Toward an Organizational Perspective on

Qualitative Sociology, Vol. 22, No. 1, 1999

Toward an Organizational Perspective on Identifying and Managing Formal Gatekeepers Calvin Morrill, David B. Buller, Mary Klein Buller, and Linda L. Larkey

In this article, we present an institutionalist organizational perspective on formal gatekeeping, arguing along two fronts: First, identifying gatekeepers provides useful analytic devices for learning about the vocabularies of structure in an organization. Second, successfully managing gatekeepers requires that one understands the vocabularies of structure in use in an organization. We ground our perspective in field experiences gleaned from eleven public sector organizations who participated in our health promotion and research program, the Arizona 5 a Day Project. Field data and insights from institutional organizational and decision-making theory frame a matrix of organizational types and formal gatekeeping. We conclude by linking our perspective with analytic perspectives on organizational charters and organizational change. KEY WORDS: qualitative methods; access; gatekeeping; organizations.

Gaining access to field settings occupies venerable territory in discussions about qualitative methods. Classical anthropological field work generated several narratives about gaining access, including Boas' s frustrating attempts to negotiate entree to the Kwakiutl (Rohner 1966), Malinowski's (1967) three-year efforts to gain access beyond the fringes of Trobriand society, and Evans-Pritchard's (1940) access difficulties with the Nuer. Access experiences commonly appear in classical sociological ethnography, as well. Whyte's (1955) description of his entree into youth groups in Boston's Calvin Morrill and Linda L. Larkey are affiliated with University of Arizona, Tucson, Arizona 85721. David Buller is affiliated with AMC Cancer Research Center, Denver, Colorado 80214. Mary Klein Buller is affiliated with Partners for Health Systems, Inc., Denver, Colorado 80214. Direct correspondence to Calvin Morrill, Department of Sociology, University of Arizona, Tucson, AZ 85721.

51 C 1999 Human Sciences Press, Inc.

52

Morrill, Buller, Buller, and Larkey

North End, Liebow's (1967) recounting of his fortuitous access to streetcorner life, and Blau's (1955) initially unsuccessful attempts to gain entree to a federal employment agency all provide poignant examples. Scores of contemporary ethnographic works and virtually every general text on qualitative field methods also contain discussions of access (e.g., Johnson 1975; Jorgensen 1989; Lincoln and Guba 1985; Lofland and Lofland 1995; Wax 1971). Central in these treatments often is the identification and management of "gatekeepers" who patrol the boundaries of formal and private settings. Whyte, for example, encountered numerous gatekeepers in the groups he sought to study. Many of them prevented his entree until he befriended one—"Doc"—who opened up the world of the "corner boys" to Whyte's ethnographic gaze. Nowhere is gatekeeping more crucial than in formal organizations, because key personnel can legitimately grant or withhold permission for access (Hammersley and Atkinson 1995: 63). In contemporary organizational society (Perrow 1991), field sites increasingly exist within formal organizations. Managing organizational gatekeeping, therefore, will assume greater importance for field workers than ever before. Despite the growing importance of organizational gatekeeping, ethnographers rarely consider it in the explicit context of organizational analysis. In this paper, we focus on a crucial aspect of the access process to organizations: the identification and management of those managers and other personnel who have formal authority to grant access to a field site. Formal gatekeeping does not capture all forms of gatekeeping that organizational ethnographers can encounter, but it is often the first set of barriers that a fieldworker must overcome before negotiating what could be called "everyday access" to informal groups and individuals within organizations. Our theoretical perspective on gatekeeping draws on ideas from institutionalist decision making (March 1988) and conceptions of organizations as "vocabularies of structure" that enable members to construct, make sense of, and carry out activities in their organization (Meyer and Rowan 1977). We specifically argue along two interrelated fronts: First, identifying gatekeepers provides useful analytic devices for learning about the vocabularies of structure in an organization. Second, successfully managing gatekeepers requires that one understands the vocabularies of structure in use in an organization. We empirically ground our perspective in experiences from our health promotion and field research project, which we dub the Arizona 5 a Day Project. In the following pages, we first highlight formal gatekeeping as both a practical and analytic problem. We then briefly describe the Arizona 5 a Day Project, discuss our initial forays into the field as they relate to understanding organizational vocabularies of structure and outline the quali-

Organizational Gatekeeping

53

tative techniques we used to analyze our experiences. We next discuss how two key components of vocabularies of structure—authority structures and decision-making routines—influence gatekeeping. The paper concludes by linking our institutionalist organizational perspective on gatekeeping with the study of "organizational charters" and discussing gatekeeping in organizations undergoing radical change.

GATEKEEPING AS A PRACTICAL AND ANALYTIC PROBLEM

Ethnographers often regard gatekeeping, like other aspects of the access process, as a practical problem one must overcome early on in a field setting. In one sense, identifying relevant gatekeepers and dealing with them are interpersonal matters. To understand who organizations deploy to police their boundaries and how to overcome these barriers requires that we draw on interpersonal acumen that we all use on a daily basis to navigate through organizations and other contexts. Yet, identifying and managing gatekeepers is not only a practical interpersonal or logistical matter; for to do so effectively involves asking analytic questions about the organization under study. The most important of these questions involves the organizational location of relevant gatekeepers. In some organizations, this question has a straightforward answer: The most relevant gatekeeper resides at the top of the organization. In other organizations, the answer could lead to a specific department or division in which the fieldwork will occur. In still other organizations, identifying relevant gatekeepers is even more difficult. Gouldner (1954: 255-256) underscored this problem in his research on the Oscar Center gypsum plant. His research team: . . . made a "double-entiy" into the plant, coming in almost simultaneously by way of the Company and the Union. But it soon became obvious that we had made a mistake, and that the problem had not been to make a double-entry, but a triple-entry; for we had left out, and failed to make independent contact with a distinct group—the management of that particular plant. In a casual way, we had assumed that main office management also spoke for the local plant management and this, as a moment's reflection might have told us, was not the case. In consequence our relations with local management were never as good as they were with the workers or the main office management.

If we look beyond Gouldner's mistakes regarding the location and importance of relevant gatekeepers, we see that his problems really began when he failed to think analytically about how his project related to the plant's organizational structure. (This error is particularly ironic given that his study focused on the plant's bureaucratic structures.) He also neglected

54


to consider the temporal aspects of organizational gatekeeping; that is, that identifying and managing organizational gatekeepers can be a long-term process rather than a one-shot accomplishment with a single set of managers at the top. What does it mean to think analytically—especially from an organizational perspective—about formal gatekeeping? What general light does formal gatekeeping shed on understanding organizations? STUDY CONTEXT: THE ARIZONA 5 A DAY PROJECT

The Arizona 5 a Day Project provided us with a comparative context to begin formulating answers for the questions posed above. The Project is a multiyear, triangulated field study (1994-1997) that belongs to a growing number of federally funded programs engaged in both health promotion and research (Havas et al., 1994). The Project derives its name from the national 5 a Day For Better Health Program that includes the consumption of "5 fruits and vegetables a day" per capita as a goal in a "healthy lifestyle." Our Project focused on evaluating the processes and effectiveness of two techniques intended to facilitate dietary and life style changes among low-income public workers. One technique used work site mass media only. A second technique used work site mass media and health peer educators who spread the word about "5 a day" within informal groups of workers. We initially contacted eleven Arizona public sector organizations to participate in the Project. Ten eventually committed to it. The ten participating organizations represent a variety of services and activities, including four large municipal and county governments, two large public school districts, two community colleges, and two public universities. They also vary in their prior commitment and resources for work site health promotion activities. When we contacted them, seven organizations operated comprehensive employee "wellness programs." These programs included components on general health promotion and education, physical fitness, disease prevention screening and support, and nutrition intervention and planning. Within the ten employers, nearly five hundred labor and trades employees (the vast majority of which are Latino and Anglo males) organized in forty departments participated in the three-year peer education/research portion of the Project. The mass media campaign reached nearly all four thousand employees in our target population. The Project combined quantitative and qualitative research components. The quantitative components comprised extensive structured surveys conducted at four time periods during each year of the research. The qualitative components captured processual aspects of on-going and health peer activities between and during survey periods. Specifically, our qualitative


55

activities consisted of collaborative observational research with health peers (conducted by the authors and a large team of additional fieldworkers), focus-group interviews of workers and managers before and during the research, on-going ethnographic interviews with health peer educators, and team field research by the authors among work groups and managers prior to, at the outset of, and during the research. Our present analysis draws from this last data source because much of it involved reconnaissance into the organizations in two-, three-, and four-person teams in order to identify and manage relevant organizational gatekeepers. It is also in this context that we began to formulate an organizational perspective on gatekeeping. FROM "ALPHA-WOLF" MANAGERS TO SITUATED ACTION AND VOCABULARIES OF STRUCTURE

When we first approached the problem of identifying organizational gatekeepers for the Arizona 5 a Day Project, we committed the same mistake that Gouldner did in studying the gypsum plant and mine. We carried a naive, tacit vision of top managers as what we called "alpha-wolf" decision makers who could decide whether or not to guarantee access to all parts of their organizations. We initially devoted a great deal of time adapting to managers' individuallevel idiosyncrasies, tastes, and preferences. In this way, we repeated the "lessons" of folk theories that instruct lawyers preparing for courtroom presentations to learn all they can about how a particular judge thinks (O'Barr 1982) or folk theories of selling in which salespersons are instructed to adapt sales strategies to customer idiosyncrasies (Prus 1989). In some organizations, we found a few top managers who closely approximated our alpha-wolf vision, sending their orders of access down the chain of command to be obeyed by subordinates. In other organizations, top managers appeared as shrewd, back room bargainers in an endless series of negotiations with colleagues from multiple levels of the organization. The more we learned, the more we shifted away from managerial idiosyncrasies to the cultural and structural dimensions of each organization. This shift led us to resonant analytic traditions in contemporary institutionalist organizational theory, related particularly to decision making. Institutionalist researchers argue that managerial decisions arise less from a "logic of consequentiality" than from a "logic of appropriateness" grounded in taken-for-granted organizational assumptions that normatively influence decision-making routines and substantive outcomes (Levitt and March 1988: 320). Decision-making routines manifest themselves in the way decisions are typically made and implemented in an

56


organization; for example, whether courses of action emerge from political coalitions, are generated in highly centralized authority structures, or have powerful "ripple effects" through an entire organization via communication networks. From this perspective, individual decision makers typically engage in decision making by invoking contextually appropriate interpretive schemata and scripts for action that ". . . construct [an] action's developing purpose and intelligibility" (Suchman 1987: 3). Chief among the sources of such schemata and scripts are vocabularies of structure that supply legitimating accounts for formal authority, decision making, and informal social relations in organizations. All organizational members' activities are thus situated in overlapping, sometimes conflicting, vocabularies of structure. In some instances vocabularies of structure can have highly localized meanings; in other instances, legitimating accounts emerge from wider institutional and cultural myths about organizational strategies and structures in which localized vocabularies of structure are embedded (Powell and DiMaggio 1991). The decision of whether an outsider can enter an organization, then, offers an occasion to collect accounts from those individuals in authority about the structure and functioning of an organization—in one sense, to access the vocabularies of structure in operation in an organization. In the next section, we outline the procedures we used to organize our realizations into a more systematic approach on organizational gatekeeping, drawing on the conceptual perspective outlined above.

PROCEDURES Our experiences of identifying managing gatekeepers derive from fieldwork conducted in 1992 and 1993 prior to the Project's main study. We draw on two primary sources: (1) in-depth interviews with top executives (n = 4), human resource managers (n = 7), health promotion directors (n = 7), and food service managers (« = 5); and (2) multiple site observations in each organization. Interviews lasted from three-quarters of an hour to well over an hour. Interviews with top managers and human resources managers concentrated on organizational support for employee wellness and cancer prevention activities. Interviews with the health promotion directors emphasized on-going health promotion programs in their organization. Interviews with food service managers focused on food preparation methods and wellness activities targeted for cafeterias. During each interview we presented information about Arizona 5 a Day and asked a number of ques-


57

tions about the manager being interviewed, most importantly how and from whom we needed to obtain access and their expectations about the overall reception of the 5 a Day Project in their organization. Our site visits consisted of observations at the physical plants of each organization and various presentations about the Project to organizational gatekeepers. (We describe these presentations in subsequent sections.) These visits lasted from two to nearly entire work shifts. For eighteen months, we became regular visitors to each of our organizations. On some visits, managers and supervisors escorted us through the premises as we engaged them in conversational interviewing. On other site visits, personnel allowed us to "hang out" on the premises by ourselves. During these visits, we occupied ourselves by eating lunch ("lunch" comes at different times during different work shifts), "doing paperwork" in lunch or break rooms, and striking up casual conversations with workers and supervisors during breaks and on- going activities. Through out this early data collection, our "team approach" placed researchers in multiple combinations at different research sites. This maximized the perspectives we gleaned from the sites, which in turn facilitated our systematically formulating the variable nature of gatekeeping across the organizations (see generally Olesen et al., 1993; Snow, Benford, and Anderson 1986). As we contacted each of the eleven organizations to which access was sought, we initially recorded information in several files of field notes about (1) how we contacted each manager (e.g., whether by letter, by phone call, but a face-to-face to visit), (2) who we contacted, (3) the content of our messages to each manager, and (4) the effects of our strategies (e.g., the receptions of our messages, whether we needed to contact other managers in addition to the ones already contacted), and (5) intra-organizational characteristics, including how decisions are made in the organization, its authority structure, and communication networks. During and subsequent to gaining access to the ten participating organizations, we began to reconstruct our field notes and develop a coding scheme for our data. We initially "open" coded (Strauss 1987) our field notes to capture key aspects of our access experiences and activities relevant to gatekeeping: (1) types of gatekeepers, (2) organizational locations of gatekeepers, (3) external factors affecting gatekeeping, (4) internal factors affecting gatekeeping. As our coding became more focused (Lofland and Lofland 1995: 192-193), we elaborated these categories and linked them to aspects of organizational vocabularies of structure for formal authority and decision-making routines.

58


A FIRST CUT ON IDENTIFYING AND MANAGING ORGANIZATIONAL GATEKEEPERS Our access process began with the development of "initial contacts" in each organization with whom we had social ties from previous projects, who we believed had insider knowledge of on-going wellness and health promotion activities in their organization, and whose experience we believed could shed light on gatekeeping activities. In each organization our initial contact either occupied an executive or a director of health promotion and/or human resources roles. Our search for gatekeepers began with the simple question, "Whose approval do we need to obtain in order to conduct our research in this organization?" As straight forward as this question seems, it stimulated increasingly complex and analytically interesting answers the further we navigated into each organization. A first cut of answers to our question led to an early set of distinctions. The first consisted of whether a formal gatekeeper was an individual or group. In some organizations supervisory councils and coalitions (including elected officials in the counties and cities) decided whether health promotion programs should be adopted or continued, and thus could grant or prevent access. In other organizations top executives made these decisions alone. This distinction also suggested another dimension: the organizational locations of gatekeeping activities. Human resource managers typically held decision power over medical components of each health promotion program, such as cancer and other disease screenings. Human resource managers also decided which of these procedures would be covered by employee health insurance and what data would be used to evaluate program success. Health promotion directors (often called "wellness coordinators") primarily participated in decisions about the actual design of health promotion activities. Finally, food service managers made decisions about food selection, except in the county governments and public universities, which used private vendors who negotiated with a variety of organizational managers about food selection. Our initial observations and interviews thus revealed formal gatekeeping as a mosaic of activities across the organizations. Another aspect of gatekeeping that emerged early in our fieldwork involved the development of access rhetorics that resonated with the accounts gatekeepers offered for activities relevant to our proposed research. As work site health promotion programs have become an increasingly taken-for-granted part of American organizations, different institutionalized accounts have emerged to legitimate their adoption and functions within public sector organizational environments (Fielding and Piserchia 1989; Worksite Health Promotion Activities Advisory Board 1992). One


59

such account focuses on the benefits such programs hold for organizational efficiency or utility (Patton 1991). In this account, health promotion programs can positively influence employee productivity by creating healthier work forces less prone to physical and psychological maladies. Another account, couched in the professional language of human resource and health promotion managers, focuses on the intrinsic benefits of health promotion for employees and the development of "strong" organizational cultures in which a "healthy lifestyle" is a component (Wolfe, Slack, and Rose-Hearn 1993). A top manager whose organization had an on-going health promotion program illustrates the "efficiency" account: He "believed" that health promotion programs could "reduce health care costs, maybe even reduce absenteeism, and lower some of the potential for health-related liability without taking too much employee time or organizational resources." By contrast, a human resources manager believed that health promotion was a "cornerstone of a strong organizational culture" that oriented employees toward "healthy lifestyles." This is not to say that such managers found efficiency arguments unimportant, but they tended to emphasize, in the phrase of one manager, ". . . the human benefit of health promotion." These differences became more acute when we talked with various managers about who they believed bore primary responsibility for maintaining good health—individual employees or their employers. Top managers tended to believe that individual employees, rather than employers, should bear primary responsibility for maintaining good health. Listen to this top manager: "Employees have a responsibility to keep themselves healthy. We [the organization] can only do so much in this regard." By contrast, human resources and health promotion managers believed that employers bore primary responsibility for creating value systems and opportunities that facilitated good health practices. One health promotion manager noted: "Organizations can't look the other way any longer with regards to health. A strong organizational culture includes the responsibility for promoting healthy life styles that will benefit every single employee." These accounts became the basis for developing access rhetorics directed at different managerial roles. As one of the authors who constructed much of the rhetoric used to gain access notes: We wanted to convince wellness coordinators [health promotion directors] that 5 a Day would complement the 'healthy lifestyles' culture they [the coordinators] were trying to develop. We also used the nationally developed literature on 5 a Day and some of our prototype local literature to illustrate what the materials and messages we were going to use in the project would look like. When we did these presentations to top managers, we emphasized how employees could take responsibility for their health through peer education. Peer education could help employees monitor their own progress toward a healthier lifestyle. We also

60


emphasized the low cost of 5 a Day to the organization in terms of worksite participation by employees and coordination of 5 a Day activities by managers and supervisors. Whenever we did group presentations, we first figured out who was going to be there and adjusted the presentation accordingly.

A SECOND CUT: DECISION-MAKING ROUTINES AND AUTHORITY STRUCTURES The longer we worked in the field, the more we realized the close linkages between gatekeeping, and managerial accounts and behavioral manifestations of vocabularies of structure. We focused especially on two components of structure: decision-making routines and authority structures. Indeed, we folded our early distinctions between individual and collective gatekeepers, as well as access rhetorics, into an overall strategy focused on identifying the relationship between gatekeeping and vocabularies of structure for each organization. Although large organizations can contain enormous cultural (Van Maanen and Barley 1985) and structural variation (Morrill 1995), we found it useful to characterize the organizations according to their predominant decision-making routines and authority structures. We conceived decision-making routines as falling on a continuum from "rationalized" to "politicized." In the ideal-typical rationalized organization, bureaucratic rules govern the means and ends of organizational action and legitimate decision making. Chains of command and spheres of responsibility in such organizations tend to be spelled out and justified by written procedures, and graphically displayed in extensive tables of organization (e.g., Weber 1946). In the ideal-typical politicized organization, decisions are political acts carried out by managers acting as brokers within and between coalitions. Shifts in policy formation tend to be a function of shifts in coalitions; the most effective managers in politicized decision-making routines are those who can build and maintain coalitions (March 1962). Our authority structure categories consisted of a continuum from "centralized" to "decentralized." These categories capture a core element of organizations: the degree to which primary authority over organizational resources is diffuse or concentrated in the hands of a few people (Vancil and Buddrus 1979). An organization is centralized to the degree that it locates authority and control over organizational resources (e.g., finances, personnel, information, and crucially, access and commitment) at the top of a managerial hierarchy. In decentralized organizations, authority resides and is legitimated in multiple places; in essence, no single set of players has monopoly control over the organization. These dimensions form a matrix, each cell of which represents an ideal-typical organizational type. Each ideal-type is discussed in the subsections below and illustrated with nor-


61

Table L Authority Structures, Decision-Making Routines, and Primary Gatekeepers AUTHORITY STRUCTURES DECISIONMAKING ROUTINES

Centralized Organizational Type:

Monocracy Primary Gatekeeper: Rationalized

Politicized

individual top manager

Decentralized Organizational Type: Heterocracy Primary Gatekeepers: multiple individual managers

Illustration: Solar County

Illustrations: Solar City Lunar City Solar University

Organizational Type: Coalarchy

Organizational Type:

Primary Gatekeeper: "ruling coalition"

Primary Gatekeepers: multiple coalitions

Illustrations: Solar School District Solar Community College

Polyarchy

Illustrations: Lunar County Lunar Community College Lunar University Lunar School District Solar Government Organization

malized descriptions and managerial accounts drawn from each organization. Table I summarizes the distinctions between each organizational type.

Monocracy A monocracy contains centralized authority structures and rationalized decision-making processes that resonate through well-defined chains of command that issue from top management down through the entire organization. Almost militaristic in tone, official, written rules and policies legitimate top management's decisions. Middle managers in monocracies enjoy few channels of upward influence. Moreover, there are few expectations or mechanisms of employee input in these organizations. Gatekeeping, like most other decision making processes, resides at the top of the organization and is closely related to everyday gatekeeping. In other words, formal authority casts a long shadow over the informal crevices of the organization. Solar County illustrates a monocracy in action.

62


In Solar County, we initially contacted the county manager via a formal letter. Several formal one-on-one meetings followed thereafter before approval for access was gained. The substantive content of our meetings with the county manager covered an array of topics, emphasizing the low cost of the Project to the organization. Once the county manager approved access for the Project, he claimed that "[the Project] will have carte blanche down the line." A middle-level manager in Solar County who we later contacted echoed these statements: "People down the line in our organization find it difficult to say 'no' when the [top executive] or their immediate boss sends down a directive. Everything pretty much follows the chain-of-command here." Indeed, these statements proved largely prophetic as we held purely informational meetings with middle- and lower-level managers, as well as employees, who rarely if at all questioned the access directives from the top. In these meetings, we focused on explaining how managers and employees could comply with the Project's requirements and schedule the baseline interview and health promotion activities. At the lower levels, moreover, managers justified their "decisions" to grant us access in terms of what had been previously decided at the top of the organization, often referring to written memos from the top and official protocols.

Coalarchy Coalarchic organizations exhibit both centralized authority structures and politicized decision-making routines. They contain well-defined, topdown chains of command as in monocracies. Unlike monocracies, however, coalarchies are directed by, in the argot of managers in one of our organizations, a "ruling coalition." Ruling coalitions centralize legitimate authority and typically have extensive, vertical patronage networks. In public organizations, such coalitions can involve appointed managers and elected officials and can have both "back" and "front stage fronts" (Goffman 1959). On the back stage, managers, in the words of several of our informants, "cut deals." On the front stage, deals are "stamped with the official seal of approval." Persuasion and politics are the rule within the ruling coalition, but obedience to directives from the coalition characterize managerial norms down the hierarchy. Understanding the boundaries of the ruling coalition—whose in and whose out—emerged as a critical factor for managing gatekeepers in a coalarchy. As we made our initial contacts and search for gatekeepers in "Solar Community College," we learned that it contained a ruling coalition comprised of several appointed top managers and members of the board of directors. Through our initial contact, we learned who needed to be at for-


63

mal presentations of the Project. We then invited members of the ruling coalition to the presentation through written invitation and cross-checked the invitation list with our initial contact lest we unintentionally convey partisanship for or against some members of the coalition. Our presentation proved successful and we gained initial access to the organization, which in turn enabled us to make informational presentations to lower-level managers about how they could cooperate with us as we implemented the Project. We followed the chain of command in contacting other managers and making further presentations. In formal invitations to managers and in the presentations themselves, we carefully referenced members of the ruling coalition to enhance our credibility. Solar School District provides an example of a coalarchy in which the ruling coalition involved both the district top manager, appointed top-managers, upper-middle managers, and elected officials on the school board. It also provides examples of the back and front stage elements of formal gatekeeping in a coalarchy. We initially contacted Solar School District's top manager, who persuaded another member of the ruling coalition, the director of community relations, to handle what the latter called, "our case." In turn, the director of community relations persuaded other members of the coalition to attend some initial presentations about the Project. Our presentations again proved successful, but access could not be granted until the elected portion of the coalition, the school board, legitimated the Project. As the top manager explained, "It's a done deal. You've got the access, but I can't officially authorize access for the Project without going public and obtaining approval from the school board." The top executive himself presented the Project to the board at a monthly public meeting (with the director of public relations, the food service manager, and a few other members of the ruling coalition in attendance). Members of the research team also attended the meeting to answer questions in a formal question-and-answer session after the presentation. Once we secured school board approval, the director of community relations provided us with lists of division and department heads who supervised eligible employees. Lower-level managers complied with the directives from the top without extensive discussions on the merits of the Project. Heterocracy

A heterocractic organization combines a decentralized authority structure with rationalized decision-making routines. Such organizations depart from the top-down authority structures of the previous two organizational types in that divisions and departments enjoy greater formal autonomy in

64

Merrill, Buller, Buller, and Larkey

the pursuit of their goals. As one Solar University manager put it: "They're rules aplenty here. They're just not coming down from the top. Every department's got their own procedures and they don't always match up with each other." Gatekeepers proliferate in such organizations and enjoy enormous latitude to grant or deny access, even when their decisions may conflict with top management. Moreover, formal rules and procedures proliferate in heterocracies, but emanate from and apply to particular divisions and departments. The heterocracy is thus rationalized within organizational units, but is "loosely coupled" across units (Weick 1976). Gatekeeping likewise follows this pattern. Formal gatekeeping is key to access to particular units, but official access and justifications for it do not typically transfer between units. "Solar City," "Lunar City," "Lunar Community College," and "Solar University" illustrate these tendencies. In Lunar City, for example, access and commitment for the Arizona 5 a Day Project required an extensive permission process in a lengthy series of meetings at multiple levels and departments in the organization. The official rules, protocols, and goals to which we adapted varied significantly across units. We met twice each with the top personnel manager, benefits manager, and employee assistance program manager to discuss Project components, costs in terms of employee time, financial benefits for the organization, and potential outcomes of the program. Each of these managers in turn came armed with a set of official (written) criteria generated from their departments that they wanted the 5 a Day Project to meet. As a result, we conducted separate negotiations for access with each department manager. These meetings effectively completed the formal approval process at the top of the authority structure, but approval had to be gained at each level in the hierarchy. With full support from top management, the employee assistance manager put us in touch with personnel officers for each department with eligible employees. At that point, the employee assistance manager receded from the approval process. We held numerous meetings with lower-level personnel officers in which we described the 5 a Day Project. Here again, we experienced variation in access decisions (some let us in; others did not) and in official rules and procedures relevant to employee release time for on-sight participation in the Project and managerial authority. The personnel and benefits managers in turn introduced us to the middle managers and supervisors in their departments who also provided localized criteria and rules for their units. These criteria and rules became relevant for each set of access negotiations we pursued. As illustrated by our description above, the approval of top management can become somewhat superfluous for gaining access to various components of a heterocracy. Such was the case in Solar City. In that organization, our initial contact, the health promotion director, helped us

6s


with approval process for managers and executives above her position. We had little participation in this process and never met with top managers. Instead, the health promotion director arranged for us to meet with upper administrators from the personnel department who in turn helped us identify eligible departments for the Project. The health promotion director also recruited relevant department heads who had the formal authority to decide whether participation in the Project was in the best interest of their own employees. Moreover, while upper management approval (which was eventually obtained after most of our negotiations were completed at the lower levels) generally did not hurt our requests for participation at the lower levels, it did not guarantee it and nor did many managers appear concerned about what upper management had or had not approved. As the health promotion director noted, "Be careful with being too strong in telling people that so-and-so approved this [project] at the top. Their [top management] word isn't always golden at the lower levels. Everybody here feels the rules in their unit are the most important, most relevant to their situation." We therefore had to gain access to each department through individual negotiations with department managers and supervisors according to the rules and procedures they adhered to in their units. Polyarchy

In many ways, polyarchies present the greatest challenge for identifying and managing organizational gatekeepers. Unlike centralized organizations, they do not have single seats of authority or power. Unlike rationalized organizations, there seem to be few if any formal rules that managers and employees actually apply with any consistency to decision making. Instead, polyarchies appear as cauldrons of organizational politics, back room deal making, and informal power bases based in coalitions and personal relationships. Accounts about structure and operations in a polyarchy focus on influence and power, as one manager in polyarchy noted: "We have rules, sure, but it all comes down to who has power, who's with you and who's against you, and who you know in this organization." Unlike coalarchies, then, power does not reside in a single coalition, but rather in multiple coalitions who are constantly vying for influence and power. Moreover, formal gatekeeping often has little to do with everyday gatekeeping in the sense that formal authority typically does not carry the weight that it does in centralized or rationalized organizations. Access to polyarchies therefore requires multiple presentations to multiple groups up and down the organization. Our initial forays into polyarchies often met with failure because we had little advance warning regarding the complexity of informal net-

66


works in these organizations. Even when we "cased" some of the polyarchies in the study prior to contacting them, we still experienced difficulty assimilating the information necessary to identify who was who until we made numerous entries into the organization. The shifting political landscape in a polyarchy also means that identifying gatekeepers is a constant concern; who authorized access last month may not have the power to ensure access this month. As such, developing trusted "sponsors" or "brokers" who offer reliable knowledge of on-going political moves becomes more critical than in other organizations. Four organizations fit the polyarchic profile: "Lunar County," "Lunar University," "Lunar School District," "Lunar Community College," and "Solar Government Organization." Our failure to gain access to Solar Government underscores the challenges presented by a polyarchic organization. This organization comprised seven powerful, independent departments, each as large or larger than any of the other organizations in the Project. We initially obtained tentative approval for access from the highest elected official within Solar Government and from one of the highest ranking appointed managers in the organization (prior to our being fully conscious of the polyarchic nature of the organization). As we intensified our efforts with these and other gatekeepers at the top and middle of the organization, we became immersed in a highly complex political system that linked elected and appointed officials, political parties, and organizational coalitions that spanned within and outside the organization. Our first difficulty arose as we attempted to ascertain simply who we should negotiate with in various departments and whether negotiating with one manager or coalition would be cast in a negative light by another manager or managerial coalition. We relied heavily on our initial contact, the director of health promotions (call him Art), to provide us with informal maps of political coalitions and updates on organizational politics. Art professed interest and commitment to the Project and offered to "pave the way for the [Project]." Even so, some of our initial visits to Solar Government were met with suspicious queries from managers with whom we had previously spoken, as illustrated by this manager: "Do you really plan on working with [a previous contact]? He may not deliver the goods." As we stumbled around the organization trying to gain access to various units (and not become pawns in enmities between various factions), our approval from one coalition or manager sometimes prevented access to units hostile with one to which we already enjoyed access. Such problems became apparent, for example, when one manager pointedly observed: "You're working with [Jones; the other unit's manager]? He's a liar and he's got a bunch of outlaws around him." To our horror, we discovered this manager was aligned in a coalition against Jones and distrusted anyone who would work with Jones. After some time, we began to realize that the


67

time and energy we devoted to this one organization outweighed all of that put into access for all of the other organizations to which we sought access. As a result, we relied more heavily on Art to broker the Project to the managers of the units to which we attempted to gain access. This strategy failed as well. In the end, Art admitted that he "might not have the kinds of contacts you're looking for to support [the Project]." When asked who might know such contacts, Art claimed they "might not exist at this time." As a result, Art suggested that we cease our access efforts. A successful brokerage example occurred in Lunar Community College. Our initial contact person at Lunar Community College (Carol, the director of health promotions) told us that she was ". . . quite committed to the Project. It's what we should be doing here and I know just the people who will support it." The last part of her statement referred to her network of contacts among various coalitions through out the organization. Carol drew on these network ties to negotiate all approval required without our direct input (beyond supplying initial information materials) or participation. Thus, the differential success of our brokers at Solar Government and Lunar City College appeared to be less a result of each individual's official position or commitment to the Project, than the volatility of competition and conflict between organizational coalitions and the brokers' network positions inside their respective organizations.

SUMMARY AND IMPLICATIONS

We began our analysis by identifying some fundamental dimensions of formal gatekeeping: whether it resides in individual or collective actors, whether it operates at the top or multiple levels of an organization, and how external factors, such as institutionalized accounts about organizational practices relevant to the research project in question, influence formal gatekeepers in different managerial roles. We then sharpened our organizational vision of formal gatekeeping by situating it in two, key components of organizational vocabularies of structure: formal authority structures and decision-making routines. In monocracies, gatekeeping is centralized and access approval by top managers essentially guarantees access to other levels of the organization. A coalarchy centralizes authority in a single political faction at the top where the principal gatekeepers reside. Decentralized organizations present additional complications because gatekeepers must be identified and managed at every organizational level. In such organizations formal and everyday gatekeeping do not enjoy the consonance they do in centralized organizations. Heterocracies contain loosely-coupled units with a plethora of local rules and policies. In a heterocracy, gatekeepers

68


abound, each armed with local procedures and protocols. Polyarchies present particular difficulties because managers view most decisions in relation to their power bases in the organization. Approval by top management in either heterocracies or polyarchies rarely leads to blanket access and can even lead to personnel distrusting researchers at the middle and lower levels of such organizations. Our analysis of gatekeeping resonates with Dingwall and Strong's (1985) attempts to reconcile organizational ethnography with contemporary institutional analysis through the study of organizational charters. According to Dingwall and Strong, an organizational charter is: . . . the concept to which organization members orient in their dealings with one another and with nonmembers to establish the limits of legitimizable action . . . . In some sense, a charter can be said to represent the constraints on a member's freedom of action that he or she experiences or depicts as exterior, objective, and given (1985: 217).

Charters, then, ground institutionalized constraints in localized sets of premises and rules about individual and organizational action. Gatekeeping, which by definition brings organizational nonmembers and members together, can provide a useful analytical device for beginning to form a concrete image of an organization's charter. Gatekeeping illuminates the various boundaries of an organization, the existence of particular organizational components, how legitimate authority is organized, and relevant informal communication networks cursing through and outside formal structures. Moreover, gatekeeping questions and responses can lead to a variety of useful kinds of data. Such data in the present context consisted of verbal accounts from managers for how we needed to gain access to each organization, how their organizations operated, and, why their organizations operated health promotion programs. We subsequently observed manifestations of these accounts in the behavioral processes by which we gained access, as well as in other organizational behaviors, such as how executives and other personnel managed the Arizona 5 a Day project during its implementation and four-year life span, organizational budget allocations, organizational restructurings, and resource procurement. Our perspective also underscores a distinction between the structural and substantive features of organizational charters. An organization's substantive charter could be said to refer to its official goals, purposes, and mandates. The substantive charters of both Lunar and Solar University, for example, encompass the broad mandates of higher education and scholarly research. Their structural charters—organizational structures, particularly "legitimizable" formal authority and decision-making routines-differ. Although we classified both as decentralized, Solar University exhibits a


69

marked tendency toward rationalized decision-making routines, while Lunar University is highly politicized. This difference, then, suggests that substantive and structural charters can be decoupled from one another, which is consistent with the decoupling of technical processes and organizational structures generally observed in public sector organizations (Mohr 1971). Such decoupling is particularly common in "institutional" sectors where conformity to social and cultural norms can take precedence over technical criteria (Meyer and Rowan 1977; Perrow 1986: 272). Can our approach to gatekeeping be extended to organizations undergoing radical change? The quick answer is a qualified yes, but let us elaborate. Radical organizational change involves the alteration of vocabularies of structure, particularly as they are instantiated in group- and individuallevel cognitive scripts and schemas (Bartunek and Reid 1992). Identifying and managing gatekeepers in organizations undergoing radical change presents practical challenges because local, legitimated accounts for authority and decision-making can change rapidly, thus invalidating the information gathered about the organization at any one time. At the same time, changes in gatekeepers and gatekeeping over the process of access negotiations can provide a chronicle of how substantive and structural organizational charters change. Although most of our organizations did not experience radical change during the study period, one of coalarchic organizations underwent a wholesale restructuring of the management structure, precipitated by a change in elected leadership and the resignation of several top managers, including our initial contact. The restructuring fractured the organization's ruling coalition, thus transforming the organization from a coalarchy to a polyarchy. In the aftermath of the change, we appealed to the organization's top executive and a few contacts who we knew from the former ruling coalition. These contacts did not have the power bases or network ties to help us and our access strategies foundered. We used our knowledge of the political factions in the organization (gained from our former contacts) to appeal to two elected officials (in a rival coalition to that of the top executive), emphasizing the low organizational costs of employee health promotion. We struck a series of bargains first with the two elected officials that covered which departments and divisions we could contact for access. That coalition then presented their bargain to the top executive's coalition as an "olive branch" for peace. This tapestry of bargains resulted in the organization participating in the Project. Once this occurred, we began the arduous process of identifying gatekeepers in individual departments. In conclusion, organizational gatekeeping remains a practical problem that every ethnographer must solve. We believe that addressing the issue of gatekeeping from an organizational perspective—particularly one

70


grounded in vocabularies of structure—sensitizes organizational ethnographers to a set of issues little discussed in qualitative methodology from an organizational perspective. More than this benefit, however, approaching organizational gatekeeping as an analytic device expands the methodological repertoire for ethnographically understanding contemporary organizations in at least three ways. First, it enables one to understand who patrols the boundaries of organizations. Second, it can provide a context in which to ask directly for managerial accounts of vocabularies of structure. Finally, it can lead to understanding the structural and substantive charters for organizations.

ACKNOWLEDGMENTS This paper is based on the Arizona 5 a Day Project (funded by a grant from the National Cancer Institute; #CA59726). Portions of this research were also supported by a planning grant from the American Cancer Society. The authors thank the management and employees of Arizona State University, City of Phoenix, City of Tucson, Maricopa Community College District, Maricopa County, Mesa Public Schools, Pima Community College, Pima County, Tucson Unified School District, and the University of Arizona for their participation. The authors also thank Ellen Richardson for her assistance. An earlier version of this paper was presented at the International Communication Association, Chicago, 1996. We thank three anonymous Qualitative Sociology reviewers for helpful comments on an earlier draft.

REFERENCES Bartunek, J. M., and Reid, R. D. (1992). The role of conflict in a second-order change attempt. In D. M. Kolb and J. M. Bartunek (Eds.), Hidden Conflict in Organizations: Uncovering Behind-The-Scenes Disputes. Newbury Park, CA: Sage. Blau, P. M. (1955). The Dynamics of Bureaucracy: A Study of Interpersonal Relationships in

Two Government Agencies. Chicago: University of Chicago Press. Cyert, R. M., and March, J. G. (1963). A Behavioral Theory of the Firm. Englewood Cliffs, NJ: Prentice-Hall. Dingwall, R., and Strong, P. M. (1985). The interactional study of organizations: a critique and reformulation. Urban Life, 14, 205-232. Douglas J. D. (1976). Investigative Social Research: Individual and Team Field Research. Beverly

Hills, CA: Sage. Evans Pritchard, E. E. (1940). The Nuer. Oxford, England: Clarendon Press. Fielding J., and Piserchia P. (1989). Frequency of worksite health promotion activities. American Journal of Health Promotion, 79, 16-20. Goffman, E. (1959). The Presentation of Self in Everyday Life. Garden City, NY: Doubleday. Gouldner, A. W. (1954). Patterns of Industrial Bureaucracy. NY: Free Press.


71

Hammersley, M., and Atkinson, P. (1995), Ethnography: Principles in Practice (second edition). London: Tavistock. Havas, S. Heimendinger, J., Reynolds, K., Baranowski, T., Nicklas, T. A., Bishop, D., Buller, D. B., Sorenson, G., Beresford, S. A. A., Cowan, A., and Damron, D. (1994). 5 a day for better health: a new research initiative. Journal of the American Dietetic Association, 94, 32-36. Johnson, J. M. (1975). Doing Field Research. NY: Free Press. Jorgensen, D. L. (1989). Participant Observation: A Methodology for Human Studies. Newbury Park, CA: Sage Publications. Levitt, B., and March, J. G. (1988). Organizational learning. Annual Review of Sociology, 14, 319-340. Liebow, E. (1967). Tatty's Comer A Study of Negro Streetcomer Men. Boston: Little, Brown. Lincoln, Y. A., and Guba E. G. (1985). Naturalistic Inquiry. Newbury Park, CA: Sage. Lofland, J., and Lofland, L. H. (1995) Analyzing Social Settings (3rd ed.). Belmont, CA: Wadsworth. Malinowski, B. (1967). A Diary in the Strict Sense of the Word. New York: Harcourt, Brace, and World. March, J. G. (1962). The business firm as a political coalition. Journal of Politics, 24, 662-678. March, J. G. (1988). Decisions and Organizations. Oxford, England: Basil Blackwell. Meyer, J. W., and Rowan B. (1977) Institutionalized organizations: formal structure as myth and ceremony. American Journal of Sociology, 83, 40-63. Meyer, J. W., and Scott, W. R. (1983). Organizational Environments: Ritual and Rationality. Beverly Hills, CA: Sage Publications. Mills, C.W. (1940). Situated actions and vocabularies of motive. American Sociological Review, 5, 439-448. Mohr, L. (1971). Organizational technology and organizational structure. Administrative Science Quarterly, 16, 444-459. Morrill, C. (1995). The Executive Way: Conflict Management in Corporations. Chicago: University of Chicago Press. O'Barr, W. M. (1982). Linguistic Evidence: Language, Power, and Strategy in the Courtroom. NY: Academic Press. Olesen, V, N. Droes, D. Hatton, N. Chico, and L. Schatzman. (1993). Analyzing together: recollections of a team approach. In R. G. Burgess and L Bryman (Eds.), Analysis of Qualitative Data. London: Routledge. Patton, J. (1991). Work-site health promotion: An economic model. Journal of Occupational Medicine, 33, 868-873. Perrow, C. (1986). Complex Organizations: A Critical Essay. NY: Random House. (third edition). Perrow, C. (1991). A society of organizations. Theory and Society, 20, 725-762. Prus, R. (1989). Making Sales: Influence as Interpersonal Accomplishment. Thousand Oaks: CA: Sage. Powell, W. W., and DiMaggio, P. J. (1991). The New Institutionalism in Organizational Analysis. Chicago: University of Chicago Press. Rohner, R. P. (1967). Franz Boas: Ethnographer on the Northwest Coast. In J. Helm (Ed.), Pioneers of American Anthropology. Seattle: University of Washington Press. Snow, D. A., Benford, R., and Anderson, L A. (1986). Fieldwork roles and informational yield: A comparison of alternative settings and roles. Urban Life, 15, 377-408. Strauss, A L. (1987). Qualitative Analysis for Social Scientists. Cambridge, England: Cambridge University Press. Suchman, L. A. (1987). Plans and Situated Actions: The Problem of Human-Machine Communication. Cambridge, England: Cambridge University Press. Van Maanen, J., and S. R. Barley (1985). Cultural organization: Fragments of a Theory." In L R. Pondy, P. J. Frost, G. Morgan, and T. C. Dandridge, Organizational Symbolism. Newbury Park, CA: Sage. Vancil, R. F., and Buddrus, L E. (1979). Decentralization: Managerial Ambiguity by Design. Homewood, IL: Dow Jones-Irwin.

72


Wax, R. H. (1971). Doing Fieldwork: Advice and Warnings. Chicago: University of Chicago Press. Weber, M. (1946). Bureaucracy. In H. H. Gerth and C. W. Mills (Eds.), From Max Weber Essays in Sociology. NY: Oxford University Press. Whyte, W. H. (1955). Street Comer Society: The Social Structure of an Italian Slum. (2nd Edition). Chicago: University of Chicago Press. Weick, K. E. (1976). Educational systems as loosely-coupled systems. Administrative Science Quarterly, 21, 1-19. Wolfe, R. Slack, T., and Rose-Hearn, T. (1993). Factors influencing the adoption and maintenance of Canadian facility-based worksite health promotion programs. American Journal of Health Promotion, 7, 189-198. Worksite Health Promotion Activities Advisory Board. (1992). National Survey of Worksite Health Promotion Activities. Washington, DC: Government Printing Office.