::''

Thepractice-based
perspective
onknowledge
What is knowledge?

-rapter 2 provided one specific answer to the question 'what is knowledge?, However,
--::robjectivist perspective
has been widely challenged, and for a number of different
::asons. Arguably the most fundamental challenge and critique of it is that it is based
.:: ilawed epistemological assumptions. Chapter 3 therefore presents an alternative
:::swer to the question 'what is knowledge?' This chapter is basedon fundamentally dif-
:::ent epistemological assumptions, and as will be seen, characterizesknowledge and
r:owledge management practices quite differently from the obiectivist perspective
:c Iable 3.1).
The practice-based perspective conceptualizes knowledge not as a codifiable
riect/entity, but instead emphasizesthe extent to which it is embeddedwithin and
.::separablefrom practice. cook and Brown (1999) labelled this perspectivean ,epi-
,:cmology of practice'due to the centrality of human activity to its conception of know-
,:j.ge. Further, Gherardi (2ooo, 218) argues that 'practice connects ,knowing, with
:':ing". Thus, the embeddednessof knowledgein human activity (practice)represents
,::e of the central characteristicsof this epistemologicalperspective.

Table3.1. Objectivist
andpractice-based
epistemologies
of knowledge

Objectivist epistemology Practice-based epistemology

- - .',ledgederivedlrom process r knowledgeis embeddedin practice

an intellectual
o knowing/doinginseparable
--,'.';edge
is a disembodied o knowledgeis embodiedin people
entity/object
r knowledgeis sociallyconstructed
- - .'.'iedgeis objective'facts' r knowledgeis culturally
embedded
o knowledgeis contestable
o knowledgeis sociallyconstructed
,: :it knowledge(objective) . tacitandexplicitknowledgeareinseparable
and
, :ged overtacitknowledge(subjective) mutuailyconstituted
:, ^ct knowledgecategories r Knowledgeis multidimensional
@ E P I S T E M O L O G I EOSF K N O W L E D G E

Table 3.2. Theoreticalperspectiverelatedto the practice-based

Dersoectrve

Writer Theoreticalperspective

E m p s o m( 2 0 0 1 ) Interprettve
B l a c k l e(r1 9 9 5 ) Activity Theory
Tsoukas(1996) Ethnom ethodoIogy/interpretive phiIosophy
Cook& Brown(1999) American Pragmatists
L a v e& W e n g e r( 1 9 9 1 ) Situated Learning Theory
Sayer(1992) CriticalRealism
Suchman(2003) Actor Network Theory

Practice to purposeful
refers humanactivity.
lt isbased
ontheassumption thatactivity
includes
both
physical
andcognitive andthatthese
elements, elements
areinseparable.
Knowledge useand
istherefore
development regarded
asa fundamentalaspectofactivity.

While the objectivist perspectivewas closely aligned with a positivistic philosophy, the
practice-basedperspective is compatible with a number of different philosophical per-
spectives(Table3.2). Another perspectivethat has much in common with the practice-
basedperspective,but has thus far not been utilized by knowledge management analysts
is Critical Realism(with the exception of Mutch 2OOr.2
The chapter follows a similar structure to Chapter 2, and begins by firstly outlining
the way knowledge is characterizedwithin the practice-basedperspective.Following this,
the chapter then examines how knowledge management processesare conceptualized.
As the chapter proceeds,the vast differencesthat exist between the practice-based,and the
objectivist perspectiveon knowledgeillustrated in Table3.1, should becomemore apparent.

Practice-basedperspectiveson knowledge

The practice-basedepistemologycan be understoodin terms of sevenspecific,but inter-

relatedfactors,eachof which are now examinedin turn (Table3.3).

The embeddednessof knowledge in practice

Perhapsthe most important difference between the objectivist and practice-basedepi-

stemologiesof knowledgeis that the practice-based
perspectivechallengesthe entitative
conception of knowledge.From this perspective,knowledge isn't regardedas a discrete

2 It is beyondthe scopeof this bookto examinein detailthe differences

betweenthesetheoretical
perspectives.
 P R A C T I C E - B A S EPDE R S P E C T I VOEN K N O W L E D G E
E
Table 3.3. Practlce-based
characteristics
of knowledge

Gharacteristics of knowledge from practice-based

epistemology

1. Knowledgeis embeddedin practice

2. Tacitandexplicitknowledgeareinseparable
3. Knowledgeis embodiedin people
4. Knowledgeis sociallyconstructed
5. Knowledgeis culturally
embedded
6 . K n o w l e d gies m u l t i d i m e n s i o n a l
7. Knowledgeis contestable

entity/object that can be codified and separatedfrom people. Instead,knowledge,or as

some of the writers from this perspectiveprefer, knowing, is inseparable.fromhuman
activity. Thus all activity is to some extent knowledgeable, involving the use and/or
developmentof knowledge.conversely,all knowledgework, whether using it, sharingit,
developing it, or creating it, will involve an element of activity. Blackler (1995, lOZ3)
rummed this up as follows, 'rather than regardingknowledge as something that people
have,it is suggestedthat knowing is better regardedas somethingthey do.,
As well as challengingthe knowing-doing dichotomy, this perspectivealso challenges
the mind-body dichotomy that is inherent to the obiectivist perspective(seeTable 3.4
'ater).As outlined, the objectivistperspective,drawing on the classicalimagesof science,
conceptualizesknowledge as being primarily derived from cognitive processes,some-
:hing involving the brain but not the body. The practice-based perspectiveinsteadviews
i.nowing and the developmentof knowledgeas occurring on an ongoing basisthrough
the routine activitiesthat peopleundertake.Knowing thus can be seenaslessof a purely
cognitive process,and more of a holistic processinvolving the whole body (Gherardi2000).
Thus, from this perspective,thinking and doing are fused in knowledgeableactivity, the
development and use of embodied knowledge in undertaking specificactivities/tasks.
Theseideascan be illustrated through consideringa number of examples.First, Orr,s
1990)widely referencedstudy of photocopier engineersemphasizeshow their knowl-
edge developed through a processof dialogue and improvization, which involved the
adaptationof existingknowledgeto new and novel situations.Similarly,Patriotta(2003),
1na study of a FiatAuto plant in Italy, showedthe embeddedness of knowledgein the nar-
rativespossessed by workers,and how thesenarrativesevolvedin the resolution of 'dis-
ruptive occurrences'(3a9).Thirdly, DeFillippi and Arthur (199g) in a study of film (i.e.
movie) production, showed that for apprentice technicians processesof learning by
rt-atchingwere crucial.Knowledgein this context tended to developthrough processes of
socialization,observation,and practice.The final example,of the traditional craft skill of
metalworking can be illustratedby a quotation:

\\'hen you have a bar of iron in front of you which has to be twisted and wrought into a certain
shape.. . . then you learn to apply ideasto things. You becomepractical.Youcannotthink the iron
E I EFSK N O W L E D G E
EPISTEMOLOGO

into the positionand shapethat is wanted,but you cannotdo it without thought.Your thoughts,
in your purpose,mustbelimited,circumscribed,
if you areto succeed bounddownto the factsof the
situation.McKinlay (7996, 86,emphasisadded)

This quotation alsoreflectswhat a growing number of authors arearguing (see,for example,

Alvesson 2000, 2001), that all work can be regarded as knowledge work, and that all
workers, whether bus drivers, cleaners, accountants, management consultants, oI
researchScientists,ale, to Someextent knowledge workers. However, this debate will be
examinedin more detail in Chapter 14.

Tacit and explicit knowledge are inseparable

Another point of departure between the objectivist and practice-basedperspectiveson

knowledge is in the way that the relationship between tacit and explicit knowledge is
conceptualized.The practice-basedperspectivesuggeststhat rather than tacit and explicit
knowledge representing separateand distinctive t)?es of knowledge, they represent two
aspects of knowledge and are in fact are inseparable, and are mutually constituted
(Tsoukas1996; Werr and Stiernberg2003). One consequenceof this is that there is no
such thing as fully explicit knowledge, as all knowledge will have tacit dimensions. Clark
(2000) usesthe term'explacit knowledge'to linguistically syrnbolizestheir inseparability
(Table3.4). For example text, which is often referred to as a form of codified knowledge,
has tacit components, without which no readercould make senseof it. Examplesof these
tacit elementsinclude an understanding of the languagein which they arewritten, or the
'The idea
grammal and syntax usedto structule them. Polanyi (1969, 195) suggeststhat,
of a strictly explicit knowledge is indeed self-contradictory; deprived of their tacit coeffi-
cients, all spoken wotds, all formulae, all maps and graphs, are strictly meaningless''

This book: partially explicit knowledge

Thisbookrepresentsa pieceof partially

explicitknowledge as an author
Firstly,
for two reasons.
I havenotbeenableto makefullyexplicitallthe ideas,assumptions, frameworks
theoretical and
valueswhichunderpin whatI havewritten.Fromthe pointof viewof the readerit canalsobecon-
sideredpartially asto readit yourequire
explicit, to havea goodgraspof the Englishlanguage,and
of otherrelevant
havesomeknowledge toplcs.
academic

T a b l e3 . 4 . C h a l l e n g i ndgi c h o t o m l e s

Ghallenging obiectivist dichotomies

Explacitknowledge(tacitand explicitknowledge)
activity(knowinganddoing)
Knowledgeable
Sensualcognition(brainand body)
 P R A C T I C E - B A SPEEDR S P E C T IO
VNEKNOWLEDGE

\Vhile, as outlined in Chapter 2, Polanyi'swork is often used to justify the tacit-explicit

::chotomy, a number of writers suggestthat this misunderstandshis analysis (Brown and
)uguid 2001; Prichard2000).Thesewriters challengethis and suSgestthat his analysisis
.rounded more in the practice-based perspective.

Knowledgeis embodied

.::e objectivist perspectiveon knowledge assumesthat knowledge can exist in a fully

.:..rlicit and codified form, that knowledge can exist independently of human beings.This
-,, sition is fundamentally challengedby the practice-based perspectiveon knowledge,
, ,::ich assumesall knowledge or knowing is personal.The practice-basedperspective
:.-,=refore assumesthat it is impossibleto totally disembodyknowledgefrom people into
, -:1ly-explicit form. This assumptionis thereforecloselyrelatedto, and flows from, the
.::'.ious two issuesexamined:that all knowledgehas tacit dimensions,and that knowl-
: :.E is embeddedin, and inseparablefrom practice.
-:e practice-basednature of knowing/knowledge assumesthat knowledge develops
'--::,ughpractice:people'sknowledgedevelopsasthey conduct activitiesand gain experi-
:.:=. Further, the inseparableand mutually constituted nature of tacit and explicit
,- -,";ledge means that it is not possibleto make such knowledge fully explicit. There will
, ,,:,r's be an element to which knowledgeresidesin the head/body of those who devel-
:":. and possessit. Thus while it may be possibleto partially convert tacit knowledge
- : an explicit form, in contradiction with the oblectivistperspective,the practice-based
:'::::ective assumesthat such plocessescan never be complete.For example,in terms of
, .:jttion most readersare likely to be familiar with from one context or another, con-
:.: :he nature of knowledgesharingin'master-apprentice'type relations,whete Some-
- = .rDeriencedattemptsto sharetheir knowledgewith a more inexperiencedcolleague.
* -: nature of the knowledge
::actice-basedperspectiveassumesthat the practice-based
'master' possesses means that this knowledge will be to some extent
on, :rpertise the
",- , .;ied, and cannot be fully articulatedand made explicit. Further,the practice-based
I ir - ::,:{tive aSSumeS that for the apprentice to learn the knowledge of the master requires
:t1i: ::€\' communicate,interact, and work together,typically over an extendedperiod

, *::her sensein which knowledge is embodied (and simultaneouslyembeddedin

'indeterminacy of practice'
rr!,: :. relatesto what Tsoukas(1996)referredto asthe ,where
llrr ::itntial distinctivenessof all situationsthat people act in requiresthem to continu-
urlll.::.-<epersonaljudgements.No matter how explicit and well deflnedthe rulesarethat
m; r;: ie action, there will alwaysbe someelementof ambiguity or uncertainty that cre-
umy r :ied for actors to make inferencesand judgements.For example, applying this
nr rr-.: :-. the perspectiveof the'apprentice' just discussed,no matter how formalized,
,[]r-rr.*-t:c.and explicit the knowledgethey have acquired,there will alwaysbe circum-
--:.atemelge where an element of judgement will be required. Thus, knowledge/
itfiiirtr.r-:.!
Umi:n,-,:,: urr-olvesthe active agencyof people making decisionsin light of the specific
iulrrr:- i::nces in which they find themselves.
@ E P I S T E M O L O G IOEFSK N O W L E D G E

The socially constructed and culturally embedded nature of knowledge

Two factors that are closely interwoven are that knowledge is socially constructed and
culturally embedded.It is thereforenecessaryto examine them simultaneously.In stark
contrast to the 'knowledge is truth' assumption of the objectivist perspective on knowl-
edge,where it is suggestedthat codifled knowledgecan exist in an objectiveform inde-
pendent of socialand cultural values,the epistemologyof practiceperspectivearguesthat
all knowledge is socially constructed in nature, which makes it somewhat subjective and
open to interpretation.Thus, knowledgeis never totally neutral and unbiased,and is, to
someextent, inseparablefrom the valuesof those who producedit.
As with the objectivistperspective,this viewpoint is basedon a particular understand-
ing about the nature of language.The objectivist perspectiveassumesthat languagehas
fixed and obiective meanings, and that there is a direct equivalence between words and
that which they denote. Instead,the practice-basedperspectivesuggeststhat language
has no such fixed meanings, and that in fact the meaning of languageis inherently
ambiguous. This subjectivity, or interpretive flexibility in language,thus.undermines any
claims about the objective status of any knowledge, whether it is highly tacit and per-
sonal, or whether it is partially explicit and codified. However,the socially negotiated
nature of languagelimits the scopeindividuals have to modify and interpret the meaning
and use of language(Sayer1992;Tsoukas1996).

istheprocess
making
Perspective through
whicha community
develops, its
andsustains
strengthens,
andvalues.
knowledge Perspective is
taking theprocess
through people
which an
develop understanding
oftheknow|edge,va|ues,and,wor|dview,ofothers.

The socially constructed nature of knowledge applies to both its production and its
interpretation. Polanyi (1969)referredto thesetwo processesas sense-givingand sense-
reading, while Boland and Tenkasi(1995) used the terms perspectivemaking and per-
spective taking. Thus both the production of knowledge, and the reading/interpretation
required to develop an understanding of it, involves an active process of meaning
construction/inference. For example, a written report is a piece of partially explicit
knowledge,whose meaning is constructedby its author/s. However,readersmay infer a
different meaning and analysis. This aspect of the practice-basedperspective therefore
has profound implications for the way knowledge is sharedand managed,asthe attractive
simolicitv of the transmitter-receivermodel is questioned.

Canyou thinkof an examplefrom yourorganizational

experienceof wherea rangeof peopleinferred
by the f luidityof meaning
partlybe explained
differentmeaningsf rom a report?Canthesedifferences
inlanguage?
 P R A C T I C E - B A SPEEDR S P E C T IO
VNEKNOWLEDGE

Further,this processof meaning construction/inferenceis typically culturally embedded.

The meanings people attach to language/eventsare shapedby the valuesand assumptions
of the social and cultural context in which people live and work. one way in which pre-
existing values and assumptions influence these processesof knowledge construction/
knowledge interpretation is through the flltering of data-information in deciding what is
considered'relevant'. A dramatic, and tragic example of such a filtering processwas one of
the contributory factorsto the ChallengerSpaceShuttle accident (Baumard 1999; Starbuck
and Milliken 1988).In this caseNASAengineersneglectedwhat tumed out to be important
information regardingO-ring erosion,asbasedon the assumptionsthey had, sucha situation
was regarded as presenting a minute risk. This cultural embeddednessresults in much
knowledgebeing context-specificand context-dependent,making its relevance,and trans-
ferability between contexts not necessarilyalwaysstraightforward.
The idea of knowledge being culturally embeddedlinks to the concept of collective
knowledge discussedin Chapter 2. Collective knowledge was shown to be culturally
embeddedin a number of different contexts/such as within communities of practice,or
tvithin the context of a national or regional culture. What distinguishesthe cultural
embeddingof knowledgein the practice-based perspectivefrom collectiveknowledgein
the obiectivist perspective,is that from the practice-basedperspectiveall knowledge is to
some extent culturally embedded.Thus from this perspective,none of the knowledge we
possess is totally separateand independentfrom the socialcontextsthat peopleoperatein.

The simultaneousmultidimensionalityof knowledge

The use of taxonomies,asillustratedin the previouschapter,suggeststhat all knowledge

can be classifiedinto distinctive categories,i.e. that it is either tacit or explicit, or that it is
tacit-collective,or explicit-individual, etc. This idea is questionedby a number of writers
trho suggestthat while such an approachmay have analytical,benefits,it misrepresents
the complexity of organizationalknowledge.Tsoukas(1996),for example,suggeststhat
dichotomies such as tacit-explicit and individual-group are unhelpful as they disguise
the extent to which theseelementsareinseparable,and mutually defined.Blackler(1995,
1032)makesa similar point by suggestingthat,

. . . it is a mistaketo assumethat embodied,embedded, embrained,encultured andencodedknow-

,:dgecansensiblybeconceived asseparate
from eachother.Knowledge
is multi-faceted
andcomplex,
:eing both situatedand abstract,implicit and explicit,distributedand individual,physicaland
:.lrental, developing andstatic,verbalandencoded.

Thus the practice-base perspectiverejectsthe taxonomy-basedapproachto categoriz-

-ng knowledge.For example,considerthe knowledge that an engineerusesto design a
car'schassis,or that a craftsman(/person)usesto assembleand build it. In both casesthis
inowledge is simultaneously individual and collective; tacit and explicit; physical and
nental; and abstractand situated.

--
nk of somespecificorganizationalknowledgethatyou possess.Canit be classified
intoa neat
:::egory,suchas tacit-collective,
or doesit havemultipledimensionssimultaneously?
E P I S T E M O L O G IOEFSK N O W L E D G E

The contestible nature of knowledge

The flnal key aspect of the practice-basedperspective is the acknowledgement that the
subjective, socially constructed, and culturally embedded nature of knowledge, means
that what constitutes knowledge is open to dispute. This therefore challengesand under-
mines the idea central to the objectivist perspective that it is possible to produce truly
'legitimate'
obiective knowledge. Thus, competing conceptions of what constitutes
knowledge can occur where different groups/individuals develop incompatible and con-
tradictory analysesof the same events, which may lead to conflict due to attempts by
these groups to have their knowledge legitimated.
One of the main consequenceswhich flow from this, therefore, is that issuesof power,
politics, and conflict becomemore important than are acknowledgedby the obiectivist
perspective.Most fundamentally, Michel Foucault'sconception of power/knowledge sug-
geststhat these concepts are inseparable(Foucault 1980; McKinlay 2000). Relatedly,
Storey and Barnett (2000) suggestthat all knowledge management initiatives require to
be seenashighly political, and are likely to be accompaniedby what they desqibe as'turf
wars'by different organizationalinterest Sroupsattempting to gain.some control over
these projects.The importance of acknowledgingand taking account of the contested
and political nature of knowledge is magnifled by the fact that this aspectof knowledge,
and knowledge management initiatives is typically either neglected or ignored by the
maiority of the knowledge management literature. Theseissuesare examined more fully
in Chapter 7.

:3ir:
I"-l:i1T::l:lT*:iig:liig::::Tr:ii9'i1Y:I
pharma-co is a UK pharmaceutical company.Untilthe early1980sit hadbeena government-
ownedresearch andby the mid-1990stherewas stillevidence
laboratory, in partof the company
of the technically focusedculturewhichhadhistorically predominated. During the mid-1990sa
decision was madeto implement a new information management system. The dominantrheto-
ric usedby the projectteamto justifythe needfor changewas thatthe changing natureof their
marketsrequiredsignificant changes to be made to improve the competitiveness of theirpro-
ductionfacilities. An importantfigureto Pharma-co's project was the World Manufacturing
Director,who stronglychampioned it. Whenthe projectstartedhe hadbeena relatlvely recent
recruitto the organrzation.As partof Pharma-co's long-term Strategy of adoptingmorecommer-
cialandcost-sensitive operatingpractices a need had beenidentified to introduce suchattitudes
to itsseniormanagement. Therecruitment of theWorldManufacturing Director was oneof these
'commercial' knowledgefrom workingoutsideof the companywas
appointments. Thus his
highlyregarded by seniormanagement. However, resistanceto the proposed changesemerged
from middlemanagers withinthe production function.They suggested the proposed changes
werefundamentally unnecessary,andthatPharma-co couldremaincompetitive through staylng
focusedon the development andproduction innovative
of technically products. Thetraditional cul-
turewhichhadbeenhistorically predominant within Pharma-co was focused around production.
 P R A C T I C E - B A S EPDE R S P E C T I VOEN K N O W L E D G E E
- - e o f t h e m a i n f a c t o r ss t r e n g t h e n i n gt h e a r g u m e n to f p r o d u c t i o nm a n a g e m e n tw a s t h e i r
:::ailed knowledge of the company's internal manufacturingpractices.Thus at the start of
r-aTma-co'schange projectthere was a highly politicalconflict between those for and against
--:nge which centred on the validityof their knowledgeand the way they used it to legitimate
--: r dlfferentanalysesof the extent to which changewas needed.

- s:ch situations, to what extentis it possibleto objectively

evaluatethe competingargumentsand
: = : : d eo n t h e ' c o r r e c t ' c o u r soef a c t i o n ?
^ of the interestsgroupsayaboutthe culturalembeddedness
" tt doesthe differentperspectives
-' <rowledge?To what extentarethe viewpointsof thosein conflictderivedfrom the valuesand
communities
:=:s of the organizational they areembeddedin?

lmplicationsfor the nature of the organizationalknowledge base

- re above outlined characteristicsof knowledge have profound implications with regard

: - the nature of organizationalknowledgebases,as a growing number of writers recogn-
.-e. The practice-based perspectiveon knowledgesuggeststhat rather than being unitary
.:d coherent, organizational knowledge basesare in fact fragmented and dispersed,
-eing made up of specializedand specificknowledge communities, which have some
:=qree of overlapping 'common knowledge' (Kogut and Zanger 1992). This led Brown
:::d Duguid (1991,53) to suggestingthat organizationsrequire to be conceptualizedas a
: rmmunity-of-communities', and Blackler et al. (2000) as decentred and distributed
,::orr'ledgesystems.Finaliy, as will be seenin the following section,theseinsights have
::tormous implications for the sharingand managementof knowledgein organizations.
The fragmentationof the organizationalknowledgebaserelatescloselyto the idea that
.::orvledgeis embeddedin practice.Typically,the practicesundertakenby organizational
.'J, and hence the knowledgethey possess, are localizedand specific,being shapedby
::-e particulardemandsof their context (local customers,market conditions, characterof
-..ional/regional regulation and legislation,etc.).The degreeof fragmentation and spe-
--"iization will alsobe relatedto the culture of the organization,and the extent to which
: -ncouragesand supportsautonomousor standardizedworking practices.

Autonomous businessunits and specializedknowledge communities

- :.ved-Truck, whichsold,rented,andserviced forklifttrucks,workhadhistorically beenorgan-

e u s i n e susn i t sw
: = t r t o s m a l ld, i s c r e t b s i t h i ns p e c i f i c
, h i c hh a dr e s p o n s i b i lfiot yr a 1bl u s i n e sw
:=:3'aphicregions. Withinthisstructure, therewas littleneedfor interaction betweenbusiness
-
, : s .a n dt h e yo p e r a t eadsv i r t u asl t a n d - a l o n be s .h i l ee a c hb u s i n e susn i ti n p r i n c i p l e
u s i n e s s eW
, - : :he samerangeof products andservices, in realitytheyhadsignificant autonomyoverhow
-:, Cidthis.Thiswas becauseboththe natureof the marketandcharacter of customers varied
E E P I S T E M O L O G I EOSF K N O W L E D G E

signlficantlyfor each business,and alsothat managementin each businessunit offered different

levelsof serviceand support.The autonomyof the businessunitswas such that the evolutionof
their working practices,the upgradingof their lT systems etc., was done purelyon the basisof
localconsiderations.Thus,discreteand specific knowledgecommunitiesdeveloped,with staff in
each businessunlt possessingsubstantialamounts of specializedknowledge,reievantto their
own localizedworking practices,and customer demands,which had limited transferabilitvand
relevance,in other businessunits.

ls the existenceof suchspecialistcommunities,with theirown knowledgebasesandways of

workingnecessarily a problemfor organizatlons?
Towhat extentis it possiblein multidivisional
corporations
to balancethe conflicting
demandsof
providingdivisionsthe autonomyto work independently and havesome levelof standardization
acrossthe corporation?

Not only can the knowledge of organizational communities be different (i.e. specialized
and specific),but it may also be basedon qualitatively different assumptions,values, and
interpretativeframeworks.Brown and Duguid (2001)referredto these as 'epistemicdif-
ferences'. For example, the communication and interaction difficulties between staff
from different functions of an organization (such as production and R&D, or finance and
R&D), or between staff from different disciplinary backgrounds (such as in a multidiscip-
linary project team) can be to some extent explained by such differences.As will be seen
in Chapter 6, where this issueis explored in detail, this significantly affectsthe dynamics
of knowledge-sharingprocesses. Finally, these issuesare again examined in Chapter 13
which examines knowledge-sharing within the context of global multinationals, where,
what Becker(2001)referredto this asthe problem ot'large numbers',meansthat asorgan-
izational size increases,so do the problems in managing an increasingly fragmented
organizationalknowledgebase.

Thinkaboutan organizationyou haveworkedin.Was its knowledgebasefragmented? Further,what

factorsinfluencedthe natureof the knowledgebasemore:the managementcultureor the diversity
of localconditions?

A practice-basedperspectiveon the managementand

sharing of knowledge

Having consideredin detail how the practice-basedepistemologyconceptualizesknowledge

it is now time to examine the implications of theseideasfor understanding the characterof
organizationalknowledge-sharingand knowledgemanagementprocesses(seeTable3.5).
One of the central components of the practice-basedperspectiveon knowledge man-
agementis that it eschewsthe idea that it is possiblefor organizationsto collect knowledge
 P R A C T I C E - B A S EPDE R S P E C T I VOEN K N O W L E D G E E
Table 3.5. A practice-based
perspectiveon knowledgemanagement

Knowledge management from a practice-based epistemology

'perspective 'perspective
requires
Knowledgesharing/acquisition making'and
of tacitassumptions
an understanding
taking'-developing
Knowledgesharlng/acquisition through
- ' r i c h ' s o c i ailn t e r a c t i o n
- immersionin practice-watchingand/ordoing

3 . Managementroleto facilitatesocialinteraction

- , .ether into a central repository,or for middle and seniormanagersto fully understand
- ,= knowledgeof those who work for them (Goodall and Roberts2003). Tsoukas(L996,
. I quoting Hayek, suggeststhat a belief in the ability to achievesuch a staterepresents
-:-: slnoptic delusion . . . that knowledgecan be suweyedby a singlemind.'Thus man-
.:.:ial understandingof organizationalknowledgewill alwaysbe fragmentedand incom-
. :.i, Zlrldattempts to collect knowledge in a central location likely to be limited. The
' .-rn'ing quotation from Tsoukas(7996,22) sums this up, and points towards the
'the key
:::;tice-based perspective/sconceptualizationof knowledge-sharingprocesses:
' ,;hieving coordinatedaction doesnot so much dependon those"higher up" collecting
* .t and more knowledge,ason those "lower down" finding more and more waysto get
, :-::ectedand interrelatingthe knowledgeeachone has.'
-:e practice-based perspectivefurther suggeststhat the transmitter*receivermodel of
- ',i1edge-sharing is questionablebecausethe sharing of knowledge does not involve
' - . '-mple transferral of a fixed entity between two people. Instead,the sharing of knowl-
: : i: involves two people actively inferring and constructing meaning. This perspective
,<,.eststhat to be effectivethe sharing of knowledgerequiresindividuals to developan
,.'::.ciation of (someof) the tacit assumptionsand valueson which the knowledge of
--.::s is based-the processesof 'perspectivemaking' and 'perspectivetaking' outlined
: r:.,:r by Boland and Tenkasi(1995).This challengesthe assumption embeddedin the
':.:-:nitter-receiver model that the knowledgeexchangedin suchprocesses is unchanged.
1, ,'."ni and Scarso (2000) suggest the practice-basedperspective on knowledge-sharing
.::t:sents a'language game'/ due to the importance of dialogue and languageto such
;-.-:sses.BolandandTenkasi(1995,358)arguethateffectiveknowledge-sharinginvolves,
r tr'-rc€ssof mutual perspective taking where distinctive individual knowledge is
: i .:3nged, evaluatedand integratedwith that of others in the organization.'
-
.= iogic of the 'languagegame' model complicatesthe nature of knowledge-sharing
r- i,:sses,as the inherent ambiguity of language,combined with the fact that those
- .',.edin the knowledge-sharingprocesshave different cognitive frameworksmeans
" ;: ::rereis alwaysscopefor differing interpretations.Thus, as you read this book, the
Tr::a-1g you take from a pieceof partially explicit knowledge may vary from the mean-
" i - -ltend to convey.
-,:se
perspective-making, and perspective-taking typically requirean extens-
processes
', : ::]tourit of socialinteraction and face-to-facecommunication, which is a conclusion
E EPISTEMOLOGO
I EFSK N O W L E D G E

reachedby a number of empirical studies(see,for example,Lam 1997,2000; Leonard-

Barton 1995;Swanet al. 1999).The acquisitionand sharingof knowledgetypically occur
through two distinct, but closelyinterrelatedprocesses:

1. Immersion in practice-for example learning by doing, or learning by watching.

2. 'Rich' socialinteraction-for example,an interaction which allows people to develop
some level of trust with each other, as well as develop some insights into the tacit
knowledge,values,and assumptionsof eachother.

Theseprocessesare interrelated becauselearning by doing is likely fo simultaneously

involve an element of social interaction, and vice versa,the sort of 'discursivepractice'
referredto by Gherardi (2OOO,221).

Swed-truck: an example of practice-basedknowledge management

In the late 1990s Swed-Truck(describedearlier-see pages 35-6) decided to implement an

organization-wideinformationmanagementsystem, with the objectiveof introducinga greater
level of coordinationand standardization
across its businessunits. To implement this it used a
sociallybased model of knowledge-sharing,
which made extensiveuse of intensivesocialinter-
action.This can be consideredby examiningthe system developmentphase only.The system
being implementedinvolvedthe introductionof a common informationsystem acrossa signific-
ant numberof differentbusinessunits.As outlinedearlier,their businessunits had operatedquite
autonomouslyfrom each other, and as a consequencehad developedtheir own specialized
knowledgebases.The projectteam decidedthat the developmentand implementationof a com-
mon informationmanagement system requiredthe utilizationof this distributedknowledge,
which was achievedthroughthe creationof a projectteam bringingtogether staff from a range
of their businessunits (who worked part time on the project).As a substantialamount of devel-
opment work was necessary,this processlastedfor a year.This interbusinessunit projectteam
worked intensivelywith consultantsto develop common systems that were compatiblewith
the diverse needs of their different business units. While the project was not without its
problems and delays, the project was deemed a success, and was implemented close to
predictedtime-scales.

To achieverichsocialinteractions to get peopletogetherface-to-face?

is it necessary

How importantto the successof knowledge-sharing

processesis the existenceof trust between
participants
to suchprocesses?

From a practice-based perspective,the managerialrole is thereforeto encourageand facil-

itate the type of communication and social interaction processesthat will allow effective
perspectivemaking and taking to occur.This can be done through an enormously diverse
 P R A C T I C E - B A SPEEDR S P E C T IO
VNEKNOWLEDGE E
' i :-:e of waysincluding (to highlight
iust a few examples):
o ::\-elop a knowledge-sharing culture (through rewarding people for sharing);

" ::;litating the development of organizational communities of practice;

. -;.rriding forums (electronicor face-to-face)
which encourageand support knowledge-
::-aring;
. :::)lement a formalized 'mentoring' system to pair experiencedand inexperienced
"._:kers

l:ese issuesare examinedin more detail in subsequentchapters,with Chapter 4 look-

- . jt generalissuesof motivation to shareknowledge,Chapters5 and 6 looking at the
:'=:inc dynamics of knowledge-sharing within and between communities, Chapter 7
, .lng at the political nature of knowledge-sharing,while Chapter 9 considersthe role
' - j- Luganizationscan play through their human resourcemanagement policies and
--:ire management practices.Finally, Chapter 8 considersthe role that information
,:=ns may be able to play in facilitating perspectivemaking and taking processes.

6onclusion

- - : rclusion, Chapters 2 and 3 have outlined two distinctive

epistemological perspect-
:: '"rhich characterizeknowledge in extremely different ways (seeTable 3.1). These
"n,:.:tctives also conceptualized knowledge-sharing and knowledge management
: - .:-tses differently. They therefore have very different managerial implications with
' ::':d to how knowledgemanagementefforts should be organizedand structured:

. :',;titist perspective:focus on the codification and collection of knowledge, cleate

--:chanisms to allow this knowledge baseto be searchedand accessed,such assetting up
. :tarchable databaseand encouraging staff to codify their knowledge and store it there.
. :-.;-iice-basedperspective:facilitate interpersonal knowledge-sharing through diverse
- ::ns of interaction and communication, such asdeveloping the levels of trust between
-::: membersof a new proiect team through allowing them to interact extensively face-
- . -:ace(perhapsin both work and socialcontexts)at the initial stagesof the project.

.nrvrrwouESTl
[

::: '' instructionson how to conducta certaintask.What tacit knowledgeis necessaryfor

: - :c make sense of them? What does this say about the inseparability
of tacit and explicit
-:,', edge?

: - : - , o u t h i n ko f a n e x a m p l ef r o m y o u r o w n e x p e r i e n c e
o f w h e r e t h e r e h a s b e e nd i s p u t ea n d
. - -= ct between competingknowledgeclaims?What politicaltacticsand strategiesdid the
'expertise'
- - -' cting partiesutilizeto justify their position?Did they use external as a way of
: : : ^ a l i z i n gc l a i m s ,e t c ?
E EPISTEMOLOGO
I EFSK N O W L E D G E

t Compare the two perspectiveson knowledge outlinedin Chapters2 and3. Whichonemore

closelymodelsthe natureof knowledgein the organizations
thatyou haveworkedin?lf
theseorganizations
implemented knowledge management initiatives,
whichepistemological
perspective
weretheybasedon?Didthisaffectthe successof theseinitiatives?

F Blackler(1995).'Knowledge,Knowledge work andorganizations:

An overviewano
lnterpretation',Organization
Studies,1616:1021|-46.
Widely referencedarticle that advocatesadopting a practice/activity-basedview of knowledge

S. CookandJ. Brown(1999).'BridgingEpistemologies:
The GenerativeDancebetween
organizational
Knowledge
and organizationalKnowing',organizationscience,1ol4:381-400.
Links together the objectivistand practice-basedperspectivesinto a unitary framework.

H. Tsoukas(1996).'The Firmas a Distributed

KnowledgeSystem:A Constructionist
Approach',
StrategicManagementJournal,1J (WinterSpeciallssue):11-2S.
Argues that organizationalknowledge basesare highly distributed.

Specialissueof Organizationon'Knowing in practice,(2OOO],712.

A collection of theoreticaland empiricalpapers atl embedded in the practice-basedperspective,but
utilizinga diversity of theoreticalframeworks.