is well illustrated by the issues that preoc-
cupied Ortner ( 1984) in her discussion of the relevance to anthropology
of theories of practice (such as Bourdieu 1978; and Giddens 1979) that
were emerging in the early 1980s. Ortner supported the attempts being
1040

made in these writings to treat societies and cultures as integrated

wholes and to avoid segregating social, economic and political factors
from values, ideals and emotions, but she emphasized how it would be
a mistake to treat all the elements of a social system as if they are of

equal analytical significance. Social systems are fundamentally unequal.

Gramsci’s ( 1957) notion of hegemony and Foucault’s ( 1980) notion of
a ’discourse of perversions’ serve as reminders that any theory of know-

ing as a cultural activity must acknowledge the, often self-reproducing,

dynamics of domination and subordination that are a feature of every-
day life.
To the suggestion just made that knowing is mediated, situated, prwi-
sional and pragmatic must therefore be added the point that it is also
contested. As noted at the start of this article, this point has not passed
unnoticed within the literature on knowledge work and knowledge-
intensive firms (see Alvesson 1993a; and Knights, Murray and Willmott
1993 ).
Applied to the of knowledge work, the approach developed here
study
suggests that, alternative to focusing on the kinds of knowledge
as an
that capitalism currently demands, attention should focus on the systems
through which knowing and doing are achieved. Because of the changes
that are occurring in capitalism touched on earlier (such as moves
towards the globalization of markets and finance, new information and
communication technologies, post-Keynesian governmental policies,
and new approaches to strategy, management and organization, etc.)
activity systems are changing significantly. Rather than asking ’what
sorts of knowledge are needed in contemporary capitalism and how
may organizations harness them’?’ the question thus becomes ’how are
systems of knowing and doing changing, and what responses would be
appropriate’?’
This revised formulation promises to establish links between the know-
ledge work literature and broader studies of economic and organiza-
tional changes. The analysis of knowing as mediated, situated, provi-
sional, pragmatic and contested, provides a basis for identifying
research priorities. Taking each of these in turn:
Knowing as mediated: Research is needed into the dynamics of activ-
ity systems and how they are currently changing. As discussed earlier.
changes associated with the new information and communication tech-
nologies are combining with other developments, such as new economic
and organizational structures and new approaches to management to
transform the contexts of action. Further work is needed into such
changes. One key consequence of these developments is that activity
systems which were previously segregated are becoming interlinked
and, therefore, are growing larger and becoming more complex.
Research is needed to document such developments. Detailed ethno-
graphic studies are needed to illuminate the ways in which people
improvize, communicate and negotiate within expanded activity
systems. , .
 1041

Knowing a,s situated: Work is needed to develop the relevance of the

notion of situated knowledge to the knowledge work debate. As already
noted, the concept of situated knowledge avoids the problems associ-
.
ated with abstract, decontextualized knowledge: it emphasizes the signi-
ficance of peoples’ interpretations of the contexts within which they act
and the key role that ’communities of practitioners’ play in the acquisi-
tion and development of skill. The knowledge work debate draws atten-
tion to the differences in approach that may develop between employees
whose work involves them in action skills or in the execution of proced-
ural routines, and those who are involved in creative problem solving.
Little is known about the ways in which peoples’ understanding of their
activities are changing as a consequence of the developing complexity
of the contexts within which they are working.
Knowing as provisional: Research is needed into the idea that know-
ing is, essentially, provisional and developing. Activity theory suggests
that developments in systems of knowing and doing will occur con-
stantly as tensions (inevitably) emerge within them. However, changes
in activity systems may or may not be planned (for example, the unan-
ticipated impacts of advanced information and communication
technologies), and may or may not be fully understood or articulated
by participants (for example, computer mediated interactions may erode
traditional practices in bureaucracies but people may continue to
describe their organizations through a terminology that is familiar to
them). Activity theory points to the opportunities that might be created
to help participants become more proactive in the development of their
activity systems. At a general level, research is needed into Engestrom’s
proposal that, by alerting people to the tensions in activity systems
that would otherwise be ignored or tolerated, a process of dialogue,
experimentation and collective learning can be triggered that may trans-
form participants’ understandings of their activities and the systems
through which they are enacted. Issues specific to the knowledge work
debate include the study of the tensions within expert-dependent and
knowledge-routinized organizations at the present time, and the ease
with which they may transform themselves into symbolic-analyst
dependent and communication-intensive organizations.
Knowing as pragmatic: Central to activity theory is the idea that col-
lective action is driven by the conceptions people have of the object
of their activities. Further research is needed into the influence that
’informated’ and ’communication-intensive’ environments have on the
approaches people take to their work. It seems likely that, as activity
systems become interrelated and complex, traditional approaches to
organizing are likely to be ineffective. Research is needed into the pos-
sibilities for developing communal narratives within expanded activity
systems. Study is also needed of the anxiety that individuals and com-
munities may experience in the face of significant, ongoing, and perhaps
conflicting demands for changes in their work methods and
priorities.
1042

Knowing as contc~.sted: Finally, as noted before, the concepts of know-

ledge and power interrelated. Conflicts are to be expected within
are
and between the new generation of symbolic analysts and problem
solvers, and established professionals and managers. Beyond this, the
far-reaching social, technological, and economic changes that are at the
heart of the knowledge work debate indicate that issues of domination
and subordination are fundamental to the development of a general
theory of knowing as praxis.
Enquiry along the lines sketched out here is unlikely to contradict the
suggestion that symbolic-analytical work and communication-intensive
organizations are of growing significance at the present time. Rather,
it promises to explain why this is so and, by reframing the problem,
to illuminate some of the difficulties associated with such develop-
ments. In summary, the approach introduced here extends the debate
about the importance of creative experts and flexible organizations to
the (more general) discussion of the nature of expertise and of the sys-
tems through which people enact their activities. The study of know-
ledge work and organizations is, in other words, best located within a
broader analysis of knowing as a cultural phenomenon.

Note *
Prepared for the E.U. Human Capital and Mobilityy Project ’European
Competitiveness in a Knowledge Society’. I am most grateful to Colin Brown, David
Courpasson, Bente Elkjaer, Manuel Graca, Henrik Holt Larson, Karen Legge,
<
.
Yves-Frederic Livian, Mike Reed and Alan Whitaker for their comments on an earlier
.
version, and also to Amdt Sorge, Mats Alvesson and anonymous reviewers for
O.S.

References Abbott, A. Arthur, W. B.

1988 The system of professions: An 1990 ’Positive feedbacks in the eco-
essay on the division of expert nomy’. Scientific American
labour. Chicago: University of (February): 80-85.
Chicago Press.
Badaracco, J.
Alvesson, M. 1991 The knowledge link: How firms
1993a ’Organization as rhetoric: know- compete through strategy alli-
ledge-intensive firms and the ances. Boston, MA: Harvard Uni-
struggle with ambiguity’. Journal versity Press.
of Management Studies 30: 997-
1016. Baer, W.
1987 ’Expertise and professional stand-
Alvesson, M. ards’. Work and Occupations 13:
1993b ’Cultural-ideological modes of 532-552.
management control: theory and
a
a case study of a professional ser- Bell, D.
vice company’ in Communication 1973 The coming of post-industrial
yearbook 16. S. Deetz (ed.), 3- society: A venture in social
42. London: Sage. forecasting. New York: Basic
Books.
Argyris, C., and D. Schon
1978 Organizational learning: A Bourdieu, P.
theory of action approach. Read- 1978 Outline of a theory of practice.
ing, MA: Addison Wesley. Cambridge: Cambridge Univer-
sity Press.
 1043

Brown, J. S. Czarniawska-Joerges, B.
1991 ’Research that invents the cor- 1990 ’Merchants of meaning: manage-
poration’. Harvard Business ment consultants in the Swedish
Review (Jan.-Feb.): 102-111. public sector’ in Organizational
symbolism. B. Turner (ed.), 139-
Brown, J. S., A. Collins, and P. Duguid 150. Berlin: Walter de Gruyter.
1989 ’Situated cognition and the cul-
ture of learning’. Educational Daft, R., and K. Weick
Researcher 18: 32-42. 1984 ’Toward a model of organizations
asinterpretation systems’. Acad-
Brown, J. S., and P. Duguid emy of Management Review 9:
1989 ’Innovation at the workplace, a 284-295.
perspective on organizational
learning’. Paper presented at the Dixon, N.
CMU Conference on ’Organiza- 1994 The organization learning cycle:
tional Learning’, May. How we can learn collectively.
London: McGraw Hill.
Brown, J., and E. Rogers
1991 ’Openness, uncertainty, and Dodgson, M.
intimacy: an epistemological 1993 ’Organizational learning: a
reformulation’ in ’Miscommun- review of some literatures’.
ication’ and problematic talk. N. Organization Studies 14/3: 375-
Coupland, H. Giles, and J. Wiem- 394.
ann (eds.), 146-165. London:
Sage. Douglas, M.
1987 How institutions think. London:
Buchanan, D., and D. Boddy Routledge and Kegan Paul.
1983 Organizations in the computer
age. Aldershot: Gower. Drucker, P.
1993 Post-capitalist society. Oxford:
Castells, M. Butterworth-Heinemann.
1989 The informational city: Informa-
tion technology, economic Duncan, R., and A. Weiss
restructuring and the urban- 1979 ’Organizational learning: implica-
regional process. Oxford: Basil tions for organizational design’ in
Blackwell. Research in organizational beha-
viour, Vol. 1. B. Staw (ed.), 75-
Chandler, A. 123. Greenwich, CN: Jai Press.
1962 Strategy and structure. Boston:
MIT Press. Engestrom, Y.
1987 Learning by expanding: An activ-
Clanchy, M. ity theoretical approach to devel-
1979 From memory to written record: opmental research. Helsinki:
England 1066-1307. Cambridge, Orienta-Konsultit.
MA: Harvard University Press.
Engestrom, Y.
Collins, H. 1989 ’Developing Expertise at the
1993 ’The structure of knowledge’. Changing Workplace; Towards a
Social Research 60: 95-116. Redefinition of Expertise’. Tech-
nical Report 130. La Jolla, Cali-
Cooper, R. fornia. Centre for Information
1992 ’Formal organization as
repres- Processing, University of San
entation : remote control, dis- Diego.
placement and abbreviation’ in
Rethinking organization: New Engestrom, Y.
directions in organization theory 1991 ’Developmental work research:
and analysis. M. Reed, and M. reconstructing expertise through
Hughes (eds.), 254-272. London: expansive learning’ in Human
Sage. jobs and computer interfaces. M.
Nurminen and G. Weir (eds.),
265-290. Amsterdam: North
Holland.
1044

Engestrom, Y. Henderson, R., and K. Clark

1993 ’Work as a testbed of
activity 1990 ’Architectural innovation: the
theory’ in Understanding prac- reconstruction of existing product
tice : Perspectives on activity and technologies and the failure of
context. S. Chaiklin and J. Lave established firms’. Administrative
(eds.), 65-103. Cambridge: Cam- Science Quarterly 35: 9-30.
bridge University Press.
Hirschhorn, L.
Foucault, M. 1984 Beyond mechanization: Work and
1980 The history of sexuality. New technology in a post-industrial
York: Vintage. age. Cambridge, MA: MIT Press.

Fiol, C., and M. Lyles Hutchins, E.

1985 ’Organizational learning’. Acad- 1983 ’Understanding Micronesian nav-
emy of Management Review 10: igation’in Mental models. D.
803-813. Gentner and A. Stevens (eds.),
191-225. Hillsdale, NJ: Erlbaum.
Florida, R., and M. Kenney
1993 ’The new age of capitalism: James, W.
innovation mediated production’. 1950 The principles of psychology.
Futures (Jul.-Aug.): 637-651. New York: Dover.

Galbraith, J. Kanter, R.
1967 The industrial
new state. Boston: 1989 When giants learn to dance. New
Houghton Mifflin. York: Simon and Schuster.

Garratt, B. Knights, D., F. Murray, and H. Willmott

1987 The learning organization, and 1993 ’Networking as knowledge work:
the need for directors who think. a study of interorganizational
London: Fontana. development in the financial ser-
vices sector’. Journal of Manage-
Giddens, A. ment Studies 30: 975-996.
1979 Central problems in social
theory: Action, structure and con- Kochan, T., and M. Useem, editors
tradiction in social analysis. 1992 Transforming organizations. Ox-
Cambridge: Cambridge Univer- University Press.
ford : Oxford
sity Press.
Kozulin, A
Gramsci, A. 1990 Vygotsky’s psychology: A biogra-
1957 The modern prince and other phy of ideas. Hemel Hempstead:
writings. New
York: International Harvester Wheatsheaf.
Publishers.
Latour, B.
Granovetter, M. 1987 Science in action: How to follow
1985 ’Economic action and social scientists and engineers through
structure: the problem of embed- society. Milton Keynes: Open
dedness’. American Journal of University Press.
Sociology 91: 481-510.
Latour, B., and S. Woolgar
Hague, D. 1979 Laboratory life: The social con-
1991 ’Knowledge society, university struction of scientific facts.
challenge’. Marxism Today London: Sage.
(September): 12-17.
Lave, J.
Hedberg, B. 1993 ’The practice of learning’ in
1981 ’How organizations learn and Understanding practice: Per-
unlearn’ in Handbook of organ- spectives on activity and context.
izational design. Volume 1: S. Chaiklin and J. Lave (eds.), 3-
Adapting organizations to their 32. Cambridge: Cambridge Uni-
environment. P. Nystrom and W. versity Press.
Starbuck (eds.), 3-27. Oxford:
Oxford University Press.
 1045

Lave, J., and E. Wenger Ortner, S.

1991 Situated learning: Legitimate 1984 ’Theory in anthropology since the
peripheral participation. Cam- sixties’. Comparative Studies in
bridge : Cambridge University Society and History 26: 126-166.
Press.
Ouchi, W.
Law, J. 1980 ’Markets, bureaucracies and
1992 ’Notes on the theory of the actor- clans’. Administrative Science
network ordering, strategy and Quarterly 25: 129-141.
homogeneity’. Systems Practice
5: 375-394. Pava, C.
1986 ’Redesigning sociotechnicalsys-
Lawson, H. tems design: concepts and
1989 ’Stories about stories’ in Dis- methods for the 1990s’. Journal
mantling truth: Reality in the of Applied Behavioural Science
postmodern world. H. Lawson 22: 201-221.
and L. Appignanesi (eds.), ix-
xxviii. London: Weidenfeld and Pear, D.
Nicolson. 1972 What is knowledge? Oxford:
Basil Blackwell.
Levitt, B., and J. March
1988 ’Organizational learning’. Amer- Pedler, M., J. Burgoyne, and T. Boydell
ican Review of Sociology 14: 1991 The learning company: A strategy
319-340. for sustainable development.
London: McGraw-Hill.
Malone, T., and J. Rockart
1991 ’Computers, networks and the Perrow, C.
corporation’. Scientific American 1984 Normal accidents: Living with
(September): 92-99. high-risk technologies. New
York: Basic Books.
Nelson, R., and S. Winter.
1982 An evolutionary theory of Peters, T.
organizational change. Cam- 1992 Liberation management: Neces-
bridge, MA: Harvard University sary disorganization for the
Press. nanosecond nineties. New York:
Alfred Knopf.
Nonaka, I.
1991 ’The knowledge creating com- Pettigrew, A.
pany’. Harvard Business Review 1979 ’On studying organizational cul-
(Nov.-Dec.): 96-104. tures’. Administrative Science
Quarterly 24: 570-581.
Nonaka, I.
1994 ’A dynamic theory of organiza- Polkinghorne, D.
tional knowledge creation’. 1992 ’Postmodern epistemology of
Organization Science 5: 14-37. practice’ in Psychology and post-
modernism. S. Kvale (ed.), 146-
Nonanka, I., and J. Johansson 165. London: Sage.
1985 ’Japanese management: what
about the ’hard’ skills?’ Academy Poster, M.
of Management Review 2: 181- 1990 The mode of information: Post-
191. structuralism and social context.
Cambridge: Polity Press.
Orr, J.
1990 ’Sharing knowledge, celebrating Prahaled, C., and G. Hamel
identity: communitymemory in a 1990 ’The core competence of the cor-
service culture’ in Collective poration’. Harvard Business
remembering. D. Middleton and Review (May-June): 79-91.
D. Edwards (eds.), 169-189.
London: Sage.
1046

Reed, M. Starbuck, W.
1991 ’Experts, professions and organ- 1993 ’Keeping butterfly and an ele-
a
izations in late modernity: the phant in ahouse of cards: the ele-
dynamics of institutional, occupa- ments of exceptional success’.
tional and organizational change Journal of Management Studies
in advanced industrial societies’. 30: 885-922.
Lancaster, Lancaster University.
Department of Behaviour in Suchman, L.
Organizations. 1987 Plans and situated actions. Cam-
bridge : Cambridge University
Reich, R. Press.
1991 The work of nations: Preparing
ourselves for 21st-century capit- Sveiby, K., and T. Lloyd
alism. London: Simon and 1987 Managing knowhow: Add value
Schuster. by valuing creativity. London:
Bloomsbury.
Ryles, G.
1949 The concept of mind. London: Swieringa, J., and A. Wierdsma
Hutchinson. 1992 Becoming a learning organiza-
tion : Beyond the learning curve.
Scribner, S. Reading, MA: Addison Wesley.
1986 ’Thinking in action: some charac-
teristics of practical thought’ in Swidler, A.
Practical intelligence: Nature 1986 ’Culture in action: symbols and
and origins of competence in the strategies’. American Sociolo-
everyday world. R. Stemberg and gical Review 51: 273-286.
R. Wagner (eds.), 13-30. Cam-
bridge : Cambridge University Unger, R.
Press. 1987 False necessity: Anti-necessita-
rian social theory in the service
Senge, P. of radical democracy. Cam-
1990 Thefifth discipline: The art and bridge : Cambridge University
practice of the learning organiza- Press.
tion. London: Century Business.
Webber, A.
Sproull, L., and S. Kiesler 1993 ’What’s so new about the new
1991 Connections: New ways of economy?’ Harvard Business
working in the networked organ- Review (Jan.-Feb.): 24-42.
ization. Cambridge, MA: MIT
Press. Weick, K.
1985 ’Cosmos versus chaos: sense and
Srivastva, S., and F. Barratt nonsense in electronic contexts’.
1988 ’The transforming nature of meta- Organizational Dynamics
phors in group development: a (Autumn): 50-64.
study in group theory’. Human
Relations 41: 31-64. Wikstrom, S., and R. Normann
1994 Knowledge and value: A new per-
Star, S. spective on corporate transforma-
1992 ’The Trojan door: organizations, tion. London: Routledge.
work, and the ’open black box".
Systems Practice 5: 395-410. Winter, S.
1988 ’Knowledge and competence as
Starbuck, W. strategic assets’ in The competit-
1992 ’Learning by knowledge intens- ive challenge. D. Teece (ed.),
ive firms’. Journal of Manage- 159-184. Cambridge, MA:
ment Studies 29: 713-740. Ballinger.
Zuboff, S.
1988 In the age of the smart machine:
The future of work and power.
New York: Basic Books.