The Educational Forum

ISSN: 0013-1725 (Print) 1938-8098 (Online) Journal homepage: https://www.tandfonline.com/loi/utef20

Distributed Leadership

James P. Spillane

To cite this article: James P. Spillane (2005) Distributed Leadership, The Educational Forum,
69:2, 143-150, DOI: 10.1080/00131720508984678

To link to this article: https://doi.org/10.1080/00131720508984678

Published online: 30 Jan 2008.

Submit your article to this journal

Article views: 34857

View related articles

Citing articles: 197 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=utef20
 Distributed Leadership
by James P. Spillane

TIJE Forum
Stories of leadership successes follo w afamiliar structure: A charismatic
leader, often the CEO or school principal, takes over a struggling school, estab-
lishing new goals and expectations and challenging business as usual within the
organization. This leader creates new organizational routines and structures
that with time transform the school's culture, contributing in turn to greater
teacher satisfaction, higher teacher expectations for students, and improved
student achievement.

Stories in the "heroics of leadership " ge nre, how ever, are p roblemati c for a t least
two reasons. First, these epics equate school leadership chiefly w ith an individual leader-
typically the school principal. Th is is inaccurat e because school p rin cip als, or any othe r
leader for that matter, do not single-ha nded ly lead scho ols to grea tness; leadership in-
volves an array of individuals w ith va rious tools and stru ctu res . Though sch ola rs ha ve
long argued for moving beyon d th ose a t th e top of organizations in stu d ies of leadership
(Barnard 1938), the "heroics of leader ship " genre per sist s. Th e sec ond problem with
these accounts is their inatten tion to leadership p ractic e. They dw ell mo stl y on the "what"
of leadership-structures, functions, rou tines, and roles- ra ther th an th e "how" of schoo l
leadership-the daily perform ance of leadership ro u tines, fu nctions, an d structu res
(Hallinger and Heck 1996). Lead ership p ract ice centers not only on w hat people d o, but
how and why they do it. Und erstand ing leader ship p racti ce is im perative if research is
to generate usable knowledge ab out and for school leader ship . Distributed leader ship is
a recent antidote, or more corr ectly a se ries of an tidotes, to th e w ork in th e heroics of
leadership.

Distributed leadership h as ga rnered cons idera ble atten tion in th e United Stat es and
abroad. It often is used interch an geably w ith "sh ar ed leader ship ," " team leader ship,"
and "democratic leadership." Some use d istributed leader ship to indicat e th at scho ol

The Educational Forum • Vall/me 69 • Wi nter 2005 • 143

Spillane

leadership involves multiple leaders; others argue that leadership is an organizational

quality, rather than an individual attribute. Still others use distributed leadership to de-
fine a way of thinking about the practice of school leadership (Gronn 2002; Spillane,
Halverson, and Diamond 2001, 2004). Distributed leadership's popularity likely has to
do with how easily people can use it to relabel familiar approaches. It is little wonder
that many observers are perplexed about the meaning of distributed leadership and
whether it is anything new. Perhaps distributed leadership is just another case of old
wine in new bottles.

My understanding of dis-
tributed leadership, based on
The Distributed Leadership Study
Distributed leadership often is (School of Education and Social
Policy at Northwestern Univer-
cast as some sort of monolithic sity 2004), an elementary school
construct when, in fact, it is leadership research study, is out-
lined. The following question is
merely an emerging set of ideas addressed: What does it mean to
take a distributed perspective on
that frequently diverge from school leadership? My intent is
one another. not to provide a comprehensive
review of different perspectives
or identify the "one best" defi-
nition, but to layout my own
definition of distributed leader-
ship. An overview of distributed leadership, in which key terms and ideas are intro-
duced and defined, is provided. I next address how leadership is distributed over an
interactive web of people and situations, examining how leadership is spread over both
leaders and followers given key aspects of their situation, including organizational rou-
tines, structures, and tools. I then illustrate how this definition of distributed leadership
is a case of new wine-not new bottles for old wine-and consider its implications for
research, practice, and leadership development.

Putting Leadership Practice Center Stage

Distributed leadership is first and foremost about leadership practice rather than
leaders or their roles, functions, routines, and structures. Though they are important
considerations, leadership practice is still the starting point. A distributed perspective
frames leadership practice in a particular way; leadership practice is viewed as a product
of the interactions of school leaders, followers, and their situation. This point is espe-
cially important, and one that is frequently glossed over in discussions of distributed
leadership. Rather than viewing leadership practice as a product of a leader's knowl-
edge and skill, the distributed perspective defines it as the interactions between people
and their situation. These interactions, rather than any particular action, are critical in
understanding leadership practice. Too frequently, discussions of distributed leadership
end prematurely with an acknowledgment that multiple individuals take responsibility
for leadership in schools. This "leader plus" view, however, is just the tip of the iceberg

144 • The Educational FOri/III • Yolume 69 • Willter 2005

 Ess ay s

because, fro m a distr ibuted p ersp ective, leadership practice that resul ts from interactions
among lead ers, followers, and their situa tion is cri tical.

Some educators m ight arg ue that th is is merely sema n tics, pointing ou t th at leader-
sh ip scholars have long recognized th e im portance of th ese interactions an d acknow l-
edged th at lea ders hi p typ icall y invol ve s more p eople than those at the top of th e o rgani-
za tio nal hierarch y. My argu men t is no t sim p ly th at situa tion is im p ort ant to leadership
p ractice, but th at it actuall y constitu tes leadership practi ce- situ ation d efin es leader-
shi p practice in in teraction w ith lead ers and foll ow er s. This way of th in kin g abou t situ -
atio n d iffers substan tia lly from prior work.

People and Practice

Equating leadership w ith th e actio ns of th ose in lead ership positions is inad equat e
for three reasons. Fir st, leadership practice typ icall y in vol ves multiple leaders, som e
with and so me w ithout formal lead ership positions. It is esse n tial, therefore, to move
beyond view ing leader ship in terms of superh u m an actions. Secon d , leadership practic e
is not so me thing done to foll ow ers. From a d istributed perspect ive, followers are one of
th e three constituting elemen ts of lead ership practice. Third, it is not th e ac tio ns of in di -
vi d uals, but th e in te ractio ns among th em, th at are cr itica l in leadership practice.

Exis ting sch olarship show s th at responsibility for leadersh ip fu nc tio ns can be di s-
tributed in variou s wa ys . Studi es have show n how thi s responsibi lity can invo lve mul-
tiple leaders-not ju st princip als or coprincipals-who work in a coo rdi na ted m anner at
times an d in p arallel at others (Heller and Fires to ne 1995). Recent work in m ore th an 100
U.S. schools showed th at resp onsib ility for leadership func tio ns typically wa s di stri b-
uted among three to se ve n peopl e, inclu di ng ad ministra tors and sp ecia lis ts (Cam bu rn,
Row an, an d Taylor 2003).

The Distributed Leadership Study also showed th at responsibility for leadership rou-
tin es involves multiple leaders, th ough th e number involved d ep ends upon th e ro u tin e
an d subject area. Some rou tines, such as m onitoring and evaluating teach in g practice,
in volve fewer leaders (typically th e princip al and assista n t principal), com pared w ith
routines such as teacher d evelopment in literacy, which often in volve th e principal, cur-
ricular specialis ts, and lead teach ers. Th e exten t to w hic h respons ibi lity for leader ship
routin es was di stributed differ ed by school subject, w ith few er leader s in volved in lead-
ershi p routines for mathemat ics th an for literacy. For exa m p le, a t Ad ams Elementa ry
Sch ool, th e principal, literacy coo rdi na to r, curricu lu m specialis t, and lead teachers were
freq ue n t an d active parti cip ants in exec u ting leadersh ip rou tines for literacy. Conversely,
lead ership ro u tin es for mathematic s instruction were typicall y d efined by one of four
lead mathemati cs teach ers (Sp illa ne , Diamond, an d [ ita 2003).

Lead ers act in si tua tio ns th at are defined by o th ers ' ac tio n s . Fro m a distribu ted
perspective, it is in these in teraction s th at lead ership practice is co ns truc ted . The
Distributed Leadership Study's an a lysis of lead ership performa nce d ocuments h o w
leadership practice is d ef in ed th rough th e in te rac tio ns of tw o or m o re lea ders. When
observing lead ership ro utines for lit eracy ins tructio n a t Ada ms Elemen tary Sch oo l,

The Educational Forum • Volume 69 • W inter 2005 • 145

Spillane

one immediately notices how leadership practice becomes defined in the interac-
tions of leaders and followers . These leadersh ip rou tines often involve some co mbi-
nation of four leade rs: the p rincip a l, th e school's li te racy coordi na to r, the African-
America n H eritage coord inator, an d a teacher lead er. A t tim es, th ese lead e rs ' ac tions
parall el or overlap one an other; a t o ther times, th ey d o n ot. Th e principal e m p ha -
siz es goa ls an d s ta n dards, keeps th e m eetings m ovin g, su m marizes comments, an d
re mi n ds p a rt icip ants of w ha t is expec te d in their classrooms. The lit er acy coo rd ina -
tor identifies p roble m s w ith lit eracy instruct ion , suggests so lu tio ns a n d resources,
a nd encourages teachers to presen t th eir id eas. The teache r lead er d escr ibes h is or
her efforts to imp lement a teach in g stra tegy th at th e li te racy coordinator sha red . The
actions of fo llowers (in this case, p ri ma rily classroom teachers) also co n tri bu te to
de fini ng lead ership p ractice . Th ey p rovide kn owledg e abou t a p articular teach in g
s tra tegy-know ledge th at so me times is u sed b y leaders to ill ustra te a p o int about
im p roving lit eracy in struction.

Leade rs hip p racti ce takes

fo rm in th e intera cti ons b e-
tw een lead ers a nd fo llowers,
rather th an as a functio n of one
Distributed leadership is first or m ore le a d e r s ' ac tio ns
and foremost about leadership (Sp illane et a l. in p ress). In d i-
v id uals p la y off one a no ther,
practice rather than leaders or cr eating a reci procal interde-
p enden cy between their actions.
their roles, functions, routines, The D ist ributed Lea ders hi p
and structures. Stud y iden tifie d in terde pe n -
d ency as th e p r im a ry charac-
ter istic of in te rac tio ns amo ng
leade rs. Th is theo ry has been
informed b y the work of orga-
n izational th eori sts (Thom pson 1967; Ma lo ne e t al. 1999) . Three types of in terdepen-
d encies identified b y Thom pson (1967) - rec iprocal, pool ed, a n d se quen tia l-served
as th e basis.

Leader ship practice can be spread across tw o or more leaders w ho work se para tely
ye t in terdependently. Th e leader ship p ractice used in m onitor in g and evalua ting teach-
ing a t Ellis Elemen tary is illustrative. The principal believes that biannual visits are in-
adequate to evalua te a teacher 's practice . She and th e assista n t p rincip al developed a
compreh ensive ro u tine for m onitor ing and evalua ting teach ing practice . The assista n t
p rincipal, w ho has a good ra p port w ith tea chers, visits classrooms fre quently to conduct
for ma tive evalu atio ns and give regul ar feedback to teachers. The prin cip al engages in
su m ma tive evalua tions th rough her biannual visi ts to cla ssrooms. Through for mal and
inform al meet in gs, th e principal and ass istan t p rinc ip al p ool thei r informa tion to d e-
velop an under standin g of tea ch ers' p ractices. Throug h thi s " pooled" interd ependency,
these tw o leader s' se parate actions interact to d efin e a collective p rac tice for monitoring
and evalua ting teach in g.

146 • Tile Educational Forum • Volume 69 • Winter 2005

 Essays

Sometimes separate leadership practices are spread over the actions of two or more
leaders and must be performed in a particular sequence. In these cases, multiple inter-
dependent tasks, arranged sequentially, are critical to the performance of a leadership
routine. For example, the five-week assessment at Adams School illustrates how leader-
ship practice can be stretched over leaders over time. This assessment involves seven
stages performed in a specific order:
• The literacy coordinator creates the student assessment instruction.
• Teachers administer the assessment.
• The literacy coordinator
and her assistant score and
analyze the results.
• The principal and literacy
coordinator meet to dis- Leadership practice takes form
cuss the assessment re-
sults, using information
in the interactions between
from classroom observa- leaders and followers, rather
tions to diagnose prob-
lems. than as afunction ofone or
• The literacy coordinator
compiles resources and
more leaders' actions.
strategies that might en-
able teachers to address
the problems identified
through the analysis of as-
sessment data.
• The literacy coordinator reports assessment results to teachers during literacy com-
mittee meetings.
• The literacy coordinator, principal, and teachers interpret assessment results and
identify instructional strategies to address problem areas.

This sequence illustrates coordinated leadership. The term "coordinated" is used to

emphasize that leadership practice that involves a sequential interdependency must be
performed in a particular sequence.

People, Place, and Practice

Leaders typically have interaction with others. They also have interaction with as-
pects of the situation including a variety of tools, routines, and structures. Tools include
everything from student assessment data to protocols for evaluating teachers. The five-
week assessment described here is an example of a routine. Structures include routines
such as grade-level meetings and the scheduling of teachers' prep periods. From a dis-
tributed perspective, these routines, tools, and structures define leadership practice; the
situation both enables and constrains leadership practice.

Aspects of the situation define and are defined by leadership practice in interaction
with leaders and followers. Structures, routines, and tools are the means through which
people act. Yet, these same structures, routines, and tools are created and remade through

The Educational Fomm • Volume 69 • Winter 2005 • 147

Spillane

leader ship practic e. Th e di stinction b etw een th e os tensive an d performati ve as pects of

organ iza tion al routines (Feld m an an d Pentland 2003) is help ful.

Th e os te nsive aspect refers to th e " ro u tin e in principle," while the id eali zed ve rs ion
of the performative aspect refers to the routine in practice in particular places and at
particular times. For example, th e seven stage s of the five-week as sessment represent
th e ost ensive aspect of thi s routine, while reporting stu d en t asses sment results to teach-
er s in a liter acy committee meeting is the performati ve aspect of the routine. Th e ost en-
sive aspect frames practice-
both ena b ling an d constrain ing
it. Pr actic e creates and recreates
th e os te ns iv e as pect. Though
Descriptive theory building Feldman a n d Pentland (2003)
is essential before causal links confined their di scussion to or-
ganizational routines, ostensive
between distributed leadership, and performative di stinctions
can be a p p lied to other aspects
instructional improvement, and of th e situa tion, including struc-
student outcomes can be tures and tools.

established. Student assessment d ata, a

widely us ed leader ship tool by
all sch ools in The Dist ributed
Leadership Study, is a good ex-
ample. In an effort to reflect the di st rict's policy of holding schools accountable for stu-
dent achievement, th e stu d en t assessment data tool framed leader ship practice in a par-
ticular w ay across all sch ools by focusing leadership practice on curriculum content
coverage. The stu d en t assessmen t d ata tool, how ev er, w as transformed differently in
and through leadership practice at ea ch sch ool. In som e schools, asses smen t data were
reported, problem areas were identified, and specific topics on w h ich teacher s sh ou ld
focus w ere presented a t faculty m eetings. In other sch ools, assessm en t data w ere used
differently. At Baxter School, for example, assessment data were di saggregated and used
as th e basis for ongoing conversations abou t instructional improvement and curricular
priorities.

Sometimes tools d esi gned for other purposes are appropriated for leadership. At
Hillside School, stu den ts ' "writing folders" -designed for classroom w riting in struc-
tion-have become a core leadership tool. The key leadership routine is th e monthly
review of these folders by the school principal. Every teach er sub m its a folder contain-
ing one composition w ritten by each stu d en t in hi s or her class. The principal reads each
student's work and provides teacher s and students wi th written feedback. The leader-
ship practice in th is exa m p le is d efined in the interactions of th e principal and th e writ-
ing fold er s, as well as those betw een teach ers and stu d en ts. Through thi s monthly rou-
tine, writing folders ha ve been redesi gned as a leadership tool. In turn, th e writing folder
fundamentally shapes a leadership practice grounded in what students are learning about
writing an d engaging tea ch ers and stu d en ts in improving writing instruction.

148 • The Educational Forum • Volum e 69 • Winter 2005

 Essays

Is th is perspective on situation new? After all, contingency th eorists have long ma in-
tained th e im portan ce of situation to leadershi p . Leadership circ umsta nces in fluence
leaders' ac tions, as well as th eir effec t on followers (Bossert et al. 1982; Murp hy 1991).
From a contingency perspective, situa tio n works independen tly to influ en ce a leader 's
beh avior or med iat e its effects. A distribu ted p erspecti ve d iffers in at lea st tw o respects.
Firs t, situa tio n d oes n ot sim p ly affect w hat school leaders d o as an in d epen den t, ex ter-
nal variab le. Rather it d efin es leadership practice in interacti on wi th leaders an d follow-
ers. Second, th ere is a tw o-w ay relatio ns h ip between situatio n and practice. As pects of
th e si tua tio n can either ena ble or cons trai n practice, w hi le practice can transform th e
si tua tion.

A Case of Old Wine in New Bottles?

The answer to thi s quest ion depends on the particul ar d efinition of di stributed lead-
ers hip being considered. Dist ributed leadership oft en is cast as so me sor t of monolithic
construct when , in fact, it is m er ely an em erging se t of id eas that frequently diverge
from one ano the r.

The di stributed p erspective on leadership in thi s p aper gives center stag e to leader-
ship p ractice. Th ough scho lars have viewed leader sh ip as a beh avior or act for so me
time (Fied ler 1973), th is work equates leadership practice wi th th e acts of individual
leaders. From a di stributed p ersp ect ive, leadership practice tak es shape in th e in te rac -
tions of leaders, foll ow ers, and th eir si tuatio n, thus breakin g new grou nd rather th an
sim ply relabeling old ideas.

Shared lea dersh ip, team lead ership , an d d emocr at ic leadership are n ot synonyms
for d ist r ibuted leadership . Dep end in g o n th e sit ua tion , a d istributed p erspecti ve al -
low s for shared leadership. A team lead ership ap proach d oes n ot n ecessarily in vol v e
s ubscribi ng to a distrib u te d p erspect ive in w hich leadership p ractice is v iewed as
th e in teractio n of leaders, fo llowers, an d situ a tion. Simi la rly, a d istributed p erspec-
tiv e a llows for leadership th at ca n be d emocratic or autocra tic. From a distributed
p ersp ective, lead ership can be stre tched over leaders in a sc hool but is not n ec essar-
ily d emocratic.

Dist ributed leadership is considered by some educators as a cure-all for all th at ails
schools-an op in ion to which I d o not subs cribe. Dist ributed leadership is a p erspec-
tive-a con cep tu al or dia gnost ic tool for th inking abou t sch oo l leadership. It is not a
blueprint for effective leadership nor a prescription for how scho ol leadership sh ou ld be
practiced .

Th e lack of em p irica l evi dence on the effectiveness of d istributed leadership in p ro-

m oting instru ctiona l im proveme n t and increasing stu den t ach ievemen t is cons idered a
wea kness. While thi s concern is underst andabl e, it is n ot crucial. What matters for in-
structio na l im provemen t and stu den t ach ieveme n t is 110t th at leadership is distributed,
but how it is d istributed. Descriptive th eory buildin g is essential before ca usal links be-
tw een d istributed leadersh ip, ins tructional im proveme n t, and student ou tcomes can be
es tablishe d .

The Educational Forum • Volume 69 • Winter 2005 • 149

Spillane

From a di stributed per spective, leadership is a sy stem of practice comprised of a

collection of interacting components: leaders, followers, and situation . These interacting
components must be understood together because the system is more than th e sum of
the component parts or practices.

R ef erences
Barn ar d , C. I. 1938. The f unctions of tile execut ive. Ca m brid ge , MA: Har vard Univ er sity Press.
Bossert, S. T., D. C. Dw yer, B. Rowan, and G. V. Lee . 1982 . The instructional ma nagem en t role of th e p rincip al. Educational
A dmi nistra tion Quart erly 18(3): 34-64.
Cam burn, E. M., B. Row an , and J. Taylor. 2003. Distribu ted lead ership in schools: The cas e of elementa ry schools ad optin g
compreh en siv e sc hoo l reform mod els. Edu cati onal Evaluation and Policy Analys is 25(4): 347-73 .
Feld man, M. S., an d B. T. Pentland. 2003. Reconceptualizing organiza tional routines as a so u rce of flexibility and cha ng e.
A dministrative Science Quarterly 48(1): 94-11 8.
Fiedler, E E. 1973. Th e con tingen cy model: A reply to Asho ur. Organi zati onal Behavior and Human Decision Processes 9(3): 356-
68.
G ronn, P. 2002. Dis tri b u ted lead ership as a unit of an aly sis. Leadership Q uarterly 13(4): 423-51.
Ha llin ger, P., an d R. H . Heck. 1996. Reassessin g th e p rinci pal's role in school effectiveness. Educational Administ ration Quar -
terly 32(1): 5-44.
Heller, M. E , and W. A. Fireston e. 1995. Who's in charge her e? Sources of lead er ship for change in eigh t schools. Elemen tary
School Journal 96(1): 65-86.
Ma lone, T. W., G. Her m an, M. Klein , E. O'Do nnell, K. Crowsto n, J. Lee, B. Pentl and , C. Dellaroca s, G. Wyn er, J. Q u imby, C.
S. Osborn , and A. Bernstein . 1999 . Tools for in ven tin g organization s: Toward a han d book o f o rga ni zation a l processes.
Management Science 45(3): 425-43.
Murp hy, J. 1991. Restructuring schools: Capturing and assessing the phenomena. New York: Teachers Co llege Press .
School of Educati on and Social Polic y. 2004. The Dist ributed Leadership Study. Chic ago : No rthwes te rn Unive rsit y. Avai lab le
at: ht tp://dl s.sesp.north westem .edu .
Sp illane, J. P., J. B. Diamond , and L. [ita , 2003. Leading instruction : Th e d istri buti on o f lead er sh ip for ins tru ct io n. [ our nal of
CurriClilum Studies 35(5): 533-43.
Spillane, J. P., R. Hal verson , and J. B. Diamond. 2001. Investi gating sch ool lead er sh ip pr act ice: A d ist ributed perspect ive.
Edu cational Researcher 30(3): 23-28.
Spillane, J. P., R. Hal verson , and J. B. Dia mo nd. 2004. Towa rds a th eor y of lead ersh ip practice: A d ist ributed per sp ect ive.
[ou rnal oj Curriculum St udies 36(1): 3-34.
Sp illa ne , J. P., J. B. Diam ond, J. Sherer, an d A. Co ld ren . in p ress . Distributing leader ship. In Developing leadership: Creating the
schools of tomorrow, ed . M. Co les an d G. Sou thworth . Milt on Key es: Ope n Uni versity Press.
Tho m pson, J. D. 1967. Organi zations in action: Social science bases of adm inistrative theory. New York: McG raw Hill.

James P. Spillane is Professor of Education and Social Policy and a Faculty Fellow at the
Institutef or Policy Research, Northwestern University, where he teaches in the Learning
Sciences and Human Development and Social Policy graduate programs. He is author of
Standards Deviation (2004), Distrib u ted Leadership (in press), and numerous journal
articles and bookchapters. Spillane is Principal Investigator of the Distributed Leadership
Study, www.distributedleadership.org.

150 • The Educational Forum • Yolume 69 • Winter 2005