Upload
tranhanh
View
215
Download
3
Embed Size (px)
Citation preview
TRIGGERS AND ENABLERS OF SENSEGIVING INORGANIZATIONS
SALLY MAITLISUniversity of British Columbia
THOMAS B. LAWRENCESimon Fraser University
Drawing on a longitudinal study of sensegiving in organizations, we investigate theconditions associated with sensegiving by stakeholders and by leaders. For each group,we identify conditions that trigger sensegiving and conditions that enable it. Integrat-ing these analyses across organizational actors, we show that, generally: (1) theperception or anticipation of a gap in organizational sensemaking processes triggerssensemaking and (2) both discursive ability, which allows leaders and stakeholders toconstruct and articulate persuasive accounts, and process facilitators—routines, prac-tices, and structures that give organizational actors time and opportunity to engage insensegiving—enable sensegiving.
Organizational life is full of attempts to affecthow others perceive and understand the world.Gioia and Chittipeddi coined the term “sensegiv-ing” to describe this “process of attempting to in-fluence the sensemaking and meaning constructionof others toward a preferred redefinition of organi-zational reality” (1991: 442). Sensegiving is an in-terpretive process (Bartunek, Krim, Necochea, &Humphries, 1999; Gioia & Chittipeddi, 1991) inwhich actors influence each other through persua-sive or evocative language (Dunford & Jones, 2000;Snell, 2002), and it is used both by organizationalleaders (Bartunek et al., 1999; Corley & Gioia, 2004;Gioia & Chittipeddi, 1991) and other stakeholders,including middle managers (Balogun, 2003; West-ley, 1990), directors (McNulty & Pettigrew, 1999),and other employees (Maitlis, 2005).
Sensegiving is not only a prevalent activity inorganizations, but also a critically important one.Gioia and Chittipeddi’s (1991) study, for instance,reveals the significant role that leader sensegivingcan play in effecting major change, a finding ech-oed in several other studies (Bartunek et al., 1999;Corley & Gioia, 2004; Dunford & Jones, 2000; Gioia& Thomas, 1996). More broadly, research on thesymbolic aspects of leadership (Pfeffer, 1981;Pondy, 1978; Smircich & Morgan, 1982) has high-lighted sensegiving as a key leadership activity intimes of both change and stability. Sensegiving bystakeholders can also have profound consequences,
affecting strategic decision making, and stakehold-ers’ integration into or exclusion from a range ofimportant organizational processes (Balogun, 2003;Balogun & Johnson, 2004; Westley, 1990). Morebroadly, the pattern of leader and stakeholdersensegiving in combination has been shown toshape the processes and outcomes of organizational“sensemaking,” or the process of social construc-tion in which individuals attempt to interpret andexplain sets of cues from their environments (Mait-lis, 2005).
In general, previous studies of sensegiving havefocused on exploring the who and the what: whothe actors engaging in sensegiving are, and whatstrategies they are using to do so. We know little,however, about the conditions associated withsensegiving in organizations—where, when, orwhy it occurs—despite the fundamental nature ofthese issues. Research has shown, for instance, thatstakeholders and leaders do not always engage insensegiving, even around issues that matter to them(Dutton, Ashford, Lawrence, & Miner-Rubino,2002; Maitlis, 2005), but scholars lack a good un-derstanding of why this is so. Most studies of sense-giving have either ignored this question, or reliedon single case studies that have made it difficult toidentify a broad range of motivators for sensegiv-ing. Even less is known about the conditions thatmight facilitate sensegiving by those stakeholdersand leaders motivated to engage in it. These are theissues we begin to address in this paper: drawingon a longitudinal, qualitative study of sensemakingin 27 issue domains across three organizations, weinvestigate the triggers and enablers associated
We would like to thank Editor Sara Rynes and ourthree anonymous reviewers for their support and veryhelpful comments on drafts of this paper.
� Academy of Management Journal2007, Vol. 50, No. 1, 57–84.
Copyright of the Academy of Management, all rights reserved. Contents may not be copied,emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download or email articles for individual use only.
57
with sensegiving by organizational leaders andstakeholders.
This large-scale comparative study is unique insensegiving research and provides the foundationfor two sets of contributions. First, we add to un-derstanding of what motivates organization mem-bers’ sensegiving by identifying two conditions thattrigger sensegiving by stakeholders and two thattrigger sensegiving by leaders. We then integratethese analyses to show that, more generally, sense-giving is triggered by the perception or anticipationof a gap in organizational sensemaking processes.Second, this study contributes to an understandingof the conditions that facilitate sensegiving by mo-tivated actors. We identify three enablers for stake-holders and two for leaders and then integrate theseanalyses to show that sensegiving by both groups isenabled by (1) the possession of a discursive abilitythat allows them to fashion persuasive accountsand (2) the presence of process facilitators—organ-izational routines, practices, and structures thatprovide the time and opportunity to engage insensegiving.
RESEARCH ON SENSEGIVING INORGANIZATIONS
Gioia and Chittipeddi’s (1991) study hasprompted a number of studies of sensegiving in avariety of contexts. Although most of these do notexamine sensegiving as deeply as Gioia and Chitti-peddi (1991), they offer concrete illustrations of arange of sensegiving actors and activities and high-light the importance of sensegiving in organiza-tions. In this section, we review this research andits contributions to our understanding of sensegiv-ing, note an important knowledge gap, and developthe research questions that guided this article.
Research on leader sensegiving has tended tofocus on identifying the sensegiving strategies thatleaders use to effect organizational change. In Gioiaand Chittipeddi’s (1991) study, for instance, theprimary sensegiving agent was a new universitypresident who used a range of sensegiving tactics,including holding meetings to “espouse his vision”and “disclosing intentions through hypotheticalscenario presentations” in order to initiate and fa-cilitate change (Gioia & Chittipeddi, 1991: 442–443). In another study, Bartunek and her colleagues(1999) examined sensegiving within the creation ofan organizational development initiative, in whichleaders constructed opportunities “in ways that ap-pealed to the values of the receivers” (Bartunek etal., 1999: 38). Corley and Gioia (2004), in theirstudy of identity change following a corporate spin-off, found that leader sensegiving represented an
attempt to provide either new labels to describe thecompany or new meanings underlying these labels(Corley & Gioia, 2004: 196). Others have looked atleader sensegiving as storytelling. Dunford andJones (2000), for instance, highlighted the signifi-cance of senior managers’ sensegiving in narrativesabout events surrounding strategic change in threeorganizations. Snell (2002) also found that narra-tives played a key role in leader sensegiving in acompany’s attempt to become a “learning organiza-tion”; the president drove change with “stories ofprogress” and “TQM-related success stories” sup-porting the learning organization initiative.
A second set of studies has pointed to the impor-tance of sensegiving by organizational stakeholdersother than leaders. In a study of a U.K. utility com-pany, for instance, Balogun (2003) showed thatsensegiving was a key function of middle managersduring organizational change. Similarly, middle-and lower-level employees have been found to in-fluence a range of decisions through the construc-tion and propagation of issues in committees(Heller, 1998) and through “issue selling,” in whichsubordinates try to “influence the organizationalagenda by getting those above them to pay attentionto an issue” (Dutton et al., 2002: 355). Sensegivinghas also been shown to be an important activity ofboard directors, who shape both the content ofcompany strategy and the processes through whichit evolves, through such sensegiving activities astesting ideas, raising issues, and questioning as-sumptions (McNulty & Pettigrew, 1999). Looking atsensegiving by leaders and by stakeholders in com-bination, Maitlis (2005) showed how the pattern ofleader and stakeholder sensegiving affects thekinds of accounts and actions produced throughorganizational sensemaking processes.
Together, these studies provide a rich descrip-tion of organizational sensegiving. We understandthat sensegiving is an important activity both ofleaders and of other organizational stakeholders,with significant impacts on organizational change,strategy, and sensemaking (Bartunek et al., 1999;Corley & Gioia, 2004; Gioia & Chittipeddi, 1991;Maitlis, 2005).Through evocative language and theconstruction of narrative, symbols, and other sense-giving devices, leaders help shape the sensemakingprocesses of organization members toward someintended definition of reality. Stakeholders alsoplay a significant role in organizational sensegiv-ing, through activities such as issue selling, ques-tioning, and the propagation of ideas in consulta-tive committees. In addition, these studieshighlight the importance of sensegiving: in leaders,it can be a crucial element in facilitating accep-tance, enthusiasm, and energy for change, and in
58 FebruaryAcademy of Management Journal
stakeholders, it can determine the issues to whichleaders and others attend and can help shape thekinds of accounts and actions that are generated. Ineffect, researchers have a good understanding ofwhat sensegiving looks like (the sets of activities byleaders and stakeholders that constitute sensegiv-ing) and its impacts on organizations and theirmembers.
What is missing from all of these studies is asystematic examination of the conditions associ-ated with organizational stakeholder and leadersensegiving. Although previous work has shownthat leaders and stakeholders do not always engagein sensegiving around issues, even when it mightbenefit them or their organizations (Dutton et al.,2002; Maitlis, 2005), there is relatively little under-standing of why this is so. Two issues in this regardconcern the triggers of sensegiving (the conditionsthat motivate organizational actors to attempt toengage in sensegiving) and the enablers of sense-giving (the conditions that facilitate sensegiving onthe part of motivated actors).
What understanding there is of sensegiving trig-gers comes primarily from the contexts in whichsensegiving research has been done. Most studiesof leader sensegiving have occurred in the contextof major organizational change and have shownleaders motivated to shape the sensemaking of or-ganization members in support of an intendedchange (Bartunek et al., 1999; Dunford & Jones,2000; Gioia & Chittipeddi, 1991; Snell, 2002). Al-though this work suggests the importance of organ-izational change as a trigger for leader sensegiving,these studies report only one or a few case studies.They cannot therefore establish this relationshipwith any certainty, examine whether change mightbe part of a larger class of triggers, or reveal otherpossible triggers of leader sensegiving. Similarly,although studies of stakeholder sensegiving suggestthat it may occur when stakeholders have an issuethey wish to sell or an interpretation of events theywant to legitimize (Dutton et al., 2002; McNulty &Pettigrew, 1999), previous research designs leavelargely unaddressed the triggers of stakeholdersensegiving.
Research on the enablers of sensegiving by stake-holders and leaders is even less developed thanthat on its triggers. There has been very little sys-tematic examination of the conditions that mightfacilitate sensegiving, despite its being a criticallyimportant organizational ability. Research onleader sensegiving has focused on the managementof language and symbols (Dunford & Jones, 2000;Gioia & Chittipeddi, 1991; Snell, 2000) and so hasimplicitly suggested the importance of leaders’ lin-guistic and interpretive skills. Descriptions of
stakeholder sensegiving highlight the role of arange of factors, including political skill and organ-izational structure (Heller, 1998), stakeholder un-derstandings of organizational elites (Dutton et al.,2002), and formal authority (McNulty & Pettigrew,1999). None of these studies, however, has system-atically examined the enablers of sensegiving overa range of issues and actors.
Thus, two research questions guided this article:(1) What conditions trigger organizational stake-holders and leaders to engage in sensegiving activ-ities? and (2) What conditions enable sensegivingon the part of stakeholders and leaders motivated toengage in sensegiving activities?
METHODS
These research questions make considerable de-mands on research design and methods. The studyof sensegiving involves observing and interpretingorganization members’ constructions and accounts,which suggests the use of intensive qualitativemethods (Gioia & Thomas, 1996; Isabella, 1990). Atthe same time, however, the identification of con-ditions associated with sensegiving implies ananalysis in which comparisons can be made amongsensegiving situations and ideally among compara-ble organizations. The study on which this article isbased met these criteria by following organizationalsensegiving processes in real time, as they unfoldedover two years in nine different issue domains foreach of three matched organizations. This designprovided a strong foundation for investigating con-ditions associated with sensegiving in organiza-tions: the similarity of the three organizations al-lowed for meaningful comparisons of the processesand actors involved, and the diversity of issue do-mains and differences between organizations pro-vided a basis for generalizability.
Research Context
We carried out this study in three British sym-phony orchestras, a context well suited to sense-giving research (see Maitlis [2005] for a more de-tailed discussion of the context). With numerous,powerful, and diverse stakeholders, orchestras of-fer a context wherein sensegiving behaviors areimportant and visible (Allmendinger & Hackman,1996). Moreover, the products and processes of or-chestras are highly subjective, creating an environ-ment in which members must influence the inter-pretations of others to manage successfully (Hirsch,1972).
Despite some distinctive characteristics, orches-tras provide a high level of generalizability to a
2007 59Maitlis and Lawrence
broader population of organizations. Like any busi-ness, an orchestra has products (concerts and re-cordings) that must be marketed and sold, and ithas a wide range of customers to satisfy. The struc-ture of an orchestra is also similar to that of manymidsized firms, with an administrative team thattypically includes finance, marketing, operations,and human resource (HR) functions, and headed bya CEO who reports to a board. Important stakehold-ers include government, customers, employees,unions, and local communities. Because of thesecomparable goals, structures, and stakeholderroles, we would expect to find sensegiving dynam-ics that are similar to those in other types of organ-izations of similar size.
We based our choice of specific orchestras on thestudy’s aims, seeking a balanced sample of organi-zations in which issues and processes could becompared but in which heterogeneity was also sig-nificant. For the sample, we took three orchestras,which are referred to hereafter as the Provincial,the Broadcasting, and the London, one from each ofthe three categories of British symphony orchestras(these categories are differentiated by their gover-nance structures and revenue sources). To increasethe study’s generalizability, we chose orchestrasthat were “midlevel performers”—that is, neitherthe strongest nor the weakest (artistically or finan-cially) performers in their categories—and thusavoided patterns of sensegiving that might only beassociated with organizations with unusually highor low performance.
Data Collection
Our data collection was intensive, extending overmore than two years and involving interviews; obser-vation of meetings, rehearsals, and orchestra tours;and extensive documentary analysis. Data collectionwas aimed at understanding the sensemaking pro-cesses associated with issues in each of the orches-tras; a particular focus was on gaining an understand-ing of who was engaged in sensegiving arounddifferent organizational issues, the ways in whichthey did so, and the reasons for their engagement ornonengagement in sensegiving around that issue.
In each orchestra, the first author conducted in-terviews with the executive director and everymember of the management team; the primary non-musician “overseer,” the principal conductor; andall musicians who served as representatives on theboard and the orchestra committee (the musicians’elected representative body). In addition, she inter-viewed those who had been in the orchestra a longtime or had previously been on the board or orches-tra committee, as well as any other musician who
expressed an interest. At the beginning of thestudy, the first author explained to all parties thatshe was conducting a study of decision making inBritish symphony orchestras.
In total, she conducted 120 interviews, of which106 were recorded and transcribed. In the remain-ing 14 formal interviews, she took detailed notesthroughout. The interviews were semistructuredand became increasingly focused over the course ofthe study. Each early interview focused on the in-terviewee’s organizational role, the main issues inwhich he or she had been involved, and the pro-cesses through which those issues had been dis-cussed and/or resolved. For the issues in whichthey had been involved, the first author asked in-terviewees what form this involvement had taken,why they had participated in the way that they had,who else was involved, what forms others’ partici-pation took, the points of conflict and cooperationassociated with the issues, on what occasions theissues were discussed, and what actions were takento understand, manage, and resolve the issues. Aswell as these formal interviews, many informal in-terviews with organization members and otherswere conducted as opportunities arose. Through-out the formal and informal interviews, respon-dents described both their own sensegiving andthat of others. The informal interviews offered re-laxed, spontaneous opportunities to investigate ac-tors’ sensegiving, but most of the interview datareported in this paper and all of the interview quo-tations come from formal interview transcripts.
The first author observed 107 meetings, includ-ing meetings of various groups within each organ-ization (e.g., the executive team, the board) andmeetings between orchestra leaders and externalstakeholders (e.g., funders, collaborating organiza-tions). In each meeting, she took verbatim notes, asfar as possible, of what was said and by whom. Themeeting observation provided many rich opportu-nities to witness leaders and stakeholders engagingin sensegiving on a range of topics and to observeconditions associated with these instances ofsensegiving. The first author also collected a widevariety of documents for analysis. For each orches-tra, these included reports, organizational appraisaldocuments, newsletters, and meeting minutes. Inaddition, she gathered all available sector-levelstrategy and policy documents that addressed cur-rent issues and changes in the British orchestrasector at large.
Data Analysis
We carried out the data analysis in three mainstages. First, we developed a set of narratives that
60 FebruaryAcademy of Management Journal
described the sensemaking processes associatedwith issues that arose in all three organizations.Second, we identified issue domains in whichstakeholders and leaders engaged in sensegiving. Inthe third stage, we focused on answering our re-search questions, identifying conditions that trig-gered stakeholder and leader sensegiving and con-ditions that enabled sensegiving on the part ofstakeholders and leaders motivated to engage insensegiving.
Stage 1: Creating narratives of the sensemak-ing processes. The output of the first stage of thedata analysis was a set of “narratives” composed ofordered, raw data (quotes from interviews, meet-ings, documents, and field notes), each chroniclingthe sensemaking associated with an issue domainin an orchestra. This stage began with the firstauthor listing every “organizational issue” thatarose in the three orchestras during the study pe-riod (150–200 issues in each). An organizationalissue was defined as a topic of discussion thatinvolved a question or concern connected in someway to the organization as whole, rather than to asmall subset of its members. Examples included“appointment of second concert master,” “improv-ing orchestra’s stage presence,” and “making sav-ings on orchestra van.” She then reduced this list toa comprehensive set of “typical” issues facing aBritish orchestra on which the analysis would fo-cus, using four criteria. First, the issues had to havearisen in all three organizations. Through data re-duction (Strauss & Corbin, 1998), it was possible tocreate categories broad enough that each capturedan issue that arose in all three orchestras. Principalconductor appointment was one such issue. Twoorchestras were considering whether to renew theirincumbent’s contract; the other faced making a newappointment. Although these issues (appointmentand renewal) are not exactly the same, they weresimilar enough for the purposes of this study to begrouped together as “principal conductor appoint-ment/contract renewal.” The second criterion wasthat, in combination, the domains should coverartistic, financial, and personnel areas and includeboth strategy and operations. Applied together, thefirst and second criteria ensured that we examinedsensegiving within a generalizable set of issue do-mains. The third criterion was that data for theselected domains should be available from multi-ple sources and largely gathered as the issue un-folded, rather than retrospectively. We used thiscriterion to maximize the validity of the data, en-suring triangulation through multiple sources andreal-time data collection to avoid biases associatedwith retrospective accounts of action (Golden,1992). The fourth criterion was that organization
members should consider all issues to be of realsignificance. This ensured that the issue alonewould not determine leaders’ or stakeholders’ ten-dency to engage in sensegiving: it reduced the like-lihood of leaders or stakeholders not engaging insensegiving simply because they regarded an issueas transient or trivial. After developing a draft set ofissues, the first author refined it on the basis ofdiscussions with key respondents in each orches-tra. All respondents confirmed that the final setcovered the major issues that had arisen during thestudy period and that it captured the main catego-ries of issues normally faced by their organization.The nine issue domains were as follows: (1) pro-gramming of repertoire, (2) principal conductor ap-pointment/contract renewal, (3) reengagement ofguest conductors, (4) dealing with unsatisfactoryplayer performance, (5) appointment of key players(e.g., section principals), (6) players’ pay and con-tract review, (7) identifying areas for cost cutting,(8) increasing income generation, and (9) collabo-rative ventures.
Following Eisenhardt and Bourgeois (1988), thefirst author then developed a narrative of 5–25 sin-gle-spaced pages that chronicled the sensemakingactivities for each issue domain in each orchestra(see the Appendix for an excerpt from one of thenarratives). She constructed these 27 narratives bytracing through all the chronologically ordered rawdata for each orchestra to identify every time agiven issue arose and then describing the organiza-tional sensemaking process through quotationsfrom interviews, meeting transcripts, archival doc-umentation, and field notes. Thus, these narrativesare composed primarily of ordered raw data, withsome additional commentary for clarity, which to-gether tell the story of each issue as it unfolded.The first author used a broad definition of sense-making at this stage to identify sensemaking pro-cesses, so that the narratives reflected all of theattempts by organizational actors to understand,interpret, or name an issue, or to affect the ways inwhich others did any of these.
Although the conventional criteria of reliabilityand validity are not easily applied to naturalisticresearch, it is still important to show why the find-ings of a qualitative study are representative of thephenomenon of interest (Lee, 1999). The longitudi-nal, ethnographic nature of this study, the numer-ous sources of data, and the diverse methods ofdata collection meant that the first author con-structed the narratives not through some idiosyn-cratic impression she gained, but through a rigor-ous procedure that made full use of the wealth andcomplexity of data.
2007 61Maitlis and Lawrence
Stage 2: Identifying issue domains in which stake-holders and leaders engaged in sensegiving. Thisstage of the data analysis had two parts. The firstpart involved identifying stakeholders and leaders.We separated stakeholders and leaders in our anal-ysis because we believed that the conditions asso-ciated with sensegiving on the part of each groupwere likely to be distinct. Previous sensegiving re-search suggests marked differences between leaderand stakeholder responsibilities for sensegiving(leaders are expected to articulate their visions andpromote directions in many organizational do-mains) and resources (Balogun, 2003; Bartunek etal., 1999; Dunford & Jones, 2000; Dutton et al.,2002; Gioia & Chittipeddi, 1991). Moreover, re-search from the stakeholder theory tradition sug-gests that leaders and stakeholders play distinctroles in organizations, attending to different sets ofissues and basing responses on different sets ofresources (Agle & Mitchell, 1999).
We identified relevant stakeholders inductivelyafter the data collection, rather than assuming a setof stakeholders in advance or identifying themearly on in the fieldwork. This was to avoid eitherfocusing upon stakeholders subsequently found tobe insignificant or excluding and therefore missingthe sensegiving of actors who might at first haveappeared peripheral. By combining certain groupsand individuals (e.g., pooling all of an orchestra’sadministrators into one “administration” category),we were able to identify sets of stakeholders thatwere comparable across organizations. After refin-ing the set through feedback from key informants,we identified ten stakeholder groups common to allthree organizations: orchestra administration,player representatives, section principals, playersas a group, principal conductor, overseers, custom-ers, funders, guest artists, and the musicians’union. For each organization, we identified as lead-ers the two people with the greatest formal author-ity. The titles and roles of these individuals variedsomewhat by orchestra, reflecting differences in theorchestras’ organizational structures: in the Provin-cial and Broadcasting Orchestras, these individualswere the executive director and the next most se-nior administrator; in the London Orchestra, theleaders were the executive director and the chair-man of the board. We chose two leaders from eachorchestra because we observed in each organizationa significant difference in organizational roles be-tween these two individuals and the rest of theorchestra members. For each orchestra, the twoleaders were both involved in and had authorityover a far greater number of issues than any othermember.
The second part of Stage 2 involved identifying
sensegiving by stakeholders and by leaders. In or-der to do this, we reviewed the sensemaking liter-ature to identify typical sensegiving activities (e.g.,Bartunek et al., 1999; Dunford & Jones, 2000; Gioia& Chittipeddi, 1991) and analyzed a subset of ournarratives to develop a list of concrete sensegivingactivities in the orchestras. Although sensegiving isa form of influence, many forms, such as exchangeand coercion, often do not involve attempts toshape the sensemaking of others or shift their def-initions of reality (Lawrence, Mauws, Dyck, & Kley-sen, 2005). Moreover, individual activities do notin themselves constitute sensegiving, since a caseof sensegiving typically involves multiple activitiescarried out by an actor to shape the sensemaking ofothers (Gioia & Chittipeddi, 1991). We used this listnot to determine what constituted sensegiving, butrather to sensitize us to potential types of activitythat might be associated with sensegiving. Some ofthese activities were more obviously associatedwith the shaping of meaning construction, such as“justifying a view” or “promoting a position,” andothers were more apparently routine acts: we re-garded calling a meeting, for example, as a sense-giving activity when the meeting was called inorder to influence others’ sensemaking (as evi-denced by, for instance, the agenda or openingstatements). In all cases, however, we only consid-ered activities as elements of sensegiving whenthey were parts of attempts by actors to influencethe sensemaking of others toward a preferred redef-inition of organizational reality. Some sensegivingactivities were unique to leaders, such as present-ing an executive director’s report to an orchestra’sboard, but most were common to leaders andstakeholders.
For each of the 27 issue domains, we categorizedleaders and each stakeholder group as sensegivingor not sensegiving. Our aim was to differentiate thedomains in which leaders or stakeholders were at-tempting to “influence the sensemaking and mean-ing construction of others toward a preferred redef-inition of organizational reality” (Gioia &Chittipeddi, 1991: 442) from those in which suchattempts did not occur. For each issue, there werealways some stakeholders who engaged in sense-giving and others who did not. In contrast, leaderswere found to engage in at least some minimalsensegiving activities in every issue domain; thiswas not surprising, since each issue concerned im-portant organization-wide matters and was trackedover an extended period of time. We did not be-lieve, however, that categorizing all of these in-stances as “sensegiving” was either consistent withthe notion of leader sensegiving as described byGioia and Chittipeddi (1991) or useful for under-
62 FebruaryAcademy of Management Journal
standing conditions associated with leader sense-giving. We therefore considered a leader to be en-gaged in sensegiving only when we could identifyhis1 “preferred redefinition of reality” and a clearset of behaviors through which the leader was at-tempting to move organization members and stake-holders toward this redefinition.
For example, in analyzing the Provincial’s guestconductor reengagement, we first identified whichstakeholders were and were not involved in sense-giving. The Appendix, taken from the narrative ofthis issue domain, shows sensegiving by threegroups: the player representatives (orchestra com-mittee); the musicians as a group (full orchestra);and the orchestra’s administration (senior manage-ment). Analysis of the full narrative also showedthat the principal conductor, guest artists, custom-ers, funders, and overseers at various times at-tempted to influence the sensemaking of otherswith respect to the orchestra’s guest conductor is-sue. Each of these groups was classified as “en-gaged in sensegiving.” Two stakeholder groups—the section principals and the musicians’union—were found not to undertake any sensegiv-ing activities in this domain and so were classifiedas “not engaged in sensegiving.” We then analyzedthe narrative to assess leader sensegiving in thisdomain. This procedure revealed that the Provin-cial’s leaders did not articulate a preferred defini-tion of reality in relation to this issue and did notengage in more than a minimal number of activitiesaimed at influencing the sensemaking of others. Wetherefore classified leaders in this issue domain as“not sensegiving.”
In contrast, Broadcasting leaders were identifiedas sensegiving in the area of cost cutting, giventheir articulation of a preferred redefinition of re-ality and their engagement in a variety of behaviorsaimed at moving others toward it. The BroadcastingCEO was clear in his vision for a new system for a“player release” (musicians’ paid time off), explain-ing that he sought a system that cost less and had aless negative impact on artistic performance. Hefirst discussed the financial implications of changewith the business manager and then brought theissue up in an orchestra committee meeting, ex-pressing his concerns about the current system. In asubsequent meeting, he informed the committee ofhis plans for “streamlining” the system, explainingthat they were going to undertake a study of otherorchestras’ release systems, on which they wouldbase their revised system. Some months later, theCEO presented the proposal for a new system to the
players’ chairman, to be communicated to the restof the orchestra.
Stage 3: Identifying conditions associated withstakeholder and leader sensegiving. The finalstage of the data analysis focused on identifyingconditions associated with leader and stakeholdersensegiving. We should first note the relationshipbetween sensegiving and sensemaking. Althoughsensegiving concerns influencing the sensemakingof others, we often observed that a person’s sense-giving also seemed to affect his or her own sense-making around an issue. In fact, this seems nearlyinevitable, since the ways in which people speakwill always inform their understandings of theirown beliefs and positions (Weick, 1995). The re-verse, however, is not the case: not all instances ofsensemaking involve sensegiving. Rather, much ofthe time, people try to work out the meaning of anevent or action without attempting to shift others’definition of reality (Weick, 1995). Weick (1995)argued that sensemaking is associated with surpris-ing or confusing situations. We argue that sense-giving is also associated with a further set of con-ditions that differentiate it from sensemaking.Thus, we focused our analysis on conditions asso-ciated with sensegiving that extend beyond thosethat Weick (1995) associated with sensemaking.This stage of the data analysis involved three steps.
First, for each of the 27 issues, we identifiedfirst-order concepts (Van Maanen, 1979): situation-ally specific factors directly connected to stake-holder and leader sensegiving that could be de-scribed with a simple phrase such as “poororganizational decision process” or “poor out-comes of leader decision making” (both first-orderconcepts that aggregated to the second-order themeof “perceptions of a lack of leader competence”).2
This step involved two rounds of analysis via theconstant comparison method (Glaser & Strauss,1967), the first focused on stakeholders and thesecond on leaders. To identify first-order conceptsdescribing conditions associated with stakeholdersensegiving, we began by examining each issue do-main (e.g., programming in the Provincial) to seewhich stakeholder groups were and were not en-gaged in sensegiving and to ascertain the condi-tions that distinguished these groups from one an-
1 All the leaders were male.
2 Unlike first-order concepts in some qualitative stud-ies (e.g., Corley & Gioia, 2004; Van Maanen, 1979), oursdo not necessarily incorporate the language of organiza-tional actors. Rather, the names of our first-order con-cepts are intended to provide simple, descriptive labelsfor commonly described conditions in our data, and sorepresent an initial aggregation of the data.
2007 63Maitlis and Lawrence
other. This was an inductive and recursive processin which we initially identified, for each issue,conditions that seemed either to trigger or to enablesensegiving by stakeholder groups engaged insensegiving and compared these conditions tothose of the stakeholder groups that did not engagein sensegiving. When we found that an identifiedcondition did not differentiate sensegiving fromnonsensegiving stakeholders for an issue, we eitherdiscarded that condition or revised it to more pre-cisely describe the condition of only the sensegiv-ing stakeholder groups. In the latter case, we cycledthrough the comparative process again to ensurethat the revised conditions discriminated betweenstakeholder groups that did and did not engage insensegiving around that issue. For example, onecondition that emerged in our early coding as seem-ing to enable stakeholder sensegiving was the pres-ence of a large number of individuals constituting astakeholder group; as we tested this condition withrespect to different stakeholder groups in differentsensemaking domains, we found that it did notconsistently differentiate those stakeholder groupsthat engaged in sensegiving from those that did not.As a result, we discarded it from the list of condi-tions associated with sensegiving. We then exam-ined each stakeholder group to identify the condi-tions that distinguished those issue domains inwhich it engaged in sensegiving from those inwhich it did not. We followed a comparative pro-cess similar to that described above, generating ini-tial first-order concepts that described conditions,comparing issue domains to check how well theseconditions discriminated between issue domainsin which the stakeholder group engaged in sense-giving and those in which it did not, discarding orrevising nondiscriminating conditions, and thencycling through the comparative process until wehad identified a set of first-order conditions for thatstakeholder group. Each of these processes led tothe development of a set of first-order conditionsthat described bases for stakeholder sensegivingand included characteristics of stakeholders, organ-ization, and issue.
To generate the list of first-order conditions as-sociated with leader sensegiving, we engaged in asimilar process of constant comparison. First, weexamined the issue domains in which leaders en-gaged in sensegiving and identified commonaltiesin issue, organization, or leaders in relation to dif-ferent issues. We then compared first-order condi-tions present in the issue domains in which leadersengaged in sensegiving with those in the issue do-mains in which leaders did not. We retained theconditions that were evident only in issue domainsin which leaders engaged in sensegiving and dis-
carded or revised them where this was not the case.We then cycled through the comparative processagain with revised first-order conditions until wehad identified a set that only occurred in the issuedomains in which leaders engaged in sensegiving.We repeated this exercise looking for commonali-ties in issues, organizations, or leaders when lead-ers did not engage in sensegiving. Through thecombination of these two sets of constant compar-ative processes, we thus produced two lists of first-order conditions, one associated with stakeholdersensegiving and the other with leader sensegiving.
In the second step of this stage, we engaged inaxial coding (Strauss & Corbin, 1998) to build upmore abstract descriptions of conditions that ap-plied over multiple issues, combining first-orderconcepts to construct a set of second-order themes(Corley & Gioia, 2004). These themes representedmore abstract and robust descriptions of the condi-tions associated with stakeholder and leader sense-giving; for example, “effect of the issue on theorganization’s performance or viability” was com-bined with “effect on stakeholders’ everyday expe-rience of work” and “effect on stakeholders’ com-pensation or security” to create the second-ordertheme “stakeholder perceptions of issue impor-tance.” This step yielded a set of second-orderthemes that described the factors that were consis-tently associated with stakeholder sensegiving andleader sensegiving over the issue domains. Figure 1shows the data structure that describes the final setof second-order themes and the first-order conceptsfrom which they were derived.
The final step in this stage of the analysis focusedon verifying the conditions identified. To do this,we repeatedly examined the relationships we wereproposing in tandem with the data from which theywere derived and checked for completeness in thetrail of evidence. In keeping with an eliminativeinduction approach to theory generation (Miles &Huberman, 1994), we also sought exceptions to therules and alternative explanations.
FINDINGS
Our research questions ask what triggers and en-ablers are associated with sensegiving by stake-holders and leaders. Our analysis of sensegiving inthe 27 cases revealed a distinct set of conditionsassociated with stakeholder sensegiving and a dis-tinct set of conditions associated with leader sense-giving. Although each condition in a given set didnot have to be present for sensegiving to occur, theexistence of each made its occurrence incremen-tally likely. In the remainder of this section, we firstreport the conditions that were associated with
64 FebruaryAcademy of Management Journal
stakeholder sensegiving and then those associatedwith leader sensegiving. We explore why theseconditions might be associated with sensegiving inthe following section, when we develop our theo-retical model of sensegiving in organizations.
Conditions Associated with StakeholderSensegiving
Stakeholder sensegiving triggers. Two sets ofconditions seemed to trigger sensegiving by stake-holders. First, we found that sensegiving was asso-ciated with stakeholders’ perceptions that an issuewas important—to themselves, to a stakeholdergroup whom they represented, or to their organiza-tion at large. “Perceptions of issue as important” isa second-order theme that aggregates several first-
order concepts: perceptions that the “issue affectsthe organization’s performance or viability,” thatthe “issue affects stakeholders’ every day experi-ence of work,” and that the “issue affects stakehold-ers’ compensation or security.” Table 1 providesrepresentative quotations from the study data thatillustrate each of these first-order concepts. Quota-tion 1.3, for example, illustrates the perception ofan issue affecting the organization’s performance:in his interview, the Provincial’s finance directordescribed his perception that the orchestra’s “mostimmediate problem” was its artistic policy. Quota-tion 1.8, from an interview with a London sectionprincipal, illustrates how player hiring was an is-sue perceived to affect stakeholders’ everyday ex-perience of work; the interviewee commented, “Ican’t think of any other job where you work in such
FIGURE 1Data Structure
2007 65Maitlis and Lawrence
TA
BL
E1
Dat
aS
up
por
tin
gth
eT
hem
e“P
erce
pti
ons
ofIs
sue
asIm
por
tan
t”a
Ass
ocia
ted
Fir
st-O
rder
Con
cep
tsR
epre
sen
tati
veQ
uot
atio
ns
Issu
e’s
effe
cton
the
orga
niz
atio
n’s
per
form
ance
orvi
abil
ity
1.1
“At
the
end
ofth
ed
ayit
’san
orch
estr
a,an
dth
ere
are
five
orch
estr
asin
Lon
don
,an
dif
we
don
’tp
lay
the
righ
tco
nce
rt,
then
they
’ll
goto
mor
row
nig
ht
and
hea
ran
oth
eron
e”(i
nte
rvie
w,
LS
Om
arke
tin
gm
anag
er)
(LS
O1)
.
1.2
On
the
nee
dfo
ra
“su
per
orch
estr
a”:“
At
that
tim
e[i
nth
e80
s],i
tfe
ltli
kea
rath
erth
reat
enin
gw
ayto
mak
era
ther
mar
gin
alch
ange
s.N
owit
feel
sli
keth
isis
som
eth
ing
that
real
lyd
oes
nee
dto
bead
dre
ssed
.”(i
nte
rvie
w,B
SO
over
seer
)(B
SO
9)
1.3
“Th
em
ost
imm
edia
tep
robl
emis
n’t
[th
ep
rin
cip
alco
nd
uct
or],
it’s
our
arti
stic
pol
icy,
asI
said
the
oth
erd
ay.
God
,I
was
gett
ing
fed
up
,I
real
lyw
as.
Ou
rar
tist
icp
olic
y—w
e’ve
got
toge
tth
atso
rted
.”(i
nte
rvie
w,
PS
Ofi
nan
ced
irec
tor)
1.4
“[T
he
pla
yers
’ch
airm
an]
exp
ress
edco
nce
rnon
beh
alf
ofth
eor
ches
tra
onth
ep
robl
emof
[th
ele
ader
]be
ing
ill
and
the
imp
lica
tion
sab
out
the
stab
ilit
yof
the
orch
estr
a.”
(min
ute
s,B
SO
man
agem
ent/
orch
estr
aco
mm
itte
em
eeti
ng)
(BS
O5)
1.5
“It
was
alw
ays
aw
orry
that
we
wer
ego
ing
tolo
seou
rin
tere
stin
g,m
ore
exot
icre
per
toir
e,w
hic
his
wh
atth
eor
ches
tra’
sre
pu
tati
onis
on,
Ith
ink.
An
dm
yte
rror
was
that
we
wer
ego
ing
toen
du
pli
keth
eR
PO
:T
sch
aik.
5,B
eeth
oven
5,an
dB
rah
ms
4ad
nau
seam
,si
mp
lybe
cau
seit
gets
the
bum
son
the
seat
s.”
(in
terv
iew
,B
SO
pla
yer)
(BS
O1)
Issu
e’s
effe
cton
stak
ehol
der
s’ev
eryd
ayex
per
ien
ceof
wor
k1.
6“I
tw
asfe
ltth
at[c
ond
uct
orC
]w
asa
very
du
llan
du
nin
spir
ing
con
du
ctor
and
the
orch
estr
aw
ould
not
like
tose
eh
imre
turn
.”(m
inu
tes,
PS
Oar
tist
icad
viso
rysu
bcom
mit
tee)
(PS
O3)
1.7
“[P
laye
rs]
wen
tfr
omtr
yin
gto
bep
osit
ive
abou
th
im.
Som
eof
them
wer
ep
osit
ive,
and
som
eof
them
trie
dto
bep
osit
ive.
An
dit
’sju
stgo
ne.
Eve
nth
ep
eop
lew
ho
like
dh
ima
year
ago
say
they
wan
tto
see
him
goin
g.T
hey
thin
kh
e’s
tota
lly
inco
nsi
der
ate,
ap
oor
mu
sic
dir
ecto
r.H
ed
oesn
’tu
nd
erst
and
the
[PS
O]
atal
l.”
(in
terv
iew
,P
SO
orch
estr
am
anag
er)
(PS
O2)
1.8
“Ica
n’t
thin
kof
any
oth
erjo
bw
her
eyo
uw
ork
insu
chcl
ose
pro
xim
ity.
We’
real
lh
erd
edto
geth
erli
kesh
eep
—yo
uh
ave
toge
ton
wit
hso
meb
ody.
It’s
both
pla
yin
gan
dh
owp
eop
leli
keth
ep
erso
n.”
(in
terv
iew
,L
SO
stri
ng
sect
ion
pri
nci
pal
)(L
SO
5)
1.9
“Ica
nn
otov
erem
ph
asiz
eth
ed
amag
ing
effe
ctth
e[a
bove
]p
oin
th
ash
adon
ever
yin
div
idu
alin
the
orch
estr
a:w
hen
we
see
ap
laye
rof
[th
ism
usi
cian
’s]
cali
ber
trea
ted
inth
isw
ay,
itis
clea
rth
atn
one
ofu
sca
nco
un
ton
the
sup
por
tof
our
man
agem
ent
agai
nst
the
wh
imof
any
pas
sin
gco
nd
uct
or.”
(let
ter
from
BS
Op
laye
rs’
chai
rman
toC
EO
)(B
SO
4)
Issu
e’s
effe
cton
stak
ehol
der
s’co
mp
ensa
tion
orse
curi
ty1.
10“B
ut
the
pro
blem
for
orch
estr
alp
laye
rsov
erth
eye
ars
isth
atth
ey’v
en
ever
been
pai
den
ough
,so
they
’re
alw
ays
grab
bin
gfo
rth
eex
tra
bits
and
wan
tin
gto
get
pai
dfo
rth
em,
soth
atat
the
end
ofth
ed
ay,
they
’ve
got
are
ason
able
amou
nt
ofin
com
e.A
nd
up
un
til
fair
lyre
cen
tly,
itst
ill
did
n’t
mea
na
reas
onab
lein
com
e.I
stil
ld
on’t
thin
kth
eyge
ten
ough
.”(i
nte
rvie
w,
BS
Oor
ches
tra
man
ager
).(B
SO
6)
1.11
“Th
ere
was
age
ner
ald
iscu
ssio
non
the
subj
ect
ofth
ead
dit
ion
alri
ghts
the
BB
Cis
aim
ing
toac
quir
e.T
he
com
mit
tee
exp
ress
edth
eco
nce
rnof
the
orch
estr
aab
out
the
un
real
isti
call
yla
rge
nu
mbe
rof
CD
sess
ion
sd
esir
edby
the
BB
C.”
(min
ute
s,B
SO
man
agem
ent/
orch
estr
aco
mm
itte
em
eeti
ng)
(BS
O6)
1.12
“Fu
rth
erd
iscu
ssio
ns
then
took
pla
ceab
out
the
pla
yers
’fe
elin
gsw
hen
[th
ep
rin
cip
alco
nd
uct
or]
aske
dto
see
ap
laye
rto
dis
cuss
thei
rp
layi
ng.
Th
isca
use
dp
laye
rsco
nsi
der
able
stre
ssan
dan
xiet
y.”
(min
ute
s,P
SO
con
sult
ativ
eco
mm
itte
em
eeti
ng)
(PS
O4)
.
aIn
Tab
les
1–9,
the
Pro
vin
cial
,Bro
adca
stin
g,an
dL
ond
onO
rch
estr
ash
ave
been
abbr
evia
ted
toP
SO
,BS
O,a
nd
LS
O.T
he
issu
ed
omai
nn
um
bers
(e.g
.,L
SO
1)id
enti
fybo
thth
eor
ches
tra
inw
hic
han
dth
eis
sue
tow
hic
ha
quot
ere
fers
.Th
eis
sue
nu
mbe
rsar
e1,
pro
gram
min
g;2,
pri
nci
pal
con
du
ctor
app
oin
tmen
t/re
new
al;3
,gu
estc
ond
uct
orre
enga
gem
ent;
4,u
nsa
tisf
acto
ryp
laye
rp
erfo
rman
ce;
5,ke
yp
laye
rap
poi
ntm
ents
;6,
pla
yer
pay
and
con
trac
ts;
7,co
stcu
ttin
g;8,
inco
me
gen
erat
ion
;9,
coll
abor
ativ
eve
ntu
res.
close proximity.” Quotation 1.11, from a meetingbetween the Broadcasting’s management and itsorchestra committee, conveys musicians’ concernabout the impact of a new employment contractthen in negotiation, illustrating the concept “issueaffects stakeholders’ compensation or security.”
The second condition that was associated withsensegiving by stakeholders was “perceptions of alack of leader competence” with respect to a given
issue. This is also a second-order theme that aggre-gates several first-order concepts: stakeholder per-ceptions of “poor organizational decision process,”“poor outcomes of leader decision making,” and“lack of leader expertise.” Table 2 provides quota-tions from the study data that illustrate each ofthese first-order concepts. Quotation 2.2, for in-stance, is from an interview with a Broadcastingorchestra committee member who commented on
TABLE 2Data Supporting the Theme “Perceptions of a Lack of Leader Competence”
Associated First-Order Concepts Representative Quotations
Poor organizational decision process 2.1 “[The associate leader] expressed concern over the lack of information from the officeand wondered whether enough was being done to seek out potential leaders to workwith the orchestra.” (minutes, BSO orchestra committee meeting) (BSO5)
2.2 Commenting on the appointment decision process for an orchestra leader: “It’s oneincredible grey area. Nobody seems to know what’s happening with that and no oneseems to know whose responsibility it is. . . . Eventually, the principals just made it soclear that basically they weren’t happy [that the appointment was not made]. . . . Butwe have a theory that he may have promised the guy the job first, and got himself intoa pickle.” (interview, BSO orchestra committee member) (BSO5)
2.3 Commenting on a decision not to terminate a player, a LSO player board membercommented: “There was a decision over this player. The vote was taken and it wentagainst the wishes of the chairman, and he said, ‘Well okay, we’ll call a council ofprincipals meeting’. . . . Most of the principals are more than happy to sit on the fence.They’ve got a hard enough job. They don’t want to put their oar in and stir things up,so of course the vote went the other way. Now I think that’s a misuse of power, if youlike. You’re widening the goal posts and moving them at the same time. I was morethan a little pissed off about that because it didn’t seem to be fair. What was the pointof having a [board]?” (interview, LSO player board member) (LSO5)
Poor outcomes of leader decisionmaking
2.4 “Programming is [the senior producer]—you couldn’t ask for better repertoire. [Thesenior producer] is very successful. He has organized some very good programmes andconcerts.” (interview, BSO player) (BSO1)
2.5 “Looking back on all this, I would say that those judgments [of the chief executive]were fatally flawed for our organization on two counts: [the principal conductor’s]availability and commitment, and his financial cost.” (interview, PSO player director)(PSO2)
2.6 “It was like lambs to the slaughter. The contract [the principal conductor] was offeredshould never have been accepted.” (interview, PSO deputy CEO) (PSO2)
2.7 “If you look at the main [home city] concerts, something has happened there, andwe’ve lost our thread, because we had three distinct series. . . . So I think the [PSO],represented by the board and the senior management, has a duty to make sure that therepertoire actually fulfils the artistic strategy.” (interview, PSO player chairman-elect)(PSO1)
Lack of leader expertise 2.8 [We need] someone who knows what they’re doing, who has sufficient commercialgrasp to know the effect of what they’re doing, and appreciates the need to create aprogramme for [the PSO home city] that will also apply in [other regional towns]. It’sthat thorough vision that is lacking at the moment, causing all sorts of orchestralproblems.” (interview, PSO finance director) (PSO1)
2.9 “You have someone here [the chief executive] who has no understanding oforchestras at all.” (observation, musicians’ union representative, PSO players meetingwith musicians’ union) (PSO6)
2007 67Maitlis and Lawrence
the confusion and inertia surrounding an importanthiring decision. Quotation 2.6 is from an interviewwith a Provincial senior administrator who be-lieved that the orchestra’s leadership had agreed toa very poor contract in previous negotiations withthe principal conductor. An example of a percep-tion of lack of leader expertise (quotation 2.8)comes from the Provincial’s finance director, whobelieved that the orchestra’s leaders lacked “suffi-cient commercial grasp” to know the effect of theirprogramming decisions.
Stakeholder sensegiving enablers. Our second setof findings in relation to stakeholder sensegiving con-cerned the three conditions that enabled it. The firstwas stakeholders’ possession of relevant expertise:stakeholders were more likely to engage in sensegiv-ing when they possessed expertise relating to an is-sue. Table 3 summarizes the types of expertise asso-ciated with the different issues and stakeholders ineach orchestra. The areas of expertise that we argueare associated with each issue are based on the long-term and intensive observation by the first author ofdecision making in each organization. We made anassessment of stakeholder expertise in each of theseareas based on the knowledge of each party’s profes-sional background and training (e.g., musician; in-vestment banker board member) and evidence of ex-pertise shown or described during fieldworkinterviews and observation. Expertise was differen-tially distributed across different stakeholders for dif-ferent issues. For example, although section princi-pals in each orchestra possessed the skill, training,and experience to judge the musical performance oftheir colleagues (expertise area 10) and of job candi-dates (expertise area 13), they did not have any par-ticular expertise in assessing the audience appeal oforchestral repertoire (expertise area 2) or knowledgeof other orchestras’ contractual agreements (expertisearea 17). In contrast, the marketing manager of eachorchestra possessed expertise in area 2, and the mu-sicians’ union representatives had expertise in area17. There were also differences in stakeholder exper-tise among the orchestras: for example, some of theBroadcasting overseers knew where organizationalslack lay in that orchestra (area 20), but overseers inthe Provincial and London did not possess this kindof knowledge about their orchestras’ operations.
The second condition that seemed to enablestakeholder sensegiving was legitimacy, which pro-vided stakeholders with an acceptable basis fromwhich to engage in sensegiving. This second-ordertheme aggregates the following first-order concepts:“formal authority/responsibility,” “representativerole,” and “principles of organizational involve-ment/participation.” Table 4 presents quotationsfrom the data that illustrate each of these first-order
concepts. In quotation 4.1, for instance, the Provin-cial’s players’ chairman described the formal au-thority of the principal conductor, who was given“hire and fire” powers, which gave him the legiti-macy to comment on the quality of players’ perfor-mance and argue for their dismissal. Quotation 4.6comes from the Provincial’s musicians’ union rep-resentative, whose legitimacy in the area of playerpay and contracts came from his role as the players’representative in this domain. Players often gainedtheir legitimacy from an orchestra’s principles ofparticipation or involvement: in quotation 4.9, aLondon player contrasts his orchestra with another,suggesting that the London players are allowedmuch greater involvement in decision makingaround such issues as principal conductorappointments.
The third condition that seemed to enable stake-holder sensegiving was the opportunity to engage insuch behavior. The second-order theme of “op-portunity” aggregates four first-order concepts:“regular meetings,” “ad hoc meetings,” “ad hocsolicitation of views,” and “ad hoc retreats”; to-gether, these represent the most common kinds ofsensegiving opportunities for stakeholders in theorchestras. Table 5 presents quotations from thedata that illustrate sensegiving in each of thesecontexts, and also, in the third column, the rela-tive frequency of each kind of opportunity ineach orchestra. Quotation 5.1 illustrates sense-giving by Provincial players in the context of aregular committee meeting, in this case the Pro-vincial’s Concerts Department subcommittee.Quotation 5.3 describes “orchestral meetings,” aless regular kind of sensegiving opportunity inthe London, in which all players had the chanceto express their views on key issues. Other op-portunities for sensegiving occurred through adhoc solicitation of stakeholder views. In quota-tion 5.8, for example, a Broadcasting player de-scribes how the senior producer (one of the or-chestra’s leaders) informally solicited playerperspectives during lunch and dinner breaks. Adhoc retreats, the last main category here, pro-vided a distinctive opportunity to engage insensegiving around strategic issues. Quotation5.10, taken from a retreat’s “Overview Docu-ment,” describes its purpose as “to brainstormwhere we believe the Orchestra should be in themedium term.”
Conditions Associated with Leader Sensegiving
Leader sensegiving triggers. Although scholarshave long argued that a central leadership functionis the construction of meaning (Pfeffer, 1981;
68 FebruaryAcademy of Management Journal
TABLE 3Types of Issue-Related Expertise Possessed by Stakeholdersa
Related Types ofExpertise Stakeholders Possessing These Types of Expertise
Programming1. Musical knowledge
of repertoirePSO: Some administrators (artistic advisor and deputy CEO); principal conductor; one board member; guestartists.BSO: Principal conductor; some overseers; guest artists.LSO: Principal conductor; guest artists.
2. Knowledge of whatrepertoire willattract audiences
PSO: Some administrators (marketing manager; deputy CEO); customers.BSO: Some administrators (marketing manager); customers.LSO: Some administrators (marketing manager); customers.
3. Understanding ofthe orchestra’sartistic strategy
PSO: Principal conductor; some funders.BSO: Principal conductor; funders.LSO: Some funders.
Principal conductorappointment
4. Knowledge ofpotentialconductors for theposition
PSO: None.BSO: None.LSO: None.
5. Musicalappreciation ofconductors’ styles,abilities, andpotential
PSO: Some administrators (deputy CEO); players; some funders.
BSO: Players; funders.
LSO: Players; some funders.
6. Sense of the fitbetween aconductor and theorchestra
PSO: One board member; players; some funders.BSO: Overseers; players; funders.LSO: Players; some funders.
7. Knowledge oftypical principalconductorcontractualarrangements
PSO: Some funders.
BSO: Overseers; funders.
LSO: Some funders.
Guest conductors8. Awareness of
potentialconductors for theposition
PSO: Some administrators (deputy CEO); players; principal conductor; one board member; guest artists.BSO: Players; principal conductor; some overseers; guest artists.LSO: Players; guest artists.
9. Musicalappreciation ofconductors’ styles,abilities, andpotential
PSO: Some administrators (deputy CEO); players; principal conductor, one board member; guest artists.
BSO: Players; principal conductor; some overseers; guest artists.
LSO: Players; guest artists.
Player performance10. Awareness of and
ability to judgemusicalperformance
PSO: Some administrators (orchestra manager); some section principals; some players; principal conductor;some customers; guest artists.BSO: Some administrators (orchestra manager); some section principals; some players; principal conductor;some customers; guest artists.LSO: Some section principals; some players; some customers; guest artists.
11. Knowledge ofreasons for poorperformance
PSO: Some administrators (orchestra manager); some section principals; some players; principal conductor.BSO: Some administrators (orchestra manager); section principals; some players; principal conductor.LSO: Some administrators (orchestra manager); section principals; some players.
12. Knowledge of theprocedures fordealing with playerperformancesissues
PSO: Some administrators (orchestra manager); section principals; player representatives; players; principalconductor.BSO: Some administrators (orchestra manager); section principals; player representatives; players; principalconductor.LSO: Some administrators (orchestra manager); section principals; player representatives; players.
Player hiring13. Ability to judge
musicalperformance of jobcandidates
PSO: Some administrators (orchestra manager); some section principals; some players; principal conductor;guest artists.BSO: Some administrators (orchestra manager); some section principals; some players; principal conductor;guest artists.LSO: Some section principals; some players; principal conductor; guest artists.
(Continued)
2007 69Maitlis and Lawrence
Pondy, 1978; Smircich & Morgan, 1982), we foundthat the leaders in our study were only triggered toengage in sensegiving under specific conditions.The first condition was their own “perceptions ofan issue as uncertain.” The second-order theme ofissue uncertainty aggregates two first-order con-cepts: ambiguous issues and unpredictable issues.
Table 6 provides representative quotations from thestudy data that illustrate these first-order concepts.Quotation 6.1, for instance, illustrates the first-order concept of ambiguous issues: here, the Pro-vincial’s CEO describes the ambiguity associatedwith assessing the strengths and weaknesses of aconductor, as demonstrated by the different per-
TABLE 3Continued
Related Types ofExpertise Stakeholders Possessing These Types of Expertise
14. Awareness ofavailablecandidates
PSO: Some section principals; some players.BSO: Some section principals; some players.LSO: Some section principals; some players.
Player pay and contracts15. Knowledge of
collective agree-ments, laborrelations
PSO: Some administrators (deputy CEO); some player representatives; musicians’ union.BSO: Some administrators (orchestra manager); some player representatives; some overseers; musicians’ union.LSO: Some player representatives; musicians’ union.
16. Negotiation skills PSO: Musicians’ union.BSO: Musicians’ union.LSO: Musicians’ union.
17. Knowledge of otherorchestras’contractualagreements
PSO: Some administrators (deputy CEO); some player representatives; musicians’ union.BSO: Some administrators (orchestra manager); some player representatives; musicians’ union.LSO: Some administrators (orchestra manager); some player representatives; musicians’ union.
Cost cutting18. Knowledge of organ-
ization’s operation-al costs
PSO: Some administrators (finance director, deputy CEO); overseers; some funders.BSO: Some administrators (business manager); overseers; some funders.LSO: Some administrators (deputy CEO); overseers; some funders.
19. Understanding ofmanagementaccounting
PSO: Some administrators (finance director); overseers.BSO: Some administrators (business manager); overseers.LSO: Some administrators (finance director); overseers.
20. Knowledge ofwhere organization-al slack lay
PSO: Some administrators (deputy CEO, finance director).BSO: Some administrators (business manager, orchestra manager); overseers.LSO: None.
Income generation21. Knowledge of
income-generatingopportunities
PSO: Some administrators (deputy CEO, development director, education director, marketing director);overseers; customers; some funders; some guest artists.BSO: Some administrators (education manager, marketing manager); overseers; customers; funders; someguest artists.LSO: Some administrators (development director, marketing manager, education officer); overseers;customers; funders; some guest artists.
Collaborative ventures22. Awareness of
opportunities forcollaboration
PSO: Some administrators (development director, marketing director); some overseers; some funders.BSO: Some overseers.LSO: None.
23. Awareness ofstrengths/weaknessesof potentialcollaborators
PSO: Some administrators (development director; marketing director); some funders.BSO: Some overseers.LSO: None.
24. Awareness ofenvironmentalfactors that (en/dis)couragecollaboration
PSO: Some administrators (development director, marketing director); overseers; some funders.
BSO: Some administrators (business manager); overseers; funders.
LSO: None.
a The types of expertise were defined through ethnographic observation by the first author. Judgments as to possession of expertise werebased on knowledge of each party’s professional background and training and evidence of expertise shown or described during fieldworkinterviews and observation.
70 FebruaryAcademy of Management Journal
ceptions of different groups (the board, the public),and his role in managing that ambiguity. Quotation6.4 illustrates the first-order concept of unpredict-able issues. In this case, a Broadcasting leader de-scribed the unpredictability of programming for theorchestra; opportunities often arose that createdsignificant changes to the orchestra’s plans.
The second condition that seemed to triggerleader sensegiving was an issue’s association withcomplex stakeholder interests. This second-ordertheme, “leaders’ perceptions of the stakeholder en-vironment as complex,” aggregated two first-orderconcepts: “numerous stakeholders connected to the
issue” and “divergent interests among connectedstakeholders.” When these were present, leaderswere more likely to engage in sensegiving in orderto construct interpretations that bridged stake-holder differences. Table 7 provides representativequotations from the study data to illustrate thetheme of stakeholder complexity. Quotation 7.1 il-lustrates the situation in which an issue is associ-ated with a large number of stakeholders: the Pro-vincial’s CEO describes programming in theorchestra as directly involving the board, the prin-cipal conductor, the part-time artistic advisor, thedeputy chief executive, and an external consultant.
TABLE 4Data Supporting the Theme “Stakeholders’ Issue-Related Legitimacy”
Associated First-Order Concepts Representative Quotations
Formal authority andresponsibility
4.1 The players’ chairman explained that the principal conductor had “hire and fire powers,”observing, “[the principal conductor] has already got rid of one person and has others he wantsto.” (interview, PSO players’ chairman) (PSO4)
4.2 “Management do have overall right to appoint who they want, the same with the leader, theydo have the right to do that, but again, that is hopefully done in conjunction with the sectionprincipals. Really I think the section principals would say it’s their responsibility.” (interview,BSO orchestra committee member) (BSO5)
4.3 Applications have been processed by a meeting of principals. Some candidates will be invitedto guest lead, others will be auditioned.” (minutes, BSO orchestra committee-managementmeeting) (BSO5)
4.4 One member described the orchestra committee as doing “the day-to-day running of theorchestra, mainly personnel things, discipline, for example, someone having to be moved.” Theseincluded “a lot of personnel decisions about people’s playing: little incidents, whether tosuspend someone, or give a verbal or written warning.” (interview, LSO board member) (LSO4)
4.5 A BSO overseer explained that the head of music would pull together a business plan,involving the senior producer and the business manager, at which point he [the overseer] wouldreview it: “I often will have a strong influence on it, if it looks as though our management orresource costs look rather high in relation to the rest of the costs or what income’s likely to be.(interview, BSO overseer) (BSO7)
Representative role 4.6 “I’m not supposed to be making a recommendation—I’m not making a recommendation—but Ican imagine there’s no way you can vote to accept this. . . . If you vote to accept, we won’t go ontalking and you have a pay freeze. If you don’t accept, we will go on talking and you have a payfreeze” (observation, musicians’ union representative, PSO players meeting with musicians’union) (PSO6)
4.7 The artistic advisory subcommittee is “[the players’] forum to say what they think is wrongwith some of our decisions. . . . So the orchestra feel they have a way of being heard, andultimately what they say is taken into consideration.” (interview, PSO part-time artistic advisor)(PSO1)
4.8 “The first meeting we had with them, there was a bit of a bump, because they [sectionprincipals] always push this hierarchy with me, and say, ‘we represent our sections,’ and I’m notconvinced of that. . . . I asked them to go back and consult their sections, and I actually checkedwith the rank-and-file players whether they’d been asked.” (interview, BSO head of music)(BSO4)
Organizationalprinciples ofinvolvement/participation
4.9 “If we want to get a conductor in, the management and [board] don’t just get the conductor in,they ask the orchestra. They don’t do that in the [other orchestra].” (interview, LSO player)(LSO2)
4.10 Now I talk to people individually—grab them at tea break and have a little chat. They don’thave to feel under any pressure and can say exactly what they want. They do say exactly whatthey want.” (interview, LSO string section principal) (LSO5)
2007 71Maitlis and Lawrence
TA
BL
E5
Dat
aS
up
por
tin
gT
hem
e“S
tak
ehol
der
s’O
pp
ortu
nit
ies
for
Sen
segi
vin
g”
Ass
ocia
ted
Fir
st-
Ord
erC
once
pts
Rep
rese
nta
tive
Qu
otes
Fre
quen
cyan
dT
ype
ofO
pp
ortu
nit
yby
Orc
hes
tra
Reg
ula
rm
eeti
ngs
5.1
“[A
pla
yer]
said
the
orch
estr
aw
asan
xiou
sto
know
that
an
ewle
ader
was
acce
pta
ble
to[t
he
pri
nci
pal
con
du
ctor
]as
this
wou
ldm
ean
ther
ew
ould
befe
wer
pro
blem
sin
futu
rew
ith
the
sch
edu
le.
...
Th
efe
elin
gin
the
orch
estr
ais
that
the
div
isio
nof
wor
kbe
twee
nth
etw
o[c
once
rtm
aste
rs]
idea
lly
shou
ldbe
50/5
0;ce
rtai
nly
no
mor
eth
an60
/40
.”(m
inu
tes,
PS
OC
once
rts
Dep
artm
ent
subc
omm
itte
em
eeti
ng)
(PS
O5)
PS
O:
Wee
kly
sen
ior
man
agem
ent
oper
atio
ns;
mon
thly
sen
ior
man
agem
ent
stra
tegy
;qu
arte
rly
boar
d;
quar
terl
yfi
nan
ce,
mar
keti
ng,
and
dev
elop
men
tbo
ard
subc
omm
itte
es;
quar
terl
yC
onsu
ltat
ive
Com
mit
tee;
quar
terl
yar
tist
ic,
con
cert
s,an
dm
arke
tin
gC
onsu
ltat
ive
Com
mit
tee
subc
omm
itte
es.
5.2
“Rep
rese
nta
tive
sfr
om[t
he
orch
estr
aco
mm
itte
e]at
ten
ded
the
quar
terl
yC
onsu
ltat
ive
Com
mit
tee
mee
tin
gs,
asw
ell
asth
ear
tist
icad
viso
rysu
bcom
mit
tee
mee
tin
gsth
atto
okp
lace
inbe
twee
nth
eC
onsu
ltat
ive
Com
mit
tee
mee
tin
gs.
Pla
yer
dir
ecto
rsan
dth
ep
laye
rs’
chai
rman
atte
nd
edth
ebi
mon
thly
boar
dm
eeti
ngs
and
also
the
boar
daw
ay-
day
,w
her
em
atte
rsof
pro
gram
min
gan
d,
mor
ew
idel
y,ar
tist
icst
rate
gy,
wer
ed
iscu
ssed
.(P
SO
,fi
eld
not
es)
(PS
O1)
BS
O:
mon
thly
adm
inis
trat
ive;
mon
thly
orch
estr
aco
mm
itte
e-m
anag
emen
t;qu
arte
rly
orch
estr
aco
mm
itte
e-se
nio
rm
anag
emen
t.
5.3
“Con
du
ctor
s”w
asfr
equ
entl
ym
enti
oned
asa
typ
ical
top
icof
boar
dd
ecis
ion
,an
dd
ebat
ear
oun
dd
iffe
ren
tin
div
idu
als
took
pla
ceat
som
eof
the
mee
tin
gsI
atte
nd
ed.
...
Th
em
inu
tes
from
abo
ard
mee
tin
gin
Oct
ober
sum
mar
ize
dis
cuss
ion
sth
ath
adta
ken
pla
ceab
out
thre
ed
iffe
ren
tin
div
idu
als.
(fie
ldn
otes
,L
SO
)(L
SO
3)
LS
O:
quar
terl
ybo
ard
;qu
arte
rly
man
agem
ent
subc
omm
itte
e.
Ad
hoc
mee
tin
gs5.
4“T
his
isa
cyn
ical
,co
mm
erci
ald
ecis
ion
we
hav
eto
mak
e:w
e’ve
just
got
tobi
teth
ebu
llet
”(O
bser
vati
on,
com
men
tfr
omu
nio
nst
ewar
din
PS
Ofu
llor
ches
tra
mee
tin
g)(P
SO
3)
PS
O:
Orc
hes
tra
com
mit
tee
ever
yfe
ww
eeks
;fu
llor
ches
tra
ever
yfe
ww
eeks
.
5.5
“Th
ep
laye
rs’
chai
rman
call
eda
full
orch
estr
am
eeti
ng
tod
iscu
ssth
ele
ader
situ
atio
n,
wh
ich
had
com
eto
ah
ead
beca
use
ofh
isla
ckof
avai
labi
lity
for
reh
ears
als
for
the
fest
ival
con
cert
s.A
tth
em
eeti
ng,
pla
yers
exp
ress
edan
ger
atth
ele
ader
,fo
rh
avin
g‘t
wo
jobs
intw
od
iffe
ren
tco
un
trie
s,’
and
atm
anag
emen
t,fo
rm
akin
ga
late
chan
geto
the
reh
ears
alsc
hed
ule
,w
hic
hth
ele
ader
was
un
able
toac
com
mod
ate.
”(f
ield
not
es,
PS
O)
(PS
O5)
BS
O:
Orc
hes
tra
com
mit
tee
atle
ast
ever
ytw
ow
eeks
;fu
llor
ches
tra
ever
yfe
ww
eeks
;se
nio
rm
anag
emen
t–fu
llor
ches
tra
abou
ttw
ice
aye
ar.
5.6
“We
hav
eor
ches
tral
mee
tin
gsw
her
eev
eryo
ne
has
asa
y...
.[t
he
chai
rman
]an
d[t
he
CE
O]
get
toge
ther
and
then
they
hav
eto
per
suad
eth
e[b
oard
],bu
tan
yof
the
maj
ord
ecis
ion
sev
entu
ally
hav
eto
com
eto
the
wh
ole
orch
estr
a.”
(in
terv
iew
,L
SO
boar
dm
embe
r)(L
SO
3)
LS
O:
Orc
hes
tra
com
mit
tee
ever
yfe
ww
eeks
;fu
llor
ches
tra
mee
tin
gsev
ery
few
wee
ks.
Ad
hoc
soli
cita
tion
ofvi
ews
5.7
“Im
eet
once
ever
ytw
om
onth
sw
ith
hal
fa
doz
enm
embe
rsof
the
orch
estr
a,an
dit
’sre
ally
thei
rfo
rum
tosa
yw
hat
they
thin
kis
wro
ng
wit
hso
me
ofou
rd
ecis
ion
s.”
(In
terv
iew
,P
SO
par
t-ti
me
arti
stic
advi
sor)
(PS
O3)
PS
O:
Boa
rdch
airm
anw
ith
pla
yer
rep
rese
nta
tive
s.
5.8
“[T
he
sen
ior
pro
du
cer]
doe
sco
me
and
sit
dow
nw
ith
pla
yers
atm
ealt
imes
som
etim
es,
soI
exp
ect
he
gets
som
efe
edba
ckth
atw
ay..
.th
at’s
pro
babl
yth
em
ain
rou
tew
her
eh
ege
tsin
form
atio
nab
out
how
peo
ple
feel
.”(i
nte
rvie
w,
BS
Oor
ches
tra
com
mit
tee
mem
ber)
(BS
O3)
BS
O:
On
goin
gd
iscu
ssio
ns
betw
een
CE
Oan
dm
anag
ers.
5.9
“We
wer
ese
ekin
ga
new
con
du
ctor
,an
dth
ere
was
tobe
ase
arch
.W
e[t
he
boar
d]
thou
ght,
‘Wh
atth
eh
ell
do
we
know
abou
tco
nd
uct
ors?
...
You
,or
ches
tra,
goaw
ayan
dd
raw
up
ali
stof
con
du
ctor
syo
uw
ould
like
toh
ave,
ortr
you
t.’”
(in
terv
iew
,P
SO
boar
dm
embe
r)(P
SO
2)
LS
O:
Fre
quen
tm
eeti
ngs
betw
een
CE
Oan
din
div
idu
alm
anag
ers.
Ad
hoc
retr
eats
5.10
Th
ep
urp
ose
ofth
eP
SO
sen
ior
man
agem
ent
away
-day
was
“to
brai
nst
orm
wh
ere
we
beli
eve
the
orch
estr
ash
ould
bein
the
med
ium
term
”(o
verv
iew
doc
um
ent,
PS
Ose
nio
rm
anag
emen
tst
rate
gyaw
ay-d
ay)
(PS
O1)
PS
O:
On
ese
nio
rm
anag
emen
tre
trea
t;tw
obo
ard
retr
eats
.
5.11
Th
ep
laye
rs’
chai
rman
com
men
ted
that
“th
em
ajor
ity
ofth
eor
ches
tra
feel
the
rela
tion
ship
wit
h[t
he
pri
nci
pal
con
du
ctor
]h
asre
ach
edth
een
dof
its
nat
ura
lli
fe”
(Obs
erva
tion
,co
mm
ent
from
pla
yer’
sch
airm
an,
PS
Obo
ard
away
-day
)(P
SO
2)
BS
O:
Man
agem
ent
retr
eat
ever
yfe
wm
onth
s;C
EO
-ove
rsee
rre
trea
tev
ery
few
mon
ths.
LS
O:
Non
e.
In quotation 7.5, the London’s CEO describes thediversity of interests associated with putting on aconcert, arguing that every stakeholder essentiallywanted a different concert.
Leader sensegiving enablers. Our second set offindings in relation to leader sensegiving concernthe two conditions that enabled it. The first was“leaders’ issue-related expertise”: when leadershad expertise relating to an issue, they were bet-ter able to shape others’ interpretations of it. Ta-ble 8 summarizes the types of expertise associ-ated with different issues and the leaders in eachorchestra who possessed that expertise. The rel-evant areas of expertise for each issue are thesame as those for stakeholders (Table 3) with theaddition of one area, “awareness of how to bal-ance different income-generating opportunitiesto create a coherent portfolio for the orchestra,”which was possessed by the orchestras’ leadersbut not by any of their stakeholders. As withstakeholder expertise, we assessed leader exper-tise in each area on the basis of knowledge of
each leader’s professional background and train-ing and evidence of expertise shown or describedduring fieldwork interviews and observation. Forexample, although the leaders in the Broadcast-ing and London orchestras had high levels ofmusical training and many years as senior artsadministrators, and thus were able to judge the“fit” between a conductor and the orchestra (ex-pertise area 6), the Provincial leaders had nar-rower musical backgrounds or none at all and sodid not possess this kind of expertise. When itcame to knowing the reasons for the poor perfor-mance of an orchestral musician, however, theonly leader with this understanding was the Lon-don’s chairman, who was a player himself andtherefore generally well acquainted with his fel-low musicians’ situations and difficulties.
The second condition that seemed to enableleader sensegiving was the “performance of the or-ganization in an issue domain” which, whenstrong, provided leaders with an acceptable basisfor engaging in sensegiving about that issue. This
TABLE 6Data Supporting the Theme “Perceptions of Issue as Highly Uncertain”
Associated First-Order Concepts Representative Quotations
Ambiguous issue 6.1 “People such as [this board member] have a very clear perception of his [the principal conductor’s]strengths and weaknesses, but that’s different from the public saying he’s great, and what seems to me tohave happened over the last few weeks has been about the managing of perceptions. Managing therelationship is second to managing perceptions.” (interview, PSO chief executive) (PSO2)
6.2 “I think the fact that people have pointed at the [PSO] and said, ‘What is it all about?’ has been faircomment, and has been largely fog that surrounds [the principal conductor], given the title of music directorand artistic director, but never being able to articulate what he’s doing.” (interview, PSO chief executive)(PSO1)
6.3 “It’s very rare to have a unanimous appointment. There’s always someone who says, ‘I can live with that’”(interview, LSO string section principal) (LSO4)
Unpredictable issue 6.4 “It’s constantly evolving—I could go and meet someone today and we’d talk about certain projects and itwould change the whole plan for a month. You suddenly get a brilliant idea and so you change the wholething.” (interview, BSO senior producer) (BSO1)
6.5 “The managing director explained that there had been a very difficult period through December 1995 whenit was reported that [a conductor] had agreed to be principal conductor of another European orchestra.However, nothing has been signed and the [LSO] remains in negotiation with him. The situation is nowclearer, but whether the orchestra will be in a position to sign the contract depends on the Arts Council’sgrant. The orchestra needs to have sufficient extra subsidy to be able to finance his retainer, substantialconcert fee and travel expenses.” (minutes, LSO Trust) (LSO2)
6.6 “We couldn’t make even the next stage of the negotiations without knowing whether the Arts Council wasgoing to give us more money to fund the residency. I needed to know that all that money was in the bag sowe could fund our ongoing programme of artistic work before actually moving.” (interview, LSO managingdirector) (LSO1)
6.7 “There’s some uncertainty about whether [the orchestra] finally accepted the pay freeze.” (comment, seniormanagement away-day, PSO, 7/96)
6.8 “[The BSO] is such a huge organization compared to what I was in before, where you had your finger onevery penny, you knew exactly what was happening. It’s very difficult when it’s such a big organization andyou’re part of it. It’s very difficult to put realistic figures on anything to be honest. It’s all a bit of aguesstimate, and I think it probably always will be.” (interview, BSO senior producer) (BSO7)
2007 73Maitlis and Lawrence
second-order theme aggregated the following first-order concepts: “artistic performance and reputa-tion,” “financial performance,” and “operationaleffectiveness and efficiency.” Table 9 presents quo-tations from the data that illustrate each of thesefirst-order concepts. For example, quotation 9.2 isan observation made by the Provincial’s CEO at ameeting with the full orchestra, acknowledging the
significant problems the orchestra had experiencedwith its artistic policy. To be associated withsuch problems clearly affected the Provincialleaders’ credibility in this issue domain. In con-trast, quotation 9.6 describes the good financialperformance achieved in the London through itsincome-generating activities. The fact that theleaders had achieved a surplus in an industry
TABLE 7Data Supporting the Theme “Perceptions of Stakeholder Environment as Complex”
Associated First-Order Concepts Representative Quotations
Numerous stakeholdersconnected to issue
7.1 “The technical answer is that the board should have the artistic strategy. The realityis that [the principal conductor] sort of produced repertoire by sparking off [the part-time artistic advisor], [the deputy chief executive], and [the external consultant withknowledge of the PSO]. . . . Emerging out of that process is ultimately [the part-timeartistic advisor] producing rolling drafts of the repertoire.” (interview, PSO CEO)(PSO1)
7.2 “This contractual thing . . . started with me saying, ‘I want more rights, I want tobuy more rights.’ And it went to London [the BSO overseers], and London picked itup and said, ‘Okay, what do we do about rights?’ It also started at the same timewhen [the new head of music of radio] came on board saying, ‘We don’t have therights on the orchestras. Ridiculous. How can we be a broadcasting orchestra if wedon’t have broadcasting rights on them? It’s unbelievable. Go back and rethink it,’which we all did. So that starts a process going where we then start talking to mymanagement, my team, the people round here, locking into other BBC orchestralmanagers, and devising a plan to say, ‘This is what we want,’ and then we have todiscuss it with the union and the players.” (interview, BSO CEO) (BSO6)
7.3 “You get a consensus, and then I can go to the [board] and say, ‘We need a Councilof Principals meeting, because I want to appoint X,’ and they say, ‘Have you spokento the section?’” (interview, LSO string section principal) (LSO5)
7.4 “Section principals make representation to the [board] if they want to do something.The [board] then has to undertake a whole series of meetings with the individual, thesection principals generally, their immediate colleagues, and the [board] then have todecide what to do, and do it.” (interview, LSO chairman) (LSO4)
Divergent interests amongconnected stakeholders
7.5 “The general public want good concerts—good conductors with pleasant repertoirewhich they know, with accessible ticket prices. Record companies only want you togive concerts of music which you’re going to record, which they can sell. . . . TheArts Council want us to do Harrison Birtwhistle for two hours nonstop, subsidizingempty seats. The problem with that is it requires lots of rehearsals, a big orchestra.The sponsors want a great night out with a star conductor, a star soloist, Beethoven 9,Dvorak 9, or Beethoven 5, so they can hum away the tune, concert as short aspossible, interval as long as possible, and get to the grub as soon as they canafterwards. Promoters outside London want something no one local can give them:high-profile conductors, for no money.” (interview, LSO CEO) (LSO8)
7.6 “There’s often a variance between the wind and the strings. Often conductors thatreally work well with the strings spend so much time with the strings, the windplayers get bored silly. But they’re usually quite good conductors.” (interview, BSOsenior producer) (BSO3)
7.7 “1) There was a general discussion on the subject of the additional rights the BBC isaiming to acquire. The committee expressed the concern of the orchestra about theunrealistically large number of CD sessions desired by the BBC. . . . The committeesuggested that [the CEO] should talk to the orchestra on these issues.” (minutes, BSOorchestra committee meeting) (BSO6)
7.8 “There’s 6 percent who can take a balanced view and look beyond their ownfeelings, 10 percent at the other end who just play the concert, and then you’ve gotthe majority in the middle who are voluble, irresponsible and expect to changeconductors every year.” (interview, PSO board chairman-elect) (PSO2)
74 FebruaryAcademy of Management Journal
where deficits are common doubtless increasedtheir credibility in the eyes of others, enablingthem to engage in sensegiving. The third first-order concept was “operational effectiveness andefficiency.” Quotation 9.7 captures the problemsthe Provincial had experienced in the area ofhiring. As a result, leaders were less likely toengage in sensegiving around this issue.
DISCUSSION
The findings we report above describe the triggersand enablers of sensegiving for stakeholders and lead-ers in 27 issue domains. In this section, we discussthese findings, develop a set of propositions that for-malize the relationships suggested by our data, andpresent an overarching model of the conditions asso-ciated with sensegiving in organizations.
TABLE 8Types of Issue-Related Expertise Possessed by Leaders
Related Types of Expertise by Issuea
Leaders Possessing These Types of Expertiseb
PSO BSO LSO
Programming8.1 Musical knowledge of repertoire Artistic director CEO, senior producer CEO, chair8.2 Knowledge of what repertoire will attract audiences None CEO, senior producer CEO8.3 Understanding of the orchestra’s artistic strategy Artistic director CEO, senior producer CEO
Principal conductor appointment8.4 Knowledge of potential conductors for the position None CEO, senior producer CEO8.5 Musical appreciation of conductors’ styles, abilities and potential Artistic director CEO, senior producer CEO, chair8.6 Sense of the “fit” between a conductor and the orchestra None CEO, senior producer CEO8.7 Knowledge of typical conductor contractual arrangements None CEO, senior producer CEO
Guest conductors8.8 Awareness of potential conductors for the position Artistic director CEO, senior producer CEO, chair8.9 Musical appreciation of conductors’ styles, abilities, andpotential
Artistic director CEO, senior producer CEO, chair
Player performance8.10 Awareness of and ability to judge musical performance Artistic director CEO, senior producer CEO, chair8.11 Knowledge of reasons for poor performance None None Chair8.12 Knowledge of the procedures for dealing with playerperformance issues
Artistic director CEO CEO, chair
Player hiring8.13 Ability to judge musical performance of job candidates Artistic director CEO, senior producer CEO, chair8.14 Awareness of available candidates Artistic director CEO Chair
Player pay and contracts8.15 Knowledge of collective agreements, labor relations Artistic director CEO CEO, chair8.16 Negotiation skills None None CEO, chair8.17 Knowledge of other orchestras’ contractual agreements Artistic director CEO, senior producer CEO, chair
Cost cutting8.18 Knowledge of organization’s operational costs CEO CEO CEO8.19 Understanding of management accounting CEO CEO CEO8.20 Knowledge of where organizational slack lies CEO CEO CEO
Income generation8.21 Knowledge of income-generating opportunities CEO, artistic director CEO, senior producer CEO8.22 Awareness of how to balance different income-generatingopportunities to create a coherent portfolio for the orchestra
Artistic director CEO CEO
Collaborative ventures8.23 Awareness of opportunities for collaboration CEO, artistic director CEO CEO8.24 Awareness of strengths/weaknesses of potential collaborators Artistic director CEO CEO8.25 Awareness of environmental factors that (en/dis)couragecollaboration
Artistic director CEO CEO
a Based on ethnographic observation by the first author.b Based on professional background and training, and evidence of expertise shown or described during fieldwork interviews and
observation.
2007 75Maitlis and Lawrence
Sensegiving Triggers
Our first research question asks what conditionstrigger sensegiving by stakeholders and by leaders.We found distinct sets of triggers for each of thesegroups. For stakeholders, sensegiving was triggeredby issues that they perceived as important either tothemselves, to a stakeholder group whom they rep-resented, or to the organization at large, and by anassessment of organizational leaders as incompe-tent with respect to these issues. For leaders, sense-giving was triggered by issues that they perceivedas ambiguous, unpredictable, and involving nu-merous diverse stakeholders. Both of these findings
contribute to existing research on what motivatesorganizational actors to engage in sensegiving.
Stakeholder sensegiving and bounded respon-sibility. Critical to understanding when stakehold-ers engaged in sensegiving is what we refer to as“bounded responsibility”: stakeholders showedlimited motivation to construct meaning for others,doing so only when they perceived a need to takesome responsibility for an issue.
The degree to which stakeholders are motivatedto influence the interpretation of issues is an un-settled question in management research. On theone hand, a common assumption in much of theliterature on political decision making and stake-
TABLE 9Data Supporting the Theme “Performance of the Organization”
Associated First-OrderConcepts Representative Quotations
Artistic performance andreputation
9.1 “The [Arts Council appraisal] team learned that there was some audience resistance to the newinitiatives. . . . In the light of the [PSO]’s experience in the 19[XX–XY] season, the team wasreassured to learn that the 19[XZ] season, while retaining an emphasis on adventurousprogramming and a number of the other new features, will return to a clearer artistic andaudience focus for the individual series.” (Arts Council PSO appraisal report) (PSO1)
9.2 “The board accept that the artistic policy hasn’t been successful in delivering an audience”(observation, PSO CEO–full orchestra meeting) (PSO1)
9.3 “[The LSO] is increasingly the first choice of orchestra for conductors and soloists performingin London. . . . [There are 11 international conductors who] choose to work with the [LSO] inpreference to other London orchestras. . . . The list of conductors working regularly with theorchestra is the strongest evidence of the status of the [LSO] in international musical circles.”(LSO submission to panel of independent evaluators) (LSO3)
Financial performance 9.4 “The board’s responsibility is to oversee the well-being of the organization, and that includesoverseeing the two main appointments of the organization, that of the chief executive, and thatof the principal conductor. And I feel personally let down by our board because we now findourselves in a financial and artistic difficulty, and they tell me that they took advice on theappointment of the principal conductor from the chief executive, and they also paid heed to theorchestra’s wishes. But my bone of contention is that they approved the contract he was given,which has proved to be almost the death knell of the organization, due to his high financialoutlay, and the lack of return artistically. . . . We just have not had a return on our investment.(interview, PSO player director) (PSO2)
9.5 “Despite the [PSO’s] confidence that it would meet its cash obligations during the year, theteam is concerned as to how the company proposes to finance the accumulated deficit of£610,000 which it is projecting at the end of the current financial year.” (Arts Council PSOappraisal report) (PSO7)
9.6 “The season had been carefully balanced to achieve a small year-end surplus.” (minutes, LSOboard meeting) (LSO8)
Operational effectivenessand efficiency
9.7 At the start of the study period, the orchestra had been without one of the principal woodwindplayers for four years. This position and the other key ones that became vacant at around thistime remained so for over a year into the fieldwork period. All three vacancies resulted fromplayer resignations that happened prior to the study period. (PSO field notes) (PSO5)
9.8 The orchestra had made severe cuts in expenditure some years before, and now operated withminimal overheads. In any case, because of the freelance nature of the London orchestras, theycarried a much smaller overhead than the regional and BBC contract orchestras, only having topay their musicians when they worked. The one major cost-cutting exercise undertakenconcerned the players’ fees. This saved the orchestra a considerable amount of money,precluding the need for further large-scale cuts. (LSO field notes) (LSO7)
76 FebruaryAcademy of Management Journal
holder theory is that organizational stakeholderswill be motivated to influence issues in which theyhave a vested interest (Agle & Mitchell, 1999;Mangham, 1986; Pettigrew, 1973). On the otherhand, research on behavioral decision making(March, 1994) and on employee participation(Heller, 1998; Pasmore & Fagans, 1992) suggeststhat employees and other stakeholders become in-volved in issues fluidly, often choosing not to par-ticipate, even in issues that directly affect theirinterests. Our findings, and particularly the notionof bounded responsibility, offer a way of reconcil-ing these conflicting views and help us understandwhen stakeholders are likely to be motivated toengage in sensegiving. Although sensegiving, con-cerned as it is with influencing the interpretationsof others, is undoubtedly a political activity, ourfindings suggest that a strictly political understand-ing of sensegiving limits understanding of stake-holder motivations. We found that sometimes self-interest motivated stakeholders to engage insensegiving but also that sometimes feelings of re-sponsibility or organizational stewardship moti-vated it: a perceived gap in leadership around is-sues that were at least as important for theorganization at large as for any individual stake-holder group often triggered stakeholder sense-giving. Moreover, the notion of bounded responsi-bility may be a useful addition to existing researchon stakeholder participation. Organizational stake-holders have limits on their available time andenergy that act as constraints on their motivation toengage in sensegiving. In the face of these con-straints, sensegiving seemed to be triggered bystakeholders’ perceived responsibility for impor-tant issues. Thus, the responsibility sought bystakeholders in organizational sensemaking is“bounded” by the combination of their perceptionsof the issue’s importance and whether they per-ceive themselves as needed to co-construct mean-ing around the issue because of their leaders’ in-competence. This formulation leads to our firstproposition.
Proposition 1a. Stakeholders will be motivatedto engage in sensegiving when they perceive anissue as having important consequences forthemselves, a stakeholder group whom theyrepresent, or their organization at large andperceive their leaders as lacking competence inrelation to that issue.
Leader sensegiving and complex sensemakingenvironments. Weick noted that environmentalcomplexity creates “occasions for sensemaking”(1995: 87). We found that complex sensemakingenvironments created occasions for leader sense-
giving, as leaders sought to construct stories thatcould make sense of unpredictable, ambiguous is-sues for stakeholder groups with divergent inter-ests. This finding advances understanding of whenleaders will be motivated to engage in sensegiving.Whereas previous studies have focused primarilyon organizational change as the trigger for leadersensegiving (Bartunek et al., 1999; Corley & Gioia,2004; Gioia & Chittipeddi, 1991), our study showsthat leaders are motivated to engage in sensegivingin a broad range of contexts: leader sensegiving wasfound in relation to issues such as income genera-tion, key appointments, and pay and contract re-view, which were not part of significant organiza-tional change but were associated with complexsensemaking environments. This finding also helpsto explain why organizational change might moti-vate leader sensegiving. Change increases the un-predictability and ambiguity of many issues andincreases the salience of divergent stakeholder in-terests. Together, these may prompt sensegiving onthe part of leaders so that the organization and itsstakeholders can construct shared, coherent ac-counts of a situation. In Corley and Gioia’s (2004)terms, complex sensemaking environments create a“sensegiving imperative” for leaders. These dy-namics lead to our next proposition.
Proposition 1b. Leaders will be motivated toengage in sensegiving when they perceive is-sues as ambiguous and unpredictable and/oras involving numerous, diverse stakeholders.
We have analyzed the triggers of stakeholder andleader sensegiving separately and identified dis-tinct triggers for each set of actors. Without separateanalyses of leaders and of stakeholders, these find-ings would not have emerged. There is also value,however, in looking at both leaders and stakehold-ers to identify common roots to their triggers. Tak-ing the commonalities and differences together, wesee our findings as suggesting that, in general, or-ganizational actors’ sensegiving is triggered by theperception or anticipation of a “sensemaking gap.”Figure 2 depicts an overview of our proposedmodel. Stakeholders commonly expect leaders tofulfill the main sensegiving roles in organizations(Corley & Gioia, 2004; Gioia & Chittipeddi, 1991)and so perceive or anticipate sensemaking gapswhen issues are important but leaders lack compe-tence with respect to those issues. Our study sug-gests that leaders perceive or anticipate sense-making gaps when issues are ambiguous andunpredictable—and hence demand sensemaking(Weick, 1995)—and relevant stakeholders are nu-merous and divergent in their interests, and sounlikely to construct a shared account of the issue
2007 77Maitlis and Lawrence
on their own. Thus, we propose “perception oranticipation of a sensemaking gap” as an aggregatedimension to describe the general condition thattriggers sensegiving on the part of both stakehold-ers and leaders.
The notion of a sensemaking gap as a trigger forsensegiving adds to sensemaking theory and re-search. First, it provides a general description ofconditions under which organizational actors willbe motivated to engage in sensegiving. Althoughthere has been significant discussion of what leadsindividuals and groups to engage in sensemaking(Weick, 1995), the conditions that trigger sense-giving have been relatively unexamined, with mostresearch tied to specific organizational contexts,such as major change (Corley & Gioia, 2004; Gioia &Chittipeddi, 1995; Snell, 2002). The concept of asensemaking gap is intended to describe a trigger-ing condition for sensegiving that can be appliedacross contexts and actors. Second, it reflects thepractical and political tensions associated withsensegiving. Sensegiving is a political process, in-fluencing the interpretations of others and conse-quently affecting their decision making. In our
study, however, we only saw sensegiving beingtriggered by the perception or anticipation of asensemaking gap, which suggests that organization-al actors interested in exerting influence will moreoften engage in less costly tactics that perhaps havemore predictable impacts than sensegiving. As aform of influence, sensegiving carries with it sev-eral uncertainties, including whether others willadopt one’s preferred definition of organizationalreality and, if they do, what the consequences willbe for their actions and decision making. Many ofthese uncertainties can be avoided with use of moredirect forms of influence, such as exchange or bar-gaining (Kipnis, Schmidt, & Wilkinson 1980; Mait-lis, 2004). More direct influence routes may not beavailable, however, when a sensemaking gap oc-curs, because actors will lack a shared definition oforganizational reality that might provide the foun-dation for exchange or bargaining. Thus, the per-ception or anticipation of a sensemaking gap maylead organizational actors to engage in sensegivingas an influence strategy because other approachesseem less likely to be successful.
FIGURE 2Conditions Associated with Sensegiving in Organizations
78 FebruaryAcademy of Management Journal
Sensegiving Enablers
Our second research question asks what condi-tions enable sensegiving by stakeholders and byleaders. As with triggers, we found distinct sets ofenablers for each of these groups. Stakeholdersensegiving was enabled by their possession of is-sue-related expertise, legitimacy, and opportunitiesto engage in sensegiving. Leader sensegiving wasenabled by leaders’ possession of issue-related ex-pertise and by their organizations already perform-ing well in the relevant issue domain. Each ofthese findings contributes to existing research onwhat helps organizational actors to engage insensegiving.
Stakeholder sensegiving capacity. First, ourfindings extend previous research on stakeholdersensegiving, providing a broader understanding ofthe relationship between stakeholder sensegivingand organizational context than previous research(Balogun & Johnson, 2004; Dutton et al., 2002;Westley, 1990). The enablers of stakeholder sense-giving that we observed—expertise, legitimacy andopportunity—suggest an image of sensegiving thatis highly “situated,” embedded in the relationalidentities of actors and in the social and technolog-ical contexts in which actors are operating. Thisview opens up the potential for an integration ofsensemaking research with the emerging practiceperspective (Schatzki, Knorr Cetina, & Von Sav-igny, 2001; Whittington, 2003). Rather than focus-ing, for instance, on sensegiving as the employmentof linguistic devices, such as metaphor or narrative,that transcend specific organizational contexts, thisperspective suggests that sensegiving be under-stood as a situated practice grounded in and depen-dent on the social and physical technologies ofspecific organizations at specific times. Althoughstorytelling, for example, has gained prominence asa form of sensegiving (Boje, 1991; Dunford & Jones,2002; Snell, 2002), our study suggests that the abil-ity to engage in sensegiving goes beyond simplytelling a good story: for stakeholders to engage insensegiving, they must tell sensible stories (draw-ing on relevant expertise) at the right time andplace (opportunity) and occupy a social positionthat leads others to listen (legitimacy). Thus, al-though sensegiving may often be prompted by or inthe service of organizational change, our study sug-gests that sensegiving itself is a highly institution-alized form of social practice, with context-specificrules and resources. From a practice perspective,sensegiving should be studied as “creative andknowledgeable work . . . which may or may notachieve its desired ends and which interacts withexisting social and technological structures in un-
intended and unexpected ways” (Lawrence & Sud-daby, 2006: 219).
The importance of expertise, legitimacy, and op-portunity for stakeholder sensegiving leads to oursecond proposition.
Proposition 2a. Stakeholders will be more ableto engage in sensegiving in domains in whichthey possess issue-related expertise and/or le-gitimacy and in which organizational pro-cesses provide them with opportunities to en-gage in sensegiving.
Leader sensegiving capacity. Second, ourfindings extend those of previous studies onleader sensegiving. Earlier work has linkedleader sensegiving to organizational changeprompted by an external threat (Dunford & Jones,2000) or a newly demanding environment (Gioia& Chittipeddi, 1991). Our findings suggest thatsuch contexts will be associated with high levelsof leader sensegiving only under certain condi-tions: when the threatened aspects of an organi-zation are those in which the leaders have exper-tise and in which the organization has beenperforming strongly. All sensegiving is in re-sponse to troubling, uncertain, or confusing is-sues, but our study suggests that if an organiza-tion’s leaders have a limited appreciation of thethreatened area, and if the threat is affecting analready poorly performing aspect of the organiza-tion, leaders will find it difficult to engage insignificant sensegiving. The role of issue-relatedorganizational performance in leader sensegivingstems from two sets of dynamics. First, sensegiv-ing in complex environments is a time-consum-ing process that demands significant attentionand energy from leaders if their accounts are to bepersuasive. In our study, in issue domains inwhich the orchestras were not performing effec-tively, leaders were typically consumed by day-to-day problems that required resolutions basedon quick compromise or immediate action. Thisleft them with little time or energy to manageinterpretations of issues through the constructionof accounts. Second, in these organizations, lead-ers were prominent figures to whom the orches-tra’s success or failure was often attributed. Theroles of expertise and issue-related organization-al performance in enabling sensegiving lead toour final proposition.
Proposition 2b. Leaders will be more able toengage in high levels of sensegiving in issuedomains in which they possess relevant exper-tise and/or in which their organizations arealready performing effectively.
2007 79Maitlis and Lawrence
Our analysis of sensegiving enablers involvedseparate analyses for stakeholders and leaders,which provided us with valuable insights intothe different conditions that allow motivatedleaders and stakeholders to engage in sensegiv-ing. Like sensegiving triggers, however, these en-ablers share important commonalties. Looking atboth stakeholders and leaders, we can break theenablers that we identified into two categories.First, there are those that affect the discursiveability of organizational actors—their ability toconstruct and articulate persuasive accounts ofthe world. For stakeholders, these conditionswere issue-related expertise and legitimacy. Wesaw that the discursive ability of different stake-holder groups varied tremendously within eachissue domain, as it did for each stakeholder groupacross different issues, as their professional back-grounds and training and organizational rolesand responsibilities interacted with the issue athand. Organizational leaders’ legitimacy largelycame with their formal roles, but their issue-related expertise and their organization’s perfor-mance around that issue affected their discursiveability: the former affected the skill with whichleaders could construct accounts, and the latteraffected their credibility in an issue domain.Thus, in Figure 2, we propose “discursive abil-ity” as the second aggregate dimension in ourmodel of the conditions associated with sense-giving in organizations. The importance of dis-cursive ability as an enabler of sensegiving acrossactors opens up to sensegiving research the meth-odologies and frameworks developed within dis-course analysis: textual analyses focusing on thecontent of sensegiving; investigations of whenand how sensegivers invoke other legitimate textsand fields of discourse; and explorations of therole of different rhetorical strategies, includingappeals to emotion, logic, or status (Phillips &Hardy, 2002).
The second set of enablers that we identifiedfacilitate the organizational processes throughwhich sensegiving occurs. For stakeholders,sensegiving was facilitated by the opportunitiesthey encountered to shape others’ interpretationsof reality; often lacking direct control of the or-ganizational processes associated with an issue,stakeholders were more able to engage in sense-giving when formal and informal opportunitiesallowed them to articulate accounts of the issuethat would be heard by relevant others. Leadersensegiving was facilitated not so much by op-portunity, over which they tended to have signif-icant control, but rather by the issue-related per-formance of their organizations, which, if strong,
made fewer demands on their time and attention,enabling them to engage in sensegiving. In viewof these commonalities, we propose “process fa-cilitators” as the third aggregate dimension in ouroverarching model. This aggregate dimensionhighlights the organizational nature of sensegiv-ing: its dependence on organizational routines,practices, and structures that can either hinder orenable sensegiving by both stakeholders andleaders. Research on sensegiving needs, there-fore, to include close attention to the organiza-tional context in which it occurs. This insightbuilds on and extends previous work that hashighlighted the importance of such elements asrole structure (Balogun, 2003; Weick, 1993) andchange management processes (Bartunek et al.,1999; Gioia & Chittipeddi, 1991) in effectingsensegiving in organizations, and it demonstratesthe value of conducting contextually sensitivestudies for future research in the area.
CONCLUSION
In this article, we have drawn on a study ofsensemaking in nine issue domains and in threecomparable organizations to identify conditions as-sociated with sensegiving by stakeholders and byleaders. Our research questions were aimed at iden-tifying the triggers and enablers of sensegiving. Forstakeholders, we found that a sense of boundedresponsibility triggered sensegiving; they engagedin high levels of sensegiving when they perceivedissues as important and leaders as incompetentwith respect to that issue. For leaders, sensegivingwas triggered by complex sensemaking environ-ments—those that were ambiguous and unpredict-able and that involved numerous stakeholders withdivergent interests. More generally, we argued thatthe perception or anticipation of a sensemaking gaptriggered sensemaking; leaders or stakeholders per-ceived the need for an account of an issue andcould not see who else would provide it. In exam-ining the enablers of sensegiving, we found issue-relevant expertise, legitimacy, and an opportunityfor sensegiving enabled stakeholder sensegiving,while issue-relevant expertise and effective organ-izational performance in the issue domain enabledleader sensegiving. Looking at both these groups,we argued that sensegiving was enabled by thecombination of a discursive ability that allowedactors to construct and articulate persuasive ac-counts of the world, and process facilitators in theform of organizational routines, practices, andperformance.
Our study does, of course, have limitations.
80 FebruaryAcademy of Management Journal
One stems from our data reduction strategy. Inorder to produce a manageable and analyzabledata set, we assessed leader and stakeholdersensegiving levels for each of the 27 issue do-mains and consequently did not address changesin those levels over the life of the issue. Althoughexamining those changes would be a worthwhileanalysis, we believe that our approach provides auseful foundation for the study of sensegivingthat could be extended in future studies. A sec-ond issue concerns the transferability of thisstudy’s findings to other research contexts. Whileorchestras are similar in important ways to manymedium-sized companies, they offer productsand services different from those of most firmsand operate in a sector with specific dynamicsand a distinctive history. We do not believe thatthese differences are likely to affect the condi-tions of sensegiving in ways that diminish thestudy’s findings, but this remains a question thatonly further research can answer. Despite theselimitations, we believe that our study points toimportant future directions for sensegivingresearch.
Future Research Directions
One issue for future research is whether sense-giving by certain organizational groups may attimes become part of a self-perpetuating system.For instance, our findings suggest that situations inwhich leaders are both lacking in issue-relatedcompetence and perceived as such by stakeholdersare likely to be associated with high levels of stake-holder sensegiving and low levels of leader sense-giving. Such a situation may create a positive feed-back cycle, reinforcing stakeholder sensegivingthrough two mechanisms: (1) stakeholder sense-giving may lead stakeholders to gain expertise andlegitimacy through their involvement with an is-sue, while leaders do not do so, reinforcing stake-holders’ ability to engage in sensegiving and (2) asstakeholders gain expertise and legitimacy regard-ing an issue, they may become more confident andcritical in their assessments of leaders’ competencewith respect to that issue, thus reinforcing thatsensegiving trigger. A similar dynamic may applyin reverse, whereby leaders are, and are perceivedas, more expert in relation to an issue than arestakeholders: stakeholder perceptions of highleader competence will diminish their own moti-vations for sensegiving, and leader expertise willenable leader sensegiving. Such patterns may beproblematic because both leader and stakeholdersensegiving have significant, positive effects (Mait-
lis, 2005), and so patterns that reduce either in aconsistent, ongoing manner may lead to less thanoptimal outcomes for organizations and their mem-bers (Maitlis & Lawrence, 2003). Sensegiving re-search might therefore valuably draw on ap-proaches that have been useful in the exploration ofother self-fueling cycles, such as “vicious circles”of bureaucracy (Crozier, 1964) and efficacy-perfor-mance spirals (Lindsley, Brass, & Thomas, 1995),which cannot be completely understood by meansof traditional, linear notions of cause and effect(Weick, 1979). Such exploration could be facili-tated by modeling tools from systems dynamics, tocapture the parallel and recursive properties of self-fueling cycles (Wolstenholme, 1990).
A second research implication is related to thepossibility that organizational stakeholders andleaders might, if interested in affecting patternsof influence around an issue and aware of theconditions associated with sensegiving, be ableto affect those conditions and, consequently, in-fluence the extent of others’ sensegiving. For ex-ample, organization members might work tostructure organizational change and strategy pro-cesses in ways that facilitate sensegiving by bothstakeholders and leaders and thus increase thepotential influence of either or both groups. Inthe context of an important change initiativesuch as a merger, leaders could alert stakeholdersto the importance of the issue, as well as supportthem with the legitimacy and opportunity thatwould facilitate their sensegiving. Similarly,stakeholders might make leaders aware of thecomplexity of their sensemaking environmentsand provide leaders with the time to developinfluential accounts of an issue and its potentialsolutions. The possibility also exists, of course,that leaders or stakeholders may work to mini-mize the triggers and enablers of sensegiving byothers, in order to more closely control the sen-semaking and decision making around an issue.Leaders might, for instance, attempt to minimizestakeholders’ perceptions of an issue’s impor-tance and work to discredit their expertise andlegitimacy. Equally, if stakeholders aim to over-whelm the influence of leaders on an issue, theymight undermine perceptions of leaders’ exper-tise, as well as perceptions of the past perfor-mance of relevant areas of an organization. Thesedynamics point to the importance of research thatexamines the strategies through which stakehold-ers and leaders shape the conditions of sensegiv-ing, and perhaps the circumstances associatedwith such strategies.
2007 81Maitlis and Lawrence
Concluding Thoughts
As Gioia and Chittipeddi (1991) first noted,sensegiving is a critical process through whichissues are constructed and interpreted in organi-zations. Issues are not objectively defined and donot appear in the same form to all organizationmembers at the same time: issues are noticed,shaped, interpreted, and sold by some membersto others (Dutton et al., 2002), with importantorganizational consequences (Maitlis, 2005).Identifying conditions that trigger and enablesensegiving by leaders and stakeholders in organ-izations fills an important gap in understandingthis process and points to interesting areas forfuture research.
REFERENCES
Agle, B. R., & Mitchell, R. K. 1999. Who matters to CEOs?An investigation of stakeholder attributes and sa-lience. Academy of Management Journal, 42: 507–525.
Allmendinger, J., & Hackman, J. R. 1996. Organizations inchanging environments: The case of East Germansymphony orchestras. Administrative ScienceQuarterly, 41: 337–369.
Balogun, J. 2003. From blaming the middle to harnessingits potential: Creating change intermediaries. BritishJournal of Management, 14: 69–83.
Balogun, J., & Johnson, G. 2004. Organizational restruc-turing and middle manager sensemaking. Academyof Management Journal, 47: 523–549.
Bartunek, J., Krim, R., Necochea, R., & Humphries, M.1999. Sensemaking, sensegiving, and leadership instrategic organizational development. In J. Wagner(Ed.), Advances in qualitative organizational re-search, vol. 2: 37–71. Greenwich, CT: JAI Press.
Boje, D. 1991. The storytelling organization: A study ofstory performance in an office-supply firm. Admin-istrative Science Quarterly, 36: 106–126.
Corley, K. G., & Gioia, D. A. 2004. Identity ambiguity andchange in the wake of a corporate spin-off. Admin-istrative Science Quarterly, 49: 173–208.
Crozier, M. 1964. The bureaucratic phenomenon. Chi-cago: University of Chicago Press.
Dunford, R., & Jones, D. 2000. Narrative in strategicchange. Human Relations, 53: 1207–1226.
Dutton, J. E., Ashford, S. J., Lawrence, K. A., & Miner-Rubino, K. 2002. Red light, green light: Making senseof the organizational context for issue selling. Organ-ization Science, 13: 355–372.
Eisenhardt, K. M., & Bourgeois, L. J. 1988. Politics ofstrategic decision making in high-velocity environ-ments: Toward a midrange theory. Academy ofManagement Journal, 31: 737–770.
Gioia, D. A., & Chittipeddi, K. 1991. Sensemaking andsensegiving in strategic change initiation. StrategicManagement Journal, 12: 433–448.
Gioia, D. A., & Thomas, J. B. 1996. Identity, image andissue interpretation: Sensemaking during strategicchange in academia. Administrative Science Quar-terly, 41: 370–403.
Glaser, B., & Strauss, A. 1967. The discovery ofgrounded theory. Chicago: Aldine.
Golden, B. R. 1992. The past is the past—Or is it? The useof retrospective accounts as indicators of past strat-egy. Academy of Management Journal, 35: 848–860.
Heller, F. A. 1998. Influence at work: A 25-year programof research. Human Relations, 51: 1425–1456.
Hirsch, P. M. 1972. Processing fads and fashions: Anorganization set analysis of cultural industry sys-tems. American Journal of Sociology, 77: 639 –659.
Isabella, L. A. 1990. Evolving interpretations as a changeunfolds: How managers construe key organizationalevents. Academy of Management Journal, 33:7–41.
Kipnis, D., Schmidt, S. M., & Wilkinson, I. 1980. Intraor-ganizational influence tactics: Explorations in get-ting one’s way. Journal of Applied Psychology, 65:440–452.
Lawrence, T. B., & Suddaby, R. 2006. Institutions andinstitutional work. In S. R. Clegg, C. Hardy, T. B.Lawrence, & W. R. Nord (Eds.), Handbook of or-ganization studies (2nd ed.): 215–254. London:Sage.
Lawrence, T. B., Mauws, M. K., Dyck, B., & Kleysen, R. F.2005. The politics of organizational learning: Inte-grating power into the 4I framework. Academy ofManagement Review, 30: 180–191.
Lee, T. W. 1999. Using qualitative methods in organi-zational research. Thousand Oaks, CA: Sage.
Lindsley, D. H., Brass, D. J., & Thomas, J. B. 1995.Efficacy-performance spirals: A multilevel per-spective. Academy of Management Review, 20:645– 678.
Maitlis, S. 2004. Taking it from the top: How CEOs in-fluence (and fail to influence) their boards. Organi-zation Studies, 25: 1275–1311.
Maitlis, S. 2005. The social processes of organizationalsensemaking. Academy of Management Journal,48: 21–49.
Maitlis, S. & Lawrence., T. B. 2003. Orchestral manoeu-vres in the dark: Discourse and politics in thefailure to develop an artistic strategy. Journal ofManagement Studies, 40: 109 –139.
Mangham, I. L. 1986. Power and performance in organ-izations: An exploration of executive process. Ox-ford, U.K.: Basil Blackwell.
82 FebruaryAcademy of Management Journal
March, J. G. 1994. A primer on decision making: Howdecisions happen. New York: Free Press.
McNulty, T., & Pettigrew, A. M. 1999. Strategists on theboard. Organization Studies, 20: 47–74.
Miles, M. B., & Huberman, A. M. 1994. Qualitative dataanalysis: An expanded sourcebook (2nd ed.). Bev-erly Hills, CA: Sage.
Pasmore, W. A., & Fagans, M. R. 1992. Participation,individual development, and organizational change:A review and synthesis. Journal of Management,18: 375–397.
Pettigrew, A. 1973. The politics of organizational deci-sion making. London: Tavistock.
Pfeffer, J. 1981. Management as symbolic action. In L. L.Cummings & B. M. Staw (Eds.), Research in organ-izational behavior, vol. 3: 1–52. Greenwich, CT: JAIPress.
Phillips, N., & Hardy, C. 2002. Discourse analysis. Lon-don: Sage.
Pondy, L. 1978. Leadership as a language game. In M. W.McCall & M. M. Lomcirdo (Eds.), Leadership:Where else can we go? 94–95. Durham, NC: DukeUniversity Press.
Schatzki, T. R., Knorr Cetina, K., & Von Savigny, E. 2001.The practice turn in contemporary theory. London:Routledge.
Smircich, L., & Morgan, G. 1982. Leadership: The man-agement of meaning. Journal of Applied Psychol-ogy, 18: 257–273.
Snell, R. S. 2002. The learning organization, sensegivingand psychological contracts: A Hong Kong case. Or-ganization Studies, 23: 549–569.
Strauss, A., & Corbin, J. 1998. Basics of qualitative re-search: Techniques and procedures for developinggrounded theory (2nd ed.). Thousand Oaks, CA:Sage.
Van Maanen, J. 1979. The fact of fiction in organizationalethnography. Administrative Science Quarterly,24: 539–550.
Weick, K. E. 1979. The social psychology of organizing.Reading, MA: Addison-Wesley.
Weick, K. E. 1993. The collapse of sensemaking in organ-izations: The Mann Gulch disaster. AdministrativeScience Quarterly, 38: 628–652.
Weick, K. E. 1995. Sensemaking in organizations.Thousand Oaks, CA: Sage.
Westley, F. 1990. Middle managers and strategy: Themicrodynamics of inclusion. Strategic Manage-ment Journal, 11: 337–351.
Whittington, R. 2003. The work of strategizing and orga-nizing: For a practice perspective. Strategic Organ-ization, 1: 55–65.
Wolstenholme, E. F. 1990. System enquiry: A systemdynamics approach. New York: Wiley.
APPENDIX
Provincial Orchestra: Reengagement ofGuest Conductor Narrative
For each issue domain in each orchestra, we devel-oped a multipage narrative composed primarily of or-dered raw data, with some additional commentary forclarity, which together tell the story of each issue as itunfolded. The following is an excerpt from one of the 27narratives.
The players’ chairman addressed the orchestra com-mittee on this matter in October. He told them that theexecutive director had spoken to him after the last Boardmeeting, explaining that conductor X had offered to do alot of work with the orchestra, including recordings anda special concert series, and he was wondering whetherto give him a title. The committee discussed this issue atsome length, considering the advantages and possibledisadvantages of such an appointment. Many of thesewere raised at the full orchestra meeting called by thechairman a few days later. Some players expressed con-cern that the title would give him power which might beabused. Another said,
“I think we shouldn’t have too much with him artisti-cally, but he is a pull. On the round country tours he’svery popular. If we need to fill the diary. . .” (full orches-tra meeting)
Another replied: “Five or six years ago we didn’t wanthim any more.”
Someone else cautioned, “I don’t think we should betoo negative. Should say we welcome him, but not toomuch.”
The exchange continued:“Why give him a title?”“As a figurehead.”And then, from the union steward: “This is a cynical,
commercial decision we have to make: we’ve just got tobite the bullet.”
One member of the orchestra committee expressedsome very clear views about the prospective appoint-ment:
“I feel quite strongly: when I was last on the players’committee, there was a lot of talk about not employinghim at all—he’s such a poor conductor—and then itturned out he was so popular, could bring in the money,but we kept pushing that it should be limited. Now theyseem want to give him all these things. That has to beseen as purely financial, and they want us to condone it.I just don’t feel we can, but there is a strong feeling,because he’s so popular. So I feel that’s a mistake.”
He went on: “I don’t think [the executive director]realizes how damaging it is to have him for three weeks.The orchestra’s playing appallingly. I think we were al-most too successful last time in curtailing him: we onlyhave him for a couple of days a week, go out of town tomake some money, and that makes people think he’salright.”
And even appealing to the judgment of the principalconductor, he continued: “. . . I don’t see how [the prin-cipal conductor] could think that appointment would be
2007 83Maitlis and Lawrence
okay—how it could be a musical decision.” (orchestracommittee member)
At this time some rumors were going around that theconductor had already been offered a position. However,when the players’ chairman asked the executive directorabout a title when he came to address the full orchestra inDecember, the executive director replied dryly:
“I haven’t found one yet” (CEO to full orchestra meet-ing)
A player said: “What about ‘not-coming-here-again’?”To the player’s evident surprise, the executive director
replied: “That’s not the prevailing view of the orchestra.”In the senior management away-day several months
later, the issue was again raised. The team was discussingpossible candidates for the “jigsaw” of guest conductors,and whether conductor X should be given a title. Onemanager said, “He has such a bad name in the business—don’t give him a title” (senior management away-day,July)
Another added vehemently, “If you give [conductor X]a title, we’ll be laughed out of court.”
Sally Maitlis ([email protected]) is an assistantprofessor of organizational behavior at the Sauder Schoolof Business, University of British Columbia. She receivedher Ph.D. from the University of Sheffield. Her researchinterests include the social and political aspects of or-ganizational sensemaking and decision-making pro-cesses, and narrative and discursive approaches to thestudy of emotion in organizations.
Thomas B. Lawrence ([email protected]) is the Wey-erhaeuser Professor of Change Management and the di-rector of the CMA Centre for Strategic Change and Per-formance Measurement at Simon Fraser University inVancouver. He received his Ph.D. in organizational anal-ysis from the University of Alberta. His research focuseson the dynamics of power, change, and institutions inorganizations and organizational fields.
84 FebruaryAcademy of Management Journal