University Senate







APRIL 18, 1997


 Sen. Eben Moglen, chairman of Faculty Affairs, began the hearing shortly after 1 p.m. in 501 Schermerhorn. A total of about 30 people attended the meeting.

Introduction by Sen. Moglen: Sen. Moglen introduced the topic of the hearing: a discussion of the university's processes of ad hoc tenure review. He explained that Faculty Affairs has a grievance investigation jurisdiction to find facts and try to settle disputes arising out of the university's personnel, appointment, and promotion policies. Over the years Faculty Affairs investigators had become familiar not necessarily with the best, or even the mean of what ad hoc procedures can do, but with their most common and distressing failures. This experience had led in 1996-97 to recommendations for improvements in this crucial set of practices.

The committee had spent much of the fall semester trying to improve the grievance investigation process, so as to render some parts of the ad hoc system more accountable when something serious goes wrong. But the committee was also interested in making things go wrong less often, in avoiding failure as well as detecting and rectifying it after it happens. For that purpose, the Faculty Affairs Committee had developed a suggested set of procedural changes, and brought them to the Senate this spring.

Sen. Moglen said the committee was not at present seeking to win Senate approval of legislation, but to engage in conversation, which might reveal the University's experience with ad hoc process, and how that experience compares with the committee's own views. The committee's present proposals were a way of structuring that conversation and soliciting opinion, a process which the committee hoped would lead to improvements in ad hoc process, either through legislation or through negotiation and agreement with the Provost.

Sen. Moglen said that he would moderate a conversation that would follow a brief introduction of the proposals by Sen. James Valentini, who along with Senators Hamid Dabashi and Eugene Litwak, served as the subcommittee of Faculty Affairs on this issue. The committee would carry a record of the present hearing for discussion at the year's final Senate meeting on April 25, and probably later meetings in the next academic year. By that time, he said, the present proposals might have changed shape, perhaps drastically.

Sen. Valentini's remarks: Sen. Valentini began by talking about the motivation of Faculty Affairs in developing its proposal. The proposal derived not only from the committee's experience with grievances, but also the experience of its members and of other senators with other facets of the ad hoc process, as chairs or former chairs, deans, or ad hoc committee members or witnesses.

But what was unique about the committee's perspective was the familiarity with failures of the procedures. Sen. Valentini gave brief descriptions of some of these:

--a case in which the chair did not present all of the candidate's scholarly material to the ad hoc. Faculty Affairs investigated a grievance about this and called for a second ad hoc, which was convened;

--a case in which the chair was unclear about what his role should be in the selection of an ad hoc;

--a instance where an ad hoc member had a conflict of interest, due to a personal involvement;

--a case in which the chair told an external candidate that the ad hoc process would be pro forma, the candidate acted accordingly, and the ad hoc then turned him down for tenure, a situation embarrassing to all involved.

--a case in which the ad hoc committee's recommendation was disregarded, because of the intrusion of other people into the process;

--cases in which ad hoc presentations were not consistent.

Referring to the last problem, Sen. Valentini said the committee's proposals offer procedures to help ensure consistent presentation of tenure cases, in a system which has no single committee to evaluate all appointments.

To the question of why it should be Faculty Affairs that proposes such changes, Sen. Valentini repeated Sen. Moglen's view that this was part of the committee's job. But he added that there is a precedent for such revisions: after a grievance investigation in 1995-96, the committee had recommended modifications in ad hoc procedure, which the Provost had incorporated over the summer in the latest edition of his tenure review guidelines.

Sen Valentini said one of the committee's goals was consistent procedures, and equal rigor, for tenure reviews, a goal endorsed by a provostial faculty committee in 1992. Another was to distinguish clearly between the two essential components in the tenure review process: review and advice on tenure candidacies by a faculty committee, representing the faculty as a whole; and a final decision on that advice by the Provost.

Sen. Valentini emphasized the importance of the dual criteria involved in a tenure decision: the department has to establish its need for a tenured position in a certain area, and also to establish that a particular person is the scholar to fill that position. The committee recognized the possibility of human error in the process, and the need to provide safeguards against its impact. That's why the proposal calls for another witness, in addition to the chair, to appear before the ad hoc.

He said that the purpose of the proposals was not to undermine the Provost, but to protect him from charges or suspicions of irregularities in the process; to protect ad hoc members similarly and help assure respect for their contributions; to protect candidates from inconsistencies in procedure; and to protect the University itself from litigation and unwise tenure decisions, both positive and negative.

Sen. Valentini stressed that the committee was offering a working document, a collection of suggestions, and not a monolithic, finished piece of work. Many of the suggestions would have to be revised, and he was already convinced that some needed to be seriously revised or abandoned as unworkable. He said the hearing was the committee's first attempt to hear from the wider community, and he welcomed questions.

Discussion: Sen. Karl Kroeber referred to a letter about the proposal from Prof. Roger Bagnall, chair of the steering committee of Arts and Science department chairs, that had been distributed for the present meeting. He said that, aside from its rhetorical posturing, the letter raised two objections that were worth addressing. The first was that the proposal seemed to assume a prescriptive right to tenure (Prof. Kroeber said he did not share this perception).

Sen. Valentini responded that the proposed changes did not assume a right to tenure and were not intended to convey the impression that the proposal sought to increase the fraction of successful tenure candidates, or to put the burden of negative proof on the ad hoc committee. If the document conveyed this impression, he said, then it needed to be clarified. He said that the level of rigor in ad hoc reviews was an entirely independent issue from that of the consistency sought in the proposal.

Sen. Kroeber then mentioned the second objection in the Bagnall letter, which he thought was a cogent one--that the proposal's requirements would further prolong and complicate an already overly cumbersome tenure review process.

Sen. Valentini said it was his understanding that much of the length of the ad hoc process was connected with assembling, convening, and scheduling the committee. He acknowledged that the proposal would require extra steps, but added that the addition of a witness would relieve the chair of some of the burden of preparing a case to the ad hoc, and thereby could expedite the process somewhat. He also said that Prof. Bagnall's letter might have viewed relieving chairs' responsibilities in this way as undermining their authority and autonomy, a perspective Sen. Valentini said was understandable.

Sen. Valentini said the pursuit of procedural regularity would inevitably incur costs. He said that it was difficult to measure those costs, but that Faculty Affairs had not thought they would appreciably lengthen the process, or add unduly to the work to be done.

Sen. Moglen offered some background on the ad hoc process. Ad hoc committees, he said, are composed by the Provost's Office, with recommendations submitted by the Tenure Review Advisory Committee (TRAC), whose membership is confidential. The membership of ad hoc committees is also secret, as are the ad hoc committee's deliberations and its report to the Provost. Until the recent release of some aggregate numbers, the disposition of ad hoc recommendations to the Provost--that is, the proportion followed or turned down--had also been a secret. The secrecy of the process was relevant to the question of its cumbersomeness or administrability. Faculty Affairs members had some information about it from grievance investigations, and some more from participation as ad hoc committee members, but these sets were confined.

Sen. Moglen discussed figures that the Provost had provided at the March Senate meeting. The rate of turndowns for tenure was stable over the long term, at 13 percent of cases reaching the ad hoc stage. One interesting statistic was that in roughly 40 percent of the cases in which the present Provost had turned down tenure candidates, the ad hoc committee had voted in favor of tenure (though almost always in split votes, according to the Provost). The Provost had never provided such data before, and Sen. Moglen said that, as a student of the system, he was surprised by the 40 percent figure.

Sen. Moglen also commented on the statement, in the second paragraph of the Bagnall letter, that "a number of chairs would strongly prefer a standing committee for tenure decisions, such as other universities have, in place of the ad hoc committee system." Thus the Bagnall letter seemed to be saying that the chairs did not think radical alteration of the ad hoc system was justified, but that perhaps it should be scrapped altogether. In this context, Sen. Moglen said, people's views on the ad hoc process should be seen as occupying a comparatively open field. He said he did not see the Bagnall letter as closing the conversation, and encouraged those present to think about the issue with some daring, perhaps including the view that the ad hoc system is not the best possible arrangement.

Gillian Lindt, Dean of the School of General Studies, said that she had sat in on many ad hocs in the 1980s as Dean of the Graduate School of Arts and Sciences, and had seen the system's strengths and weaknesses. She expressed concern about what she understood to be a suggestion in the proposal that chairs do not adequately speak up for tenure candidates, and that therefore adding a witness as a representative for the candidate would somehow redress the balance. She said this view was a fundamental misreading of what happens in most cases. Typically, the involvement of the chair, as leader of the department, is very important in making the case for the candidate; furthermore, she knew of no case where the chair prepared the case alone, and said there were always faculty committees involved. Once a department decides to recommend someone for tenure, there is usually a group solidly supporting the candidate, so the idea of a separate witness was troubling. Problems do arise, she said, when a department is split, or when a member of a department that has voted unanimously in favor of a candidate comes before the ad hoc committee as a negative witness. But she didn't see how a separate witness would improve this situation.

Without trying to speak for the chairs, Dean Lindt also commented on Prof. Bagnall's statement about standing tenure committees. The chairs may have thought that the ad hoc system has problems and advantages, but that if the Faculty Affairs Committee really wanted major change, there was a different system worth considering.

She said one understandable reason why a Provost might disagree with some ad hoc recommendations is that ad hocs can't always achieve consistency of standards. Standing committees are better suited to that purpose, she said, since they have to consider a whole range of cases over an extended period of time. But Dean Lindt said she didn't see how adding a witness would help toward achieving such uniformity.

Sen. Moglen noted that ad hocs are supposed to provide a broader, University-wide view of tenure candidacies than experts within a department can provide. But if, as Dean Lindt had said, they were not as suited to the search for uniform standards as standing committees, then what were their advantages?

Dean Lindt replied that the problem she had identified arose in perhaps 10 percent of tenure cases. In a most of cases, she said, the outcome would be the same as with a standing committee. The presumed advantage of the ad hoc system is that it can achieve much more focused reviews of particular disciplines than a standing committee could. Only in a small number of cases, usually on the margins, does the problem of consistent standards arise.

Sen. Moglen asked if it was an advantage not only to have confidential records of ad hoc deliberations, but to have hidden decision makers as well.

Dean Lindt said that ad hoc membership had to remain confidential, because the pressure would otherwise be enormous if it were known beforehand.

Sen. Moglen asked if it would be a good measure of accountability for that information to be public afterward.

Dean Lindt said that if the system inspired so little confidence, perhaps it ought to be scrapped altogether. Letters of recommendation, she said, are inflated. An honest letter is very hard to get. That's why the confidentiality of the ad hoc process is important. Without it, she said, candid evaluation is rare.

Sen. Valentini responded to Dean Lindt's criticisms of the idea of a witness. He said that grievances revealed that in some cases chairs had been ineffectual or inattentive in preparing tenure cases, and it seemed wise to safeguard against human error in this important endeavor by building in a certain redundancy. With two people responsible, he said, there is always a backup. Expert witnesses are already a part of the process in many tenure presentations, but having the candidate choose the witness, as the proposal does, not only lightens the chair's burden but also gives the candidate a feeling of participating in the process, to the extent that that is possible.

It was to address the problem of incorporating a range of opinions from split departments, Sen. Valentini said, that the Faculty Affairs proposal had called for allowing any faculty member to appear before the ad hoc. Since then he had realized, even before receiving Prof. Bagnall's letter, that that was an unworkable idea. Sen. Valentini said a mechanism was still needed to allow for an organized presentation of different views to the ad hoc, and not just to the Provost after the committee has met, as is inevitable in the present system.

Dean Lindt said she thought that a witness representing the political interests of the candidate would set up an unnecessary tension within a department, at a time when the chair and the candidate have to be together. As to the problem of chairs presenting inadequate tenure cases, she said there was a kind of review system in which deans in Arts and Sciences go to the Vice President to make sure that a good case was being mounted.

Dean Lindt said she was not denying that some chairs do a better job than others. What worried her were cases in which warfare might erupt between a department and a candidate who thinks the chair is making a terrible tenure case. In a very small number of cases a candidate is confident of having earned tenure, but there is no consensus in the department. She worried about an arrangement that might inadvertently alienate candidates instead of supporting them.

Sen. Valentini asked if Dean Lindt agreed with the idea of mandating the participation of a witness, but having the chair coordinate the selection. She said she could support such an arrangement, adding that appointing witnesses in addition to the chair was already common practice, and that it was often done in consultation with the candidate.

Prof. Gauri Viswanathan (English and Comparative Literature) expressed concern about inconsistencies between ad hoc reviews for internal and external candidates. She said there was something inherently wrong with a procedure in which ad hocs are appointed after offers have been made to external candidates. In contrast to the case Sen. Valentini had mentioned, of an external candidate turned down by an ad hoc, Prof. Viswanathan was familiar with cases of external appointments in which ad hoc members feel they are serving merely as a rubber stamp. She suggested that there should be a different procedure, or at least different timing, for tenuring external candidates.

Sen. Ralph Holloway (Ten., GSAS/SS) said he had had the misfortune as a chair to defend someone at an ad hoc many years earlier, and that experience still burned. He said he thought it would be possible to improve conditions, and that was all the proposal asked for. Ad hocs varied widely depending on political and other conditions. In his 32 years at Columbia he had seen candidates who were denied tenure, and others brought in who were less qualified. Departments fall apart, go into receivership, and need to be built up. He agreed with the Prof. Viswanathan that standards were not the same for external candidates as for internal candidates.

Prof. Allen Staley, chairman of the Art History Dept, said he had missed the meeting of chairs that had led to Prof. Bagnall's letter and would have disagreed with a number of the views in the letter. He said that most of the Faculty Affairs proposals were desirable, though some fine-tuning was needed for some of the more cumbersome procedures. He said his department, since he had been chair, had almost always had a witness in addition to the chairman appear before the ad hoc committee. He said a mandated witness in some form was a good idea, and consultation with the candidate in choosing the witness was also desirable. He said his biggest concern about the system was who serves on ad hocs and how are they selected. Although the committee had clearly given much thought to this issue, but admitted he didn't understand its recommendations on this point, and asked for clarification.

Sen. Valentini said he had heard this question before, a sign that this section of the proposal had to be clearer. He said the proposal does not depart from current practice in making the Tenure Review Advisory Committee (TRAC) responsible for selecting the ad hoc committee. But there would be two new restrictions: one of the five members of the ad hoc would be chosen from a list of five submitted by the department chair; and the witness could veto one of the names submitted by TRAC. But the TRAC would be newly constituted: instead of a secret committee of seven appointed by the Provost, it would be a larger committee appointed by deans. Currently, from what Sen. Valentini understood, the department chair is engaged in the process of ad hoc selection, but not uniformly. The chair is supposedly asked his opinion about candidates for the ad hoc but does not have the right to veto any selections.

The rationale behind the changes proposed by Faculty Affairs for appointment of the members of TRAC,, Sen. Valentini said, is to emphasize the ad hoc committee review role--expressed in documents about the ad hoc system--of representing the faculty as a whole, by having its members selected by the faculty as a whole.

Prof. Staley said that confidentiality complicated the selection process for ad hoc members. It was often difficult, without a trained assistant, for a chairman to do the research to find out if a proposed ad hoc member is suitable. He saw this as a crucial issue: how could a chair make use of this power if confidentiality requirements deprived him of an opportunity to exercise it?

Prof. Valentini said that the proposal actually gives the witness, not the chair, veto power over one name submitted by TRAC. The logic behind this provision was that the witness, an expert in the candidate's field, would be the one most likely to be aware of scholarly biases, competitions, and personal influences in that field. Prof. Staley said this might be true in principle, but not necessarily in practice.

Sen. Moglen said that if expertise is the reason for preferring ad hoc to standing committees of tenure review, as Dean Lindt had suggested, then how one learns who the experts are becomes a matter of great importance. Faculty Affairs had in the past investigated a case in which the composition of the ad hoc committee was of great importance. The academic field in question was a small one, and the committee wondered how TRAC knew whom to appoint to the ad hoc. It turned out that one member of TRAC was an amateur scholar in the field, who offered the names of some experts. Those names turned out to match others that emerged in a search that went on in the Provost's office. But Sen. Moglen said he was a left a little breathless by this answer to the question of what TRAC does and how. He also wondered whether this process of knowing who the experts are and how they relate to one another is an accountable process at all if it is carried out at current levels of confidentiality. He said it might be a perfectly suitable way to converge on the names of the experts, but maybe it wasn't, and in the world of confidentiality one doesn't know.

Prof. Valentini asked Prof. Staley if he thought the idea of veto power was warranted or if he had an idea about how to handle the confidentiality issue. Prof. Staley said he didn't know that a formal veto power was necessary. He thought an informed opinion about names proposed for ad hocs would be taken seriously; but how did one get the information, given confidentiality restrictions?

Prof. Teodolinda Barolini, chair of the Italian Dept., said she had also missed the chairs' meeting on April 15 that had led to the Bagnall letter. Had she been there, she said she would have abstained on the vote on the proposal, because she did not think five minutes was enough time to discuss it. She said she agreed with a number of Dean Lindt's points, though her experience was not as extensive. She believed it was important for chairs to feel fully invested in bringing a candidate up for tenure. To set up an adversarial situation in which the chair feels the candidate might not trust him or her seemed potentially harmful to the candidate. In her experience, furthermore, candidates did not always have the best judgment about who are their best supporters.

She also said she was perplexed by Prof. Moglen's emphasis on secrecy as a basic problem, when he was surely aware that the pressure to base one's vote on feelings of friendship and collegiality and not on scholarship is intense. The secrecy, even after votes, is there to protect friendships and collegial relations. She wished he could at least rhetorically acknowledge this human dilemma, based on more than a nefarious desire to create secrecy.

Sen. Moglen said he appreciated the bearing of the comment. He said he did not support the existing ad hoc process, in part because he did not think such mechanisms of decision making are defensible. In a public university, an ad hoc system would not pass review in the federal courts, because such decision processes don't meet the requirements of the due process clause when public employment is terminate. He acknowledged that membership on a publicly known standing committee on promotion and tenure is a difficult form of political service, one of the heaviest kinds of responsibility, in human terms, that a university can impose on anyone, as the burden of public, accountable judgment can be wrenching for people. But he added position that this kind of judgment is in the end an easier burden to bear, precisely because one can point to the evidence on which judgment is based. He said that these differences of opinion depend in part on our professional might be a more common professional role for someone in the law to make public judgments than for others. He added that department chairs must often bear exactly the same burden, and he did not think people would propose being chairs in secret to redistribute the burden. He acknowledged that it was a tough problem. Public universities bear the burden, and get more accountable and defensible decision making. He believed, as one senator, that Columbia should make a similar decision.

Prof. Barolini said that she had not been advocating the ad hoc system, but trying to explain, based on her own experience, the reasons for some aspects of the process. But she hadn't thought that changing the whole system was the topic of the present hearing.

Prof. Valentini agreed, and added that Faculty Affairs had not taken the position that the ad hoc system should be abandoned. In fact, the present proposal assumed the continued existence of the system, which is why its first word was "Addendum."

A student said he was disturbed at the secrecy of tenure decisions, and at the fact that as a student he had no access to any of the tenure process.

Sen. Moglen pointed out that student evaluations are mentioned in the Code of Academic Freedom and Tenure of the University Statutes as one necessary basis for tenure decisions, and that failure to consider them is grounds for a grievance.

Sen. Valentini pointed out that the Faculty Affairs proposal makes no attempt to remove the Provost's right to overturn any ad hoc committee recommendation. He added that he was hesitant to attach excessive significance to statistics about provostial turndowns of ad hoc recommendations. Turndowns of divided (3-2) ad hoc committee votes, for example, should be viewed differently than turndowns of unanimous votes.

Sen. Moglen said the committee had no evidence that the Provost had ever overturned a unanimous ad hoc vote.

Sen. Jim Carey (Journalism), a member of the committee, agreed with Sen. Valentini that the meaning of those statistics was hard to judge. He pointed out that the vast majority of the Provost's turndowns of positive ad hoc recommendations were of votes of 3-2 or 3-1 with one abstention. He added that even in universities where there are standing tenure committees, the Provost can still overturn those committees, but because such an action would be immediately made public, there is much more pressure on him not to overturn. Secrecy therefore isolates not only faculty members from pressure, but also the Provost himself. He said one might feel differently about these two kinds of isolation depending on one's view of how centralized a system Columbia ought to be using.


Dean Lindt raised questions about the part of the committee's proposal calling for a reconstituted TRAC, to be appointed by deans. She suggested that the newly defined committee, with 18 members, would be unwieldy. She also asked on what basis the deans would make their appointments. She predicted that complaints would arise about lack of representation on TRAC.

She suggested that a stronger case could be made for making TRAC a public committee, but keeping ad hocs secret.

Dean Lindt's final point was that there are remarkable differences in degrees of consensus on excellence among different disciplines on what constitutes excellence in a candidate. There was virtual unanimity among scientists, while in the humanities there were deep divisions of opinion. She didn't see how any adjustment of witnesses could solve this problem.

Following up on Dean Lindt's suggestion of a public TRAC, Sen. Moglen asked if it should become the center for the faculty viewpoint on tenure matters. Should it collect information more publicly? Could it become the locus of faculty governance?

Dean Lindt said TRAC always included faculty input. She added that she thought Sen. Moglen was underestimating the degree of public scrutiny of the Provost. She said that ad hoc members know what goes on. If there were a sense of excessive overturning by the Provost, faculty would stand up and protest.

Sen. Moglen noted that there had been civil disobedience among some faculty members about ad hoc secrecy, and the committee had a list of those blacklisted for such acts.

Dean Lindt said she believed in the possibility of reforming the ad hoc system, and that's why she was at the hearing.

Sen. Valentini explained that the idea of having TRAC appointed by deans was to decouple the faculty advisory process from the actual decision making by the Provost. But he added that separation of faculty advice from provostial decisions was the important matter to him, not necessarily the details of how it would be accomplished.

Elbert Garcia, a student, noted the possibility that certain disciplines, like certain individuals, could be excluded unfairly in tenure deliberations.

Sen. Moglen said that he would leave the hearing to cast a tenure vote at a faculty meeting at the Law School, which does not use the ad hoc system. What he liked about the Law School system, as in most departmental decision making elsewhere in the University, was casting and defending his vote publicly, giving reasons based on shared evidence. He said he thought such accountability was the most important kind.

He adjourned the hearing at 2:23 p.m.

Respectfully submitted,


Tom Mathewson, Senate Staff