College and Research Libraries The Organizational Misfits Patricia A. Suozzi and Sandra S. Kerbel Departmental libraries in large systems are often viewed as outsiders and as not fitting neatly into the organizational structure. Most of tlze writing about these units has focused on whether or not they should exist, rather than on their nature and value to the institutional mission. This article examines the nature of these libraries, their role in the organization, and the type of organizational structure that best enhances their value to the organization. proclamation has been issued. Libraries must evolve from being collection owners to be- coming information providers capable of serving as network channelers or coordinators. Librarians have been warned that their traditional organiza- tional structure will need to change if they are to survive and be effective in a future environment of remote users, electronic access, and rapid technological change. Furthermore, they are told that they must look for ways to promote connections be- tween databases, communication net- works, scholarly communication, and clients in order to meet what Dilys Morris has called the "greater need for access, interpretation, and brokering." 1 The search is on for a new structure or paradigm to effect this role transforma- tion. There is general agreement that "the more important truth about aca- demic libraries is that they are encum- bered by record systems and by financial and organizational structures which dis- courage innovation and make it difficult for them to manage uncertainty." 2 Yet an appropriate or effective role model for the transformation has not been found and validated. However, the authors believe that a model already exists and should be ex- amined. The model the authors are pro- posing is that of departmental libraries in a college or university. These libraries are already performing many of these future functions-specialized and tailored ~rv­ ices, intensive knowledge of clients, and document delivery. Further, many have operated in an environment where these services were developed without much support from the senior library adminis- tration. An examination of departmental libraries will provide insight into how li- brary systems can adopt or integrate the entrepreneurial and flexible mode of operation that characterizes many de- partmental or branch libraries and seems so necessary to the future growth and effectiveness of libraries in general. This idea is not as controversial as it may seem. In discussing the future struc- ture of libraries, scholars have actually been describing the operation of many departmental libraries. Some are sug- gesting that the new structure be orga- nized around groups with many self- contained tasks and that "the basic man- agement of the groups should be col- legial and participative."3 Others suggest that "one might also consider access in smaller research-oriented branch libraries, the staff in such branches often seem to understand intuitively the library's place Patricia A. Suozzi is the Director at Jennie King Mellon Library, Cat/zam College, Pittsburgh, Pennsylvania 15232-2814. Sandra S. Kerbel is Acting Head, Physical Science and Engineering Libraries, University of Pittsburgh, Pennsylvania 15261. 513 514 College & Research Libraries in the network of research information communication, when it is situated as one 'service station' among many in a department or school."4 David W. Lewis observes that "when access replaces ownership as a significant means of pro- viding scholars and students informa- tion, high-quality reference and con- sultation services will become part of expected institutional support. This will likely lead to an organizational structure based on client-centered work groups as suggested by Charles Martell."5 D. Kaye Gapen predicts that "library organizations will become evolutionary, nonhierarchi- cal, entrepreneurial, and horizontal."6 In spite of such claims, few librarians have analyzed the role departmental li- braries may play in shaping the library of the future. This dearth of analysis can be attributed to the fact that the role of departmental libraries in the library or- ganization is neither well defined nor well understood. As is too often the case, what is not understood is dismissed or conveniently ignored.7 In this case, ig- norance may well be terminal. Continu- ing to view organizational structure through a myopic haze will lead not to evolution but to extinction. This, how- ever, need not be the case. Given the future libraries are facing, librarians should consider whether or not organiza- tional units that exemplify these future paradigms and can act as the foundation upon which to build the "new library" already exist. The first step in this process, then, is to examin·~ and clarify the nature of departmental libraries and their role in present library organiza- tional structure. CHARACTERISTICS Departmental libraries developed from the specialized needs of specific clienteles. In fact, many of these libraries grew from departmental reading rooms and were formed independently of li- brary administrative plans. This phe- nomenon of ad hoc libraries is not specific to any discipline but does characterize many departmental librar- ies. As Edward G. Holley states, "the presence of numerous bootleg or 'broom November1992 closet' libraries on campus ought to in- dicate to somebody that most university library systems are failing to provide adequate service to segments of the uni- versity community."8 Departmental li- braries in a library system can be viewed as the structured o~tcome of unmet ser- vice needs. As a result, they tend to maintain a close working relationship with their originating departments. Such origins-outside of administrative channels and in close association with departments-can cause library adminis- trators to view departmental libraries as troublesome outsiders. And yet many de- partmental libraries would not be likely to exist unless they were fulfilling needs common to most academic institutions. Departmental libraries can be divided into two types: those libraries that are in buildings entirely separate from the cen- trallibrary, and those that are physically located in the central library building but that maintain separate collections. While this paper is mainly concerned with the first type, the second, or hybrid type, also has similar origins and shares many of the same characteristics as the separate departmental library. The chief difference between the two is that the hybrid does not support all of the oper- ational activities, such as circulation or facilities maintenance, that the separate library supports. Following are some characteristics that tend to identify departmental librar- ies regardless of type or discipline: • Readily identifia~le and vocal clientele. • Tightly focused goals and objectives. • Inter-relatedness of functions. • Holistic view of service. • High degree of collegiality and flexi- bility among staff. • Close physical proximity to primary user community. • Entrepreneurial management style, ne- cessitated by both physical and spiritual isolation from other library units. • Ability to develop and personalize service. • Identification by primary clientele and staff as part of that academic unit. Given these characteristics, many de- partmental libraries do not fit well into the current hierarchical organization of most library systems. The reporting lines for these libraries are usually vertical and unidirectional. Yet departmental li- braries encompass boundary-spanning functions. They are responsible for pub- lic services, collection development, and technical services, and such functions often overlap with those of units in a main or central library. As a result, they do not fit well into a hierarchical struc- ture, and as Barbara B. Moran states, "almost all academic libraries are struc- tured in a hierarchicalmanner." 9 There- fore, they are, in a sense, misfits. THE DEBATE Little in the literature addresses the role of departmental libraries in the knowledge-information process or iden- tifies their place in the organizational structure. Yet, departmental libraries are not being completely ignored. In fact, a large amount of writing has been done on whether departmental libraries should even exist. The problem with these discus- sions, though, is that the unit of analysis is not the departmental library's effective- ness, but the system's need for tight bureaucratic control. Thus the assump- tions on which these discussions are based are biased in that no consideration is given to the needs that promulgated these libraries in the first place or to the ability of these libraries to contribute to the effectiveness of the system. For example, J. Michael Bruno states that the main disadvantages with depart- mental organization are administrative: 1. Administrative control (coordina- tion, cooperation, and communica- tion) is difficult to achieve, 2. The cost of administering such branches, 3. The problems of access and secu- rity increase. 10 He dismisses their advantages within two sentences and concludes that "the disadvantages of having such small units as departmental libraries far out- weigh any of these advantages." 11 The conclusion of many others is similar: that departmental libraries may fill a need but that they are far too costly. Organizational Misfits 515 In 1987 the Journal of Academic Librari- anship presented a symposium entitled "Centralization or Decentralization of Library Collections." Thomas D. Watts presented a user's opposition to branch or departmental libraries. He listed five major concerns: 1. The growing interdependence of knowledge · 2. Tremendous inconvenience to the user 3. Isolation of collections 4. Expense 5. Communication between depart- ments.12 Implicitly, both Bruno and Watts ac- knowledge the users' desire for de- partmental libraries. Yet they see no formal role for them in a library's or institution's mission. Nevertheless, the assumptions and conclusions of centralization advocates have been questioned and, in a few cases, tested. In terms of administrative control, Anne Woodsworth concludes that "with the acceptance of common citizenship within a university, coordi- nation of policies, personnel practices, budgets, and planning can bring about a decentralized organization that is strong- er through its diversity." 13 More impor- tantly, Hugh Atkinson predicts that the organizational structure of the future does seem to require an increasing num- ber of smaller, decentralized units and that "these changes will require a new kind of central administrator, a coordi- nator or evaluator ... " rather than a controller. 14 COSTS The supposed high cost of maintain- ing departmental libraries, although espoused as dogma, has not been sub- jected to many empirical studies. One of the few is a study by Snunith Shoham of the Library School Library at the Univer- sity of California at Berkeley. He studied the cost of maintaining a departmental library in relation to the costs to the user and the university. In his analysis of the costs of the Library School Library, he estimated that 44 percent of the total labor costs for that library would exist 516 College & Research Libraries even if it were not a branch.15 In examin- ing material costs, Shoham concluded that only about 7 percent of the materials cost is for duplication of resources.16 He also examined users' costs (i.e., travel costs, time necessary to locate informa- tion, waiting time) and concluded that "one would have to impute extremely low values to the hourly value of users' time for their preferences for a con- venient, decentralized provision not to be economically justified, once all costs are taken into account." 17 Charles R. Martell also calls into question this focus on administrative costs when he states that "the tendency of libraries to measure inputs rather than outputs has been a problem of continuing concern to many librarians." 18 The costs, depending on the organizational structure, can be substantial. In addition to materials duplication ... the costs of maintaining departmental libraries can include personnel, equipment, and support for electronic resources. It may well be that belief in the cost effectiveness of decentralized libraries depends on world view. For example, in the book Austerity Management in Aca- demic Libraries, Joseph Z. Nitecki states that management in an "economy of de- cline requires efficient operation which in tum implies strictly enforced dead- lines and centralization," while Atkin- son and Patricia F. Stenstrom in the same book suggest using a decentralized on- line system to avoid duplication of effort and to recover costs.19 Indeed, Atkinson elsewhere proposes that not only may the departmental library be cost-effec- tive but that the amount of administra- tive managerial overhead may be reduced because "in smaller units much of that 'institutional overhead' is elimi- nated because the ability to have direct interpersonal interaction is at a high enough level to eliminate the need for formal com:munication, management and support activities." 20 November 1992 Undeniably, costs are associated with departmental libraries. The costs, de- pending on the organizational structure, can be substantial. In addition to materi- als duplication (which, as Shoham showed, is probably quite low), the costs of maintaining departmental libraries can include personnel, equipment, and support for electronic resources. Personnel costs tend to be the largest proportion of most library budgets. Many proponents of centralization cite this cost, based on a 1960s estimate by N. Orwin Rush, as one of the main reasons for centralization. In his article, Rush quotes from a letter by Director of the Univer- sity of Michigan Library Frederick H. Wagman that "fully 30 percent of the personnel budget of my library system is spent in staffing the many branches in less than adequate fashion." 21 However, this statement minimizes the fact that personnel would still be needed to per- form these functions in a centralized li- brary and ignores users' costs. In fact, a study by Charlene Renner and Barton M. Clark at the University of Illinois showed that staffing patterns were most highly correlated with the circulation and number of volumes in a collection.22 This study would seem to indicate that whether a collection was centralized or not has no significant bearing on the number of staff needed. Thus, given that collection size and circulation levels should not vary when collections are centralized, it is unlikely that great savings in personnel costs would be achieved through such a move. Equipment duplication will exist to some extent in a decentralized system. Additional fax machines, circulation and OPAC terminals, and microform readers will be needed. However, eliminating a departmental library will not completely eliminate these costs since the number of machines for public use (OPAC, micro- form readers) is depen-dent on the num- ber of users rather than on those users' geographic location. Conversely, some decrease in the number of fax machines and circulation terminals could be realized by centralization. While these savings would be small, nevertheless they could allow purchase of fewer but higher qu- ality items that could be shared by more staff in a centralized location. Duplication of electronic resources and workstations, such as CD-ROM databases, can be expensive. However, recent techno- logical advances will eliminate this prob- lem. The networking of CO-ROMs, the loading of databases into online cata- logs, the accessibility of the Internet, and other such developments are providing a decentralized, distributed information environment that permits users to access information from the location of their own choosing rather than forcing them to come to some particular structure. As Woodsworth notes, "When today's tech- nology is utilized fully, the issue of cen- tralized collections will fade into oblivion."23 Even in the smallest academic institutions, networking is becoming available and is, therefore, eliminating the need for duplication of resources. Another criticism of departmental li- braries is that access is limited and knowledge is dispersed. Automated union catalogs, shared utilities, and in- creased reliance by all libraries on re- source sharing have made library col- lections more accessible and decreased re- liance on comprehensive, centralized col- lections. Furthermore, departmental li- braries do tend to have integrated collec- tions (contrary to Watts' view) in that the interdisciplinary instructional and re- search interests of their user communi- ties are reflected in the collection. An engineering collection will not contain only materials from the Library of Con- gress (LC) T classification, just as an economics collection will contain mate- rials from LC classifications other than H. In fact, as Holley states, "One might well make a case that small well-selected collections located conveniently to users would be more valuable to most of the university community than a large stack of infrequently consulted material cen- trally housed and scrupulously but ex- pensively cataloged."24 Indeed, brows- ing and serendipity may actually be en- hanced in departmental libraries. For ex- ample, users trying to browse only for business journals in a current periodicals Organizational Misfits 517 room filled with thousands of titles from scores of disciplines may find the ex- perience quite frustrating. THE DATA In the ARL SPEC Kit 99, "Branch Li- braries in ARL Institutions," ninety-four university libraries reported on their branches, with a total of 1,008 branches reported. Sixty-eight percent of the re- sponding libraries have centralized library systems and 32 percent have decentralized systems. The average number of branches per library in a centralized system was six, while thirteen was the average in a de- centralized system. The most common branches were music, mathematics, en- gineering, physics, chemistry, business, architecture, geology, rare books/special collections, science, and undergraduate libraries. Finally, more libraries reported establishing new branches in the pre- vious five years than closing existing branches.25 At least in ARL libraries, branch libraries are a fact and do not seem to be disappearing rapidly. The idea of maintaining service- oriented decentralized units is no longer considered heretical. Martell, in his Client- Centered Library, suggests a model of de- centralized units that closely resemble the roles and functions of the typical de- partmentallibrary, while not explicitly dis- cussing departmental libraries. All of this would imply that the real question for departmental libraries is not whether they should exist. They do and obviously will continue to do so. Nor is the question whether they pro- vide services that fulfill articulated needs-clearly, they do. Rather, the issue is how to design an organizational struc- ture that will allow these nontraditional entities to fit, enhancing and supporting their creative, client-centered character. Resolving this issue will help begin to move the entire library system toward the more proactive, flexible structure needed to fit into a fast-changing infor- mation technology environment. However, changing to a more appro- priate structure may be difficult. As Ken Jones points out, ''Libraries have been embedded in stable and predictable public 518 College & Research Libraries and academic organizations concerned more with internal regulation and input control than with outgoing enterprise and maximization of effectiveness" and as a result, ''bureaucratic organizations tend to suffer from goal displacement, that is, displacement of terminal values by instrumental values. The work, the social relations and the organization it- self become their own justifications, with little reference to the clients."26 ORGANIZATIONAL STRUCTURES AND MODELS The placement of departmental librar- ies within the organizational structure exemplifies this phenomenon. Little con- sideration seems to be given to the actual characteristics of the units or the optimal way t9 position them within the struc- ture to best fulfill user needs. Just as little agreement on the nature of departmen- tal libraries exists, little agreement on their place in the organization occurs. This lack of agreement can be seen in the ARL survey on organization charts (SPEC Kit 129). In analyzing the 61 charts for those libraries that reported having branches, the authors found that 38 percent of the branches reported to an . administrator for public services, 13 per- cent reported to administrators for sub- ject libraries, 10 percent reported to the director, and, in 18 percent, the reporting lines varied by departmentallibrary.27 These structures and reporting lines tend to be of four types, which we have labeled: single-dimension functionalism, administrative, subject-divisional, and in- tegrated-collegial. All of them have some advantages, but all of them have the dis- advantage of denying (or suppressing) the multifunctional, boundary-spanning na- ture of these units. Following are general- ized descriptions of these three models. Single-Dimension Functionalism The main characteristic of this model is that all departmental libraries report to an administrator of a single functional area in the library system, such as an assistant director for public service. This structure is the predominant one re- ported by ARL libraries in SPEC Kit 129. November 1992 It has the benefit of keeping departmen- tal libraries together in one administra- tive unit. This type of structure implies that the functional unit within which de- partmental libraries are placed is the pri- mary, or only, focus of their operations. The problem, of course, is that this place- ment denies the multifunctional aspects of these units. Departmental libraries by definition are libraries that encompass all of the varied library functions, such as public services, technical services, ac- quisitions, and outreach. Departmental libraries must regularly interact with many departments on many different levels. Thus, their problems and needs are quite different from those of a depart- ment that has only one primary function. Denial of the malfunctional departmen- tal libraries is part of the reason so much conflict and managerial difficulty sur- rounds these units. Administrative Similar to our first model, this model is also characterized by a hierarchical structure. However, in this case, de- partmental libraries report to an admin- istrator who has more generalized administrative responsibilities than re- sponsibility for just one function. For ex- ample, departmental libraries may report to the head of the largest or central library. To some extent, this model does address the issue of the multifunctional- ism of departmental libraries. However, it assumes that the largest library is nec- essarily the heart of the system. In fact, conflict can arise when libraries are placed in superior/inferior relationships to each other. The head of the central library's main job is to provide services to that clien- tele. This commitment may result in serv- ices to other clienteles being lessened, not because that is objectively appropriate but because that is required to allow heads of the central libraries to do their jobs. Some balance can be achieved by creat- ing an administrator for departmental libraries and an administrator for the central library and placing them on the same hierarchical level. The problem here, especially in large systems, is again the penchant for a hierarchical, control- oriented management style. User needs become difficult, if not impossible, to communicate up the hierarchy, and flex- ibility and creativity in the frontline units are easily suppressed. Subject-Divisional Characterized by dividing the library into units similar to academic depart- ments on the basis of disciplinary boun- daries, this structure would organize the system by subject areas with each subject area having an administrator. This model attempts to unify the knowledge base of libraries by basing the structure on the organization's intellectual and edu- cational mission. Thus it draws together resources in terms of knowledge rather than architecture. However, the location of certain functions, such as general ref- erence or technical services, can become problematic. Also, compartmentalizing the entire library system can foster further bureaucratization, leading to a frag- mented and overspecialized structure. These models parallel Henry Mintz- berg's structural configurations. His ma- chine bureaucracy configuration in which much formalization occurs and the flow of decision making is top down is similar to the administrative model described above.28 His professional bureaucracy with its emphasis on standardization of skills and horizontal and vertical de- centralization is similar in some ways to the single dimension functionalism .. Our subject-divisional model shares traits with his divisionalized form in that both em- phasize standardized outputs and have little vertical · decentralization and limited liaison devices.29 These structures are based on the con- figurational view of organizations de- scribed by Gregory K. Dow. He defines the configurational view as one that "emphasizes the integration of work tasks under common managerial author- ity."30 Thus, the focus in most organiza- tions on control and centralization of services in analyzing and dealing with departmental libraries clearly fits into the configurational type of structure. Another view of organizational struc- ture that Dow discusses is coactivational. Organizational Misfits 519 ''The coactivational view stresses recur- rent patterns of interaction among or- ganizational participants."31 In this type of analysis, the units in an organization are analyzed by the type and intensity of their interactions with other units in the organization. In the coactivational view depart- mental libraries would be con- sidered nodes in a communication net- work rather than vertical lines on a chart. This type of structure permits a greater flexibility that is better suited to units that span traditional functional boundaries. Resolving this issue will help begin to move the entire library system toward the more proactive, flexible structure needed to fit into a fast-changing information technology environment. Mintzberg describes a similar type of structure when he states that "sophisti- cated innovation requires a ... very different structural configuration, one that is able to fuse experts drawn from different disciplines into smoothly functioning ad hoc project teams." He calls this structural configuration ad- hocracy. One type is the operating ad- hocracy that "innovates and solves problems directly on behalf of its clients," and a key feature is "that its administrative and operating work tend to blend into a single effort."32 This description seems ac- curate for the boundary-spanning, client- centered characteristics of the departmental library and is certainly a role model to be emulated not only by departmental li- braries but also by all libraries as they begin to position themselves for a very different future. Mintzberg continues, ''The adhocracy is clearly positioned in an environment that is both dynamic and complex," and "a dynamic environ- ment calls for organic structure and a complex one calls for decentralized structure. "33 The adhocracy is characterized by little formalization but much liaison, with decision making spread among units, 520 College & Research Libraries and by horizontal job specialization. Within the academic context it could be called an integrated-collegial model. This is our fourth model. Integrated-Collegial This model is characterized by a flat organizational structure in which de- partmentallibraries directly participate in the policy-making management of the or- ganization, rather than reporting through a pyramidal or divisional structure. This model presup- poses a high level of pro- fessionalism and a high degree of re- sponsibility, openness, and a holistic world view. Each library or work unit would be responsible as a group for developing services and meeting client needs. The senior administrator for de- partmental libraries would be replaced by a coordinator, possibly rotating every few years. This model provides a mech- anism for considerable input and com- munication among all members of the organization and allows individuals to contribute their strengths. However, the differing abilities of the various mem- bers of the group as well as their willing- ness to participate fully in the process can be a source of conflict. In addition, maintaining a collective unity can be problematic with a rotating leadership. In spite of some problems, this model allows multifunctional, client-centered units, such as departmental libraries, to fit into the organizational structure without necessarily sacrificing flexi- bility and creativity. While the other models focus on control, this model em- phasizes shared responsibility and en- hanced creativity and flexibility. Such an organizational structure can accommo- date the autonomous, multifunctional nature of departmental li- braries and channel these energies into support for system-wide goals. By reducing bureau- cracy and encouraging participation, much of the divisiveness and fragmenta- tion arising from structural constraints can be eliminated. At the same time, it positions the entire library organization to move into a more proactive, boun- dary-spanning role within the larger or- ganization. November 1992 CONCLUSION These models and analyses are in- tended to open discussion on designing an organizational structure that will fit de- partmental libraries and will provide the flexibility and entrepreneurial manage- ment environment increasingly necessary for libraries, especially large academic li- braries, to retain their viability in a time of such great change. In describing the fu- ture model for university libraries, D. Kaye Gapen states that "we have to par- ticipate, we have to say what we think; we have to put some of our responsibility on the line and be assessed by the people we work with, and that's not so easy. We have to be open and show initiative and courage, and those are qualities that are difficult at times to develop."34 Indeed, library users, and especially academic users, are becoming so information and computer literate that with distributed computing and networks, they, rightly, expect to be limited no longer to accessing information and library resources in only one monolithic location. Thus, the regular call for elimination of departmental libraries may not only be myopic but also illogical and ulti- mately self-destructive. The call derives more from received wisdom than from a highly centralized organizational struc- ture making sense in the era of a computer on every desk. Indeed, library users, and especially academic users, are becoming so information- and computer-literate that with distributed computing and net- works, they rightly expect to be limited no longer to accessing information and library resources in only one monolithic location. This, then, is the challenge for library organizations. It is time for de- partmental libraries to be considered as models for the library organization rather than as misfits. The systems approach to services and organizational structure of today's departmental libraries may be the characteristics of tomorrow's libraries. Organizational Misfits 521 REFERENCES AND NOTES 1. Dilys Morris, "Electronic Information and Technology: Impact and Potential for Aca- demic Libraries," College & Research Libraries 50:56-64 (Jan. 1989). 2. David W. Lewis, "An Organizational Paradigm for Effective Academic Libraries," College & Research Libraries 47:337-53 (July 1986). 3. Ibid, p.349. 4. John R. Sack, "Open Systems for Open Minds: Building the Library without Walls," College & Research Libraries 47:535-44 (Nov. 1986). 5. David W. Lewis, "Inventing the Electronic Library," College & Research Libraries 49:291- 304 (July 1988). 6. D. Kaye Gapen, "Myths and Realities: University Libraries," College & Research Libraries 45:350-61 (Sept. 1984). 7. For another recent example of this phenomenon, see Irene B. Hoadley and John Corbin, "Up the Beanstalk: An Evolutionary Organizational Structure for Libraries," American Libraries 21:676-78 (July/ Aug. 1990). 8. Edward G. Holley, "Reaction to 'A Brief ... ,"' Journal of Academic Librarianship 9:201-2 (Sept. 1983). . 9. Barbara B. Moran, Academic Libraries: The Changing Knowledge Centers of Colleges and Universities, ASHE-ERIC Higher Education Research Report No. 8 (Washington, D.C.: Assn. for the Study of Higher Education, 1984), p.32 10. J. Michael Bruno, "Decentralization in Academic Libraries," Library Trends 311-17 (Jan. 1971). 11. Ibid., p.314. 12. Thomas D. Watts, "A Brief for Centralized Library Collections," Journal of Academic Librarians/zip 9:196-97 (Sept. 1983). 13. Anne Woodsworth, "Decentralization Is the Best Principle of Organization Design Where It Fits," Journal of Academic Librarians/zip 9:198-99 (Sept. 1983). 14. Hugh C. Atkinson, "Impact of New Technology on Library Organization," The Bowker Annual of Library and Book Trade Information, 29th ed. (New York: R.R. Bowker, 1984), p.113. 15. Snunith Shoham, "A Cost-Preference Study of the Decentralization of Academic Li- brary Services," Library Research 4:175-94 (Summer 1982). 16. Ibid., p.182. 17. Ibid., p.188. 18. Charles R. Martell, The Client-Centered Academic Library (Westport, Conn.: Greenwood, 1983), p.71. 19. Joseph Z. Nitecki, "Creative Management in Austerity," Austerity Management in 20. Academic Libraries, p.43-61; Hugh C. Atkinson and Patricia F. Stenstrom," Automation in Austerity," Austerity Management in Academic Libraries, p.277-85. Hugh C. Atkinson, "A Brief for the Other Side," Journal of Academic Librarians/zip 9:200-201 (Sept. 1983). 21. N. Orwin Rush, "Central vs. Departmental Libraries," Mountain Plains Library Quar- terly 7:3-4 (July 1962). Charlene Renner and Barton M. Clark, "Professional and Nonprofessional Staffing Patterns in Departmental Libraries," Library Research 1:153-70 (1979). 23. 24. 25. 26. 27. Woodsworth, "Decentralization Is the Best Principle," p.199. Holley, "Reaction to 'A Brief ... ,"' p.202. Branch Libraries in ARL Institutions, ARL SPEC Kit 99. (Washington, D.C.: Assn. of Research Libraries, 1983). Ken Jones, Conflict and Change in Library Organizations: People, Power, and Service (London: Clive Bingley, 1984), p.9, 29. Organization Charts, ARL SPEC Kit 129 (Washington, D.C.: Assn. of Research Libraries, 1986). 28. Henry Mintzberg, The Structuring of Organizations: A Synthesis of tlze Research (Englewood Cliffs, N.J.: Prentice-Hall, 1979). Machine bureaucracy is defined as a structure in which operating work is routine and work processes are standardized, p.315. 29. Ibid., p.305-467. 522 College & Research Libraries November1992 30. Gregory K. Dow, "Configurational and Coactivational Views of Organizational Struc- ture," Academy of Management Review, 13:53-64 (Jan. 1988). 31. Ibid, p.53. 32. Mintzberg, Structuring of Organizations, p.437. 33. Ibid., p.449. 34. Gapen, "Myths and Realities," p.360. OCLC/AMIGOS Collection Analysis Systems Make a wise investment. Choose from three options to analyze your library's data: Collection Analysis CD compares quantitative data BCL3 Tape Match measures against a standard Tape Analysis fits individual specifications Available exclusively from AMIGOS Bibliographic Council, Inc. 12200 Park Ce ntral Drive, Suite 500 Dallas, Texas 75251 214/851-8000 or 800/843-8482