College and Research Libraries Research Notes Library Cooperation: A Serials Model Based on Philosophical Principles Kurt Pond and Dwight F. Burlingame Library cooperation has been an often- stated goal m· library annual reports, at conventions and workshops involving li- brarians, and particularly in recent library literature. Noted examples of success are evident across the country; 1 however, few library scholars, if any, hilVe examined the philosophical principles of cooperation as the basis for a successful foundation for in- terinstitutional library cooperation. Bowling Green State University (BG) and the University of Toledo (UT) libraries utilized an applied philosophy model in developing their cooperative protocols. A philosophy intern was employed by both institutions to facilitate the process. In the summer of 1981, the two library staffs dis- cussed the possibility of beginning greater cooperation in the area of serials. The proj- ect was implemented during January of 1982, and shortly thereafter the intern was brought into the project to be the major partner in formulating, presenting, and evaluating a serials survey that was dis- tributed to faculty members of the En- glish, biology, and economics depart- ments of Toledo and Bowling Green. The experience gained by applying phil- osophical principles to library cooperation has been instructive and beneficial to the su<::cess of BG-UT cooperative ventures . Indeed, many of the principles are general enough to be applied not only to other co- operative situations in a library setting, but to almost any cooperativeventure. · PHILOSOPHICAL VIEWS CONCERNING COOPERATION Examining philosophical perspectives of cooperation and coordination and how such philosophical principles and models can be applicable and transferrable to li- brary cooperation is desirable in achieving success in a cooperative venture. Once li- brary staffs are willing to cooperate with each other, and each library has the tech- nology to support cooperation, the next logical step is to analyze and develop pro- cedures and principles that will be effec- tive in making cooperation ~ actuality. To cooperate means to work together for a common objective or to unite in produc- ing a desired effect. This definition implies that cooperation involves more than one agent, that the agents involved have coor- dinated their actions, and the agents have a shared goal. Further, a coordination of beliefs usually needs to have occurred in order for cooperation to occur. For Donald Rega.n, coordination "sug- gests a primary focus on the parties' ulti- mate behavior," 2 whereas we have stated cooperation means to work together to- warq a common goal. "~orking to- Kurt Pond is a ph~losophy intern and Dwight F. Burlingame is dean of libraries and learning resources, Bowl- ing Green State University, Bowling Green, Ohio 43403. 299 300 -college & Research Libraries gether'' connotes behavior or actions that are harmonious. Thus, a.gents involved -in a cooperative venture have successfully transformed individual action into collec- tive action. Regan suggests that-coopera- tion involves not on1y coord-inated actions, but -coordinated beliefs and atti- tudes as welL For him, coordinated actionS are usually the result of -coordi- nated beliefs and attitudes. Actions can be seen as putting into prac- tice certain heliefs and attitudes. There- fore, if we wish to modify actions we must modify the beliefs and attitudes responsi- ble .for such actions. If we desire to coordi- nate _actions we must attempt to coordi- nate the beliefs and attitudes behind such actions. When speaking .of the objectives .of co- operation, a .distinction -needs to be made "between~ooper-ation as an activity and the objectives-Or -g.oals -of -cooperation. Coop- eration is an activity that is a means !o an end. Edna -Ullmann-Margalit states, .liThe achievement uf-coordination is Itself but a means to a ·further end, that of 1lttaining a cooperative goal.H3 The ..activity and the objectives of -the ~activity must be --distin- -guished, for without making such a rlis- tinc-t1on, false inferences may result. Some may assumeihat to be cooperating Is .con- -tingentu_pon reaching selec-ted objectiv-es, or that agents cannot be said to be cooper-:- ating until those objectives have been reached. In addition to coor-dinating beliefs, -alt~­ -tudes, and actions, the selection uf coop- er-ative objectives and goals itseU is cruCial in ,aetermining (aj whether .cooperation-is t-o occur and :(b) whether the c-ooperative goal will be reached. There are cases in whkh the act of cooperation is an end in itself. Yet .sum situations seem to 1>e the :exception ratber lhan tbe norm. In many situations# coeperation can be a burden- some -activity. Even in many -situations where the act.of~-ooperation brings :enjoy- ment., it is usually-the case that suchoenjoy- ment is engendered from the agents• ex- pedatioru; of .obtaining the benefits deriv,ed Irom cooperatiflg. Cooperation will oocur if the rooperalive objectives are worthwhile aad will compensate £or the sacrifices ilw'Olved in eooperating. July 1984 Agents cannot always know a pnon whether cooperative objectives can be de- finitely reached or not. Agents must rely -on probabilities rather than certainties. If a desirable objective has a good probability of being obtained, then it is reasonable to pursue such an objective. Yet, selecting desirable and obtainable objectives does not guarantee that such objectives will be reached. David Lewis states, ''A cooperative goal will be reached if the nature of the situa- tion is clear enough. ''4 Lewis is saying that if the cooperative situation is correctly un- derstood, then appropriate {i.e., desirable and obtainable) goals and Gbjectives will usually he selected. Thus, in order for a co- operative objectiv-e to be reached, a c-oor- dination of beliefs and actions is not suffi- dent; rather, the -coordinated belief-s, atti- tudes, and actions mu-st be appropriate for the cooperative situation. ·In summary, reaching a cooperative ob- jective involves (l) c~onfination of be1iefs~ attitudes .and actions; (2) a -.correct under- standing of -the cooperative situation; ~3) selection of "objecliv,es that are desirable and ~btainable; and ( 4) .dev-elopment .and implementation .of an appr-opriate method, procedure, ,or course of action .lhat is instrumental toward -reaching the selected cooperative obJective. Meeting aH -of -the above conditions l eads to a high probabiiigr of successful cooperalion. Cemmunication is an effective activity · in coordinating belief-sJ attitu-de-s.., an-d actions. Agents can communicate with each other about-the cooperative Situation and joit\lly -select -appr.opriate 'Objectives and methods for obtaining 'Such objec- tives. -An eifectiv..e cooperative procedure is one in which ,c-ommunication is both :full and honest. By., .. full,.,, we mean -that-com- munication between .agents is -operative thr-oughout the <:<>o~per-ative enterpi-ise. Honest communication engenders c-oop- eratien.. When -agents realize they are -be- ing honestly wormed, they will tend to trust and :acquir-e "CCnfidence in tbe others in theceooperative project. The agents invo1v~ -in n cooperative -project must -reach an agreement as to what form the cooperative project is to take and what objectives :should be -pur- . sued. Without such an agreement, coop- eration will not occur. Whether all of the agents take part in formulating the proce- dures and objectives in a cooperative proj- ect, or just a few, the agents involved must agree to cooperate. If agents believe that a cooperative project is worthwhile then they will jointly att~mpt to coordinate their preferences, for ''conflicting prefer- ences are overwhelmed by our desire to concert our choices. Rational agents will realize that it is to their interest to cooper- ate for without such activi~ a cooperative goal will not be reached.'' Finally, coordinating preferences can of- ten be accomplished through enhance- ment. By enhancement we mean there- shaping of the precooperative situation in such a way that a more desirable outcome becomes possible. Defining what kind of cooperation is to occur can determine (a) if agents will cooperate and (b) the level of enthusiasm the agents will have toward the cooperative project. The way the coop- erative project is presented can determine whether the potential cooperators will be willing to cooperate with each other. In summary, it seems that engendering cooperation is a nearly impossible task given all the above requirements. Yet, we contend that an awareness of such princi- ples or necessary conditions increases the likelihood of being successful in a cooper- ative project. BG-UT SERIALS PROJECT During the summer of 1981, members of the library advisory committees from BG and UT met to discuss several areas of pos- sible cooperation between the institu- tions . From this meeting the formulation of two objectives transpired: (1) to initiate a van service that would transport library materials and individuals between cam- puses, and (2) to concentrate on serials de- velopment. In the fall of 1981, the van service was successfully implemented and continues today. In May of 1982, the library advisory committees met again to discuss the feasi- bility of joint sharing, acquisition, and cancellation of serials. The major reasons for concentrating on serials were the large Research Notes 301 share of the acquisitions budget devoted to serials and that each library was able to hold only a relatively small number of seri- als titles available for purchase. Through cooperation both libraries could make available locally more titles to their clien- tele as well as to other citizens in the re- gion. Also, library cooperation had been encouraged by the Ohio Board of Regents in their Master Plan of Higher Education, 1982. Several conditions led to the success of cooperation between BG and UT. First, the staffs of both libraries were willing to cooperate. This favorable consensus is far from trivial when one considers the num- ber of articles in the literature that discuss the difficulty of getting library staffs to co- operate with each other. Second, the two libraries are in rela- tively close proximity with each other (twenty miles). Third, a van service had been developed as a way of transporting the serials between the two institutions in a timely manner. The van operates be- tween each campus tWice daily (Monday through Friday). Fourth, the OCLC sys- tem allows each institution to know what the other's holdings are. Fifth, the project involved not only li- brary personnel at both universities, but faculty members from selected academic departments. Three departments were se- lected to participate in the experimental project: English, biology, and economics. The rationale behind selecting these de- partments was that the English depart- ment would represent the humanities and a doctoral-degree-granting department, biology would represent the natural sci- ences as well as a doctoral department, and economics would represent the social sciences and a master-degree-granting de- partment. Sixth, the serials were evaluated interin- stitutionally rather than intrainstitu- tionally. Joint lists of the serials held at both institutions were compiled. For ex- ample, a joint list was formulated for all of the "English department related" serials held at both universities. The faculty members in both university English de- partments were given the opportunity to evaluate the serials located on their cam- 302 College & Research Libraries pus, and the serials held on the other cam- pusas well. PROCEDURES FOR SERIAL COOPERATION Recognizing that an effective method or procedure for a cooperative project will be one that is both informative and instruc- tive, the following procedure was imple- mented for the serials project. The se- lected faculty members from the English, biology, and economics departments were given the opportunity to cooperate by evaluating the serials pertaining to their fields of study. The project began with the English department, the ration- ale being that the procedure would run · more effectively if just one department was surveyed at a time. Also, lessons could be learned from the successes and failures with the English department and implemented with the other departments. Additionally, if the English project was successful, this success would then serve as a testimony to the other departments that such a cooperative serials project was worthwhile. Each serial was to be evaluated accord- ing to the following scale of recommenda- tions: A = the faculty member wanted the serial to remain on the campus at all times; B = the serial may be transferrable (i.e., tradable) on the condition that it be available within a twenty-four hour period; C = the serial may be transferrable on the condition that it be available through interlibrary loan (i.e., longer than a twenty-four hour period); D =· the faculty member believed that the serial had little or no research value;* and E = the faculty member had little or no basis for evaluating the serial (i.e., abstaining from evaluating the se- rial). The faculty r·esponse sheets were com- July 1984 puted and tabulated, and a sample title printout looked as follows: Insti- tution BG UT Serial Number 117 117 A 13 8 B 13 8 c 9 8 D 0 0 E 59 66 Numerous "well-founded" inferences could then be derived from the data. For serial 117, many of the faculty members from the English department at Bowling Green refrained from evaluating the serial (59 percent in the E category), while 13 percent felt that the serial should be lo- cated on the campus at all times. Many of the English faculty at Toledo felt that they had no basis for evaluating the serial ( 66 percent), while 8 percent felt that the serial should be on their campus at all times. Though it appeared that the serial was not used by most of the English faculty, thi~ was not to say that the serial had little or no value. Eight percent at one institution and 13 percent at the other believed the se- rial to be nontradable. Compare this serial evaluation to a serial numbered 125: Insti- Serial tution BG UT Number 125 125 A 0 0 B 9 0 c 0 8 D 4 0 E 81 83 For serial125, there appears to be no inter- est in holding the serial on either of the campuses at all times (i.e., A = 0 at both institutions). At Bowling Green, only 9 percent felt that if the serial were shared, it should be made available within twenty- four hours. Eighty-one percent refrained from evaluating the ~erial, meaning that the faculty members either never heard of the serial or never used the serial, or per- haps used the serial only a few times. Re- gardless, the serial was not highly valued for the English faculty at Bowling Green. The results were similar for the English faculty at Toledo. General inferences can be drawn from such results. The serials with no percent- ages of responses under the A column meant that the serials were possible candi- · *The D response category has subsequently been omitted from th~ survey because of low respon~es based on the assumption that faculty members chose not to say a htle had no research value, but m- stead chose to refrain from evaluating the serial (E category). dates for sharing. If the total printout re- vealed that high percentages appeared under the A column, then the success of the cooperative project would probably be in jeopardy, for it would seem to indicate that faculty members were unwilling to cooperate by way of sharing serials. Fortu- nately this was not the case. The rationale for adding the B and C columns was that if there appeared to be no interest in holding a particular serial on either of the campuses at all times (i.e., A = 0), the question arose of how the serial was to be shared. How a serial was to be shared depended upon the B (twenty four-hour period) and C (interlibrary loan) responses given by the faculty members. For example, if 0 percent chose A and 0 percent chose the B category, then this would seem to indicate that the respond- ing faculty members felt that the serial could be shared and be made available through interlibrary loan. The benefit in such a serials survey is that the value of a serial can be ascertained through the responses of the teaching fac- ulty members evaluating the serials. Fre- quency of use is one indicator of the value of a serial, but by itself it seems to be insuf- ficient as a criterion of judgment. A low frequency of usage does not guarantee that a particular serial is not in demand. Librarians asking teaching faculty mem- bers to help evaluate serials is an effective method for ascertaining the value of a se- rial. PROPOSALS BASED UPON THE COMPUTER ANALYSIS After the results of the serial survey were evaluated, further categories were formulated in order to classify and pro- pose a future status for each of the serials. The categories were as follows: first, there was a "wrong institution" category. These were serials that were held at one university in which the faculty members of the English department at that univer- sity expressed no desire to have that serial located on their campus at all times (i.e., A = 0 at the institution holding the se- rial). In addition, the faculty members of the English department at the other uni- versity where the serial was not held, ex- Research Notes 303 pressed interest in having the serial lo- cated on their campus at all times (i.e., A = 0). Fifteen of the English serial titles fell under this "wrong institution" category. The second category concerned serials in which there was no interest in holding the serial on either of the campuses at all times (i.e., A = 0 at both institutions). Nine English serials fell under this classifi- cation. The third category concerned serials that were held at both universities, but only one university expressed a desire to have the serial held on their campus at all times (i.e., A = 0 at one institution, and A > 0 at the other institution). Nine of the English serials fell under this third cate- gory. The first proposal, based upon the sur- vey results, was as follows: Serials for which a change of status was considered were the serials in which A was not greater than zero at either institution (A = 0). In other words, if any faculty mem- ber desired to have the serial located on his/her campus at all times, then that se- rial was not considered shareable. When a serial did not receive any A evaluations, and it fell under a B or C classification, then that serial was considered negotiable and a change of status was proposed. How the serial was to be traded depended upon the Band C responses. The proposal stated that if any faculty member chose B, and none chose A, then the physical availability of the serial within a twenty-four hour period was pro- posed. If no faculty member chose B as well as A, then it was proposed that these- rial be available through interlibrary loan. If a particular serial was held at one insti- tution and the faculty members of the En- glish department at that institution ex- pressed no desire to have the serial located on their campus at all times (i.e., A = 0), and if the other institution, not holding the serial, expressed the desire to have the serial located on their campus (A > 0), the proposal was then for the transfer of the serial (both back issues and subscriptions) to the institution that expressed the desire to hold the serial. If such a transfer oc- curred, then the serial could be available at the institution that transferred the serial 304 College & Research Libraries within a twenty-four-hour period or through interlibrary loan, depending upon the~ and C responses. The English serials considered for can- cellation were the ones in which (1) A = 0 at both institutions, and (2) B = 0 at both institutions. Such a serial, if cancelled, could then be made available to both insti- tutions through interlibrary loan from a third institution. If one faculty member chose B at only one institution (i.e., B > 0), then the proposal was to keep the serial on the campus expressing a need and then to make the serial available to the other in- stitution through interlibrary loan. Implementation of the above proposals would redirect library funds, which could result in actualizing the goal of making more serials available to library users. It should be noted thanhe above proposals were suggestions, not decrees. Whether such proposals would be accepted de- pended upon the consent of the faculty members involved in the project, as well as the librarians involved in the serials project. The attractive features of the BG-UT se- rials project were as follows: First, the sur- vey contained clear directions on how to evaluate the serials. What the A toE cate- gories meant was clearly and explicitly stated in the directions. Also, theE cate- gory gave the faculty members the option of refraining from evaluating a serial if they felt unqualified to do so. Second, the consequences of actions were clearly defined . The agents involved in the project knew what making certain evaluations entailed. For example, in fill- ing out the serial survey, if a faculty mem- ber chose B for a serial, that meant that the faculty member wished to have the serial shared and be made available within twenty-four hours. If that faculty member would be dissatisfied with such a status, then that faculty member could select a different classification for that serial. Third, the agents not only knew the con- sequences of evaluating serials, but they had a great deal of control over the conse- quences of their actions. In addition to evaluating serials, the faculty members were given the opportunity to evaluate the proposals that were based upon the July 1984 survey responses. The faculty members were, in fact, encouraged to offer sugges- tions, criticisms, and general feedback about the proposals and the project as a whole. Modifications were made upon such suggestions and because of this the project was flexible; such open communi- cation and flexibility engenders trust and cooperation. The fourth attractive feature of the pro- cedure was that responding to the serials survey (i.e., cooperating) involved a mini- mal amount of sacrifice. Completing the survey took at most twenty minutes. This time period was minimal in comparison to, fifth, the benefits that could be derived from responding to the survey. For exam- ple, if by evaluating a serial, a faculty member gained access to a desired serial that previously was not available, then co- operating would have proven worth- while. If a faculty member gained greater access to two or more serials, cooperation was that much more beneficial. Sixth, the position of the agents in the cooperative project was attractive. The agents began and remained on an equal status with one another. Favoritism or an imbalance of power was absent, for each survey response carried equal weight and the faculty members remained anony- mous. Thus, it was not a situation in which some agents acted out of a position of strength and others out of weakness. Also, it was not a situation of conflict be- tween agents. In addition to equality among agents, there was a relative equality of library in- stitutions. Serial holdings at each library were fairly equivalent. Thus, like the co- operating agents, it was not a situation in which one library was acting from a posi- tion of strength and the other from a posi- tion of weakness. Both library institutions had something to offer the other and this was another factor that made serial coop- eration attractive to all parties. Another feature about the procedure was that it worked toward getting non- cooperating agents to cooperate. For ex- ample, suppose a faculty member intially refused to cooperate by not responding to the serial survey. Then, a proposal was presented in which a serial that the non- . . cooperating agent wished to remain at his home institution was to be cancelled or shared. It would then be in the best inter- est of that agent to respond, and in this case, object, to such a proposal. In doing so, the noncooperating agent has begun cooperating. This, in addition to the pro- cedure implemented being attractive, such a procedure engenders and con- tinues cooperation. SUMMARY The general goal of the cooperative proj- ect at BG and UT was to achieve an in- creased number of available serials to us- ers at both campuses. This outcome was perceived as desirable by all of the agents involved in the project. Yet, the task con- sisted of formulating and initiating a pro- cedure that would prove instrumental in achieving an increased number of avail- able serials. This was where the philo- sophical principles of cooperation came into play. The procedure that was imple- mented for the serials project was based upon the philosophical principles of coop- eration. The procedure was informative, instructive, attractive, fair, engendered cooperation, and proved instrumental in obtaining the cooperative goal. As discussed, cooperation involves the coordination of beliefs, attitudes, and actions. Those responsible for initiating the cooperative project were cognizant of the philosophical principles discussed. Thus, the procedure was informative and appropriate for the cooperative situation, which worked toward coordinating be- liefs. Also, the procedure was fair to the Research Notes 305 agents involved, and the benefits from co- operating far outweighed the sacrifices in- volved in cooperating. Thus, the coopera- tors were enthusiastic about the project, which worked toward coordinating their attitudes. Finally, the procedure was eas- ily understood and the agents knew what specifically was required of them in the co- operative project. Thus, actions were co- ordinated and appropriate. As mentioned, cooperation involves more than coordinating beliefs, attitudes, and actions. The selected cooperative ob- jectives must not only be desirable but ob- tainable. In short, such selected objectives must be appropriate. The general goal of an increased number of available serials was a goal that was both desirable and ob- tainable. Specifically, how many more se- rials would be available was dependent upon the judgment of the participating faculty members and the library staffs. Thus, the goal of cooperation was general enough to gain acceptance and the proce- dure implemented to obtain the goal proved effective. The philosophical principles of coopera- tion discussed can be applied to coopera- tive ventures other than serials. Library cooperation could extend to a joint ap- proval plan for the purchase of books and other library materials. In addition, li- braries could exchange staff. Library cooperation is a rational re- sponse to better serve library user needs and to foster sharing of serials. The model or procedure discussed will, it is hoped, be instructive for other library staffs desir- ing to initiate a cooperative program. REFERENCES 1. See Joint Serials Control System Project for the Libraries of Cornell University, University of Rochester and the State University of New York at Buffalo. Phase I, feasibility study, final report. (New York : Five Associated University Libraries Syracuse, 1971). Also Carlos A. Cuadra, Final Report on Phase II, Study of Academic Library Consortia (Santa Monica, Calif.: Systems Development Corp. , 1971); Carlos A. Cuadra andRuthJ . Patrick, ' ' Surveyof Academic Library Consortia in the U.S.,'' College & Research Libraries 33:271-83 (July 1972); and Report on Library Cooperation, 1982, 4th ed. Ed. by Nancy Wareham (Chicago: ASCLA, 1982). 2. Donald Regan, Utilitarianism and Cooperation (Oxford : Clarendon Pr ., 1980), p.133 . 3. Edna Ullmann-Margalit, The Emergence of Norms (Oxford: Clarendon Pr. , 1977), p .133. 4. David K. Lewis, Convention: A Philosophical Study (Cambridge: Harvard Univ. Pr., 1969), p.16. 5. Ullmann-Margalit, p.80.