Faculty Goverannce: Election of Dept Chairs (Kunc to Tierney)

MEMORANDUM

DATE:         29 September 1999
TO:              William G. Tierney,
                    President: Academic Senate

FROM:        Joseph Kunc,
                    Chair: Engineering Faculty Council

                    Bart Kosko,
                    Vice-Chair: Engineering Faculty Council

SUBJECT:   Faculty Governance: Response to Provost Lloyd Armstrong s 16 August 1999
                    letter stating that he will not accept the Senate s May resolution on electing
                    department chairs.

SUMMARY

The President of the Academic Senate asked each faculty council chair to respond to Provost Armstrong s letter of 16 August 1999. The Provost states in this letter that he will not accept the Senate s May resolution to modify the language of the Faculty Handbook dealing with selecting department chairs. The Senate voted on May 12th to replace "and/or" with "and" in the following sentence: "The chair of a department is appointed by the dean, following selection by (1) departmental election and/or (2) thorough consultation with the full-time faculty of the department and of related departments." This memorandum gives the response of the Engineering Faculty Council (EFC) on this core issue of faculty governance. The EFC has approved this response in a majority vote of its members. The response has three parts.

The first part states the history and context of the resolution. The senators from the School of Engineering proposed the resolution. Senate Past President Lawford Anderson seconded the motion. The resolution arose from a March referendum of the engineering faculty on a like resolution. Roughly half of the tenure-track faculty voted with nearly 80% approving the measure. The Senate duly debated the resolution and considered the two arguments that the Provost gives in his letter.

The second part analyzes the Provost s letter. The letter omits the crucial "extraordinary circumstances" clause in the very next sentence of the Handbook. This clause gives the administration full discretion to ignore or even bypass any election. The letter states that faculty votes "make no sense" in "small" departments. But even appellate courts decide most appeals by the vote of three-judge panels. The letter further states that elections may "lead to a destructive polarization." But administrative appointment offers no more legitimacy than does a democratic vote and may well produce even more polarization. Neither the Senate nor the Board of Trustees invokes such a provision when they elect their members and leaders.

The third part concludes with a suggestion for future action: The Academic Senate should conduct a faculty-wide referendum on the resolution. Electing department chairs is so central to faculty governance that the Senate should hear from the entire faculty before it further discusses or negotiates this issue with the administration. The gravity of this issue makes the issue too important for any Senate committee to decide it.

I. HISTORY AND CONTEXT OF THE CHAIR-ELECTION RESOLUTION

The selection of department chairs is a core issue of faculty governance. That is why we believe that the current members of the University s faculty councils and the newly convened Academic Senate should know the history and context of the 12 May 1999 vote (Resolution 98/99-11) in the Academic Senate on the election of department chairs. We also believe it is important to make clear what the resolution does and does not state.

1. The Resolution Replaces "OR" with "AND" not with "EITHER-OR"

The Engineering Faculty Council s (EFC s) Academic Senate representatives proposed the motion to amend the Faculty Handbook to replace the inclusive-or disjunctive connective "and/or" with the conjunctive connective "and" in Section 2-1-C Academic Organization: "The chair of a department is appointed by the dean, following selection by (1) departmental election and/or (2) thorough consultation with the full-time faculty of the department and of related departments" (emphasis added). Hence the resolution amounts to a one-word change.

A key point is that the motion did not replace the "and/or" connective with an exclusive "either/or" connective whereby a vote would replace faculty consultation in the process of selecting a department chair. Nor did the resolution unconditionally omit "thorough consultation." So it is wrong to assume that voting would replace faculty consultation. Chair selection would instead depend on both. The current language of the Handbook permits deans to appoint chairs without a secret election of the department faculty. This occurs routinely in some schools and occurs occasionally in others. Voting to select department chairs is common but not uniform across schools.

A further and crucial point is that the subsequent language of the Handbook would still allow the administration to select chairs in "extraordinary circumstances." We discuss this point below in Section II.

2. The Senate Motion Arose from a Vote of the School of Engineering Faculty

An EFC member brought the issue of chair election to the EFC in the early spring of 1999 because of that member s department s past experience with the dean appointing department chairs. The EFC agreed then in an unanimous vote that the all departments in the School of Engineering should elect their chairs. The vote was not binding on the Dean because the by-laws of the EFC authorize the EFC only to advise the Dean. But the same by-laws authorize the EFC to conduct faculty polls: "IV. Role. The Council will&C. Call special meetings of the faculty or conduct faculty polls." The EFC voted to conduct such a poll because of the importance of chair selection to faculty governance and because a secret-ballot poll gives a direct measure of the will of the faculty.

The EFC carried out a school-wide referendum in March on the following yes-no proposal: "All departments in the School of Engineering shall elect their chairs by secret majority vote of all department faculty. Elected chairs shall be tenured, senior faculty and shall serve for three years. Nominations and elections shall take place during the spring semester of the third year of the current chair s term. The three-year term of the chair-elect shall commence on the following September 1st."

The Dean of the School of Engineering sent a broadcast email to all engineering faculty during the voting period to explain the language of the Faculty Handbook on chair selection and to state that "I fully recognize the need for appointing chairs who have the strong support of their department faculty. In the School of Engineering, elections have almost always been held and the individual chosen by the department normally is appointed chair. The dean s office has overseen the process when necessary."

One engineering faculty sent a broadcast email to the faculty as a follow-up to the dean s broadcast message near the end of the voting period. His email message criticized the language of the ballot measure as "very poorly formulated since it does not address the issue of recruitment of a chair from outside the department." His message further suggested the change to the Handbook that formed the basis for the Senate resolution: "If you want a revision that makes election central to the process, why not vote on changing and/or in the current Faculty Handbook wording to and ?"

The vote was overwhelmingly in favor of the EFC s proposal. Roughly half of the tenure-track faculty casts votes with nearly 80% approving the measure. The vote count was 49 yes to 13 no (with 3 votes declared invalid).

That favorable vote outcome led the EFC officers and Academic Senate representatives to propose a like measure to the Senate. They based their proposed resolution on the earlier suggestion of replacing "and/or" with "and" and on the language of the Provost s and Academic Senate s Joint Sub-Committee on the Review of Deans and Department Chairs.

Then on 14 April 1999 the EFC submitted the following resolution to the Academic Senate s Executive Committee: "Amending Chair Selection Procedure in Faculty Handbook. Whereas, the Provost s and Academic Senate s Joint Sub-Committee on the Review of Deans and Department Chairs has found that: (1) Fundamental to the success of a Chair is a capacity to coalesce interest and commitment to this objective with his or her faculty and (2) The core ingredient is that whatever steps are taken are believed by those who take time to be serving their interest, and that their stake in a common venture is enhanced by taking them ; whereas, elections add credibility and serve to imbue a Chair with increased power to coalesce interest among the faculty; therefore, it is resolved that the Academic Senate recommends to the President of the University of Southern California that the Faculty Handbook, Section 2-1-C Academic Organization be revised by replacing words and/or with the word and in the paragraph concerning procedures for appointment of department chairs."

3. The Senate Vote Followed Due Deliberation and was Entirely Valid

The Senate heard the motion to approve the chair-election resolution at its last meeting of the semester on 12 May 1999. Senate Past President Lawford Anderson seconded the motion. The minutes of that meeting show that the Senate duly debated the proposed resolution. Each senator was fully informed on the matter because each had a copy of the motion and the relevant section of the Faculty Handbook.

Critics of the resolution stated the two objections to it that deal with voting in small departments and in "polarized" departments that the Provost cited in his August 16th letter to explain why he does "not believe that an election will always be the best way to evaluate the results of that [faculty] consideration" and that we discuss in the next section. The Senate voted 12 to 8 in favor of the resolution (with 7 abstentions) despite these two objections. Hence the vote was 60% in favor and 40% opposed.

We note that a dissenting senator observed at the first meeting of the Senate this semester on 15 September 1999 that the combined 8 no votes and 7 abstentions exceed the 12 yes votes (56% no/abstain to 44% yes). That is true. But it is equally true that adding the 7 abstentions to the 12 yes votes would result in an even greater majority in favor of the motion (70% yes/abstain to 30% no). Splitting the abstention "votes" would seem the fairer way to treat them. But splitting them would not have changed the vote outcome.

Nor does a "large" proportion of abstentions invalidate a vote. A "large" proportion of American voters abstains from voting for President and for other officials because these voters do not vote at all. Indeed only a minority of the electorate elects the President. Yet no one suggests that the subsequent Presidential vote is not valid.

II. ANALYSIS OF THE PROVOST S LETTER

Provost Lloyd Armstrong sent Academic Senate President William Tierney a letter on 16 August 1999 that states that the Provost and President will not accept the Senate s proposed change of "and/or" to "and" in the Handbook. The Provost s two-paragraph letter is short enough to restate it here in full:

"On May 13, 1999, Greg Thalman sent me a resolution passed by the Academic Senate regarding a proposed Handbook amendment on Chair Selection (Resolution 98/99-11). As required by the Trustees as part of the consultative process that might lead to a change in the Faculty Handbook, I have discussed this proposed amendment with the members of the Provost s Council as well as with several other members of the senior administration, including President Sample.

"The consensus is that it is not in the best interest of the University to accept this proposed amendment to the Faculty Handbook. While everyone with whom I consulted supports the idea that chairs should only be appointed following full consideration by the faculty of their department, we do not believe that an election will always be the best way to evaluate the results of that consideration. For example, in some cases, departments are so small that elections make no sense; in others, the climate might be such that an election is likely to lead to a destructive polarization. I personally believe that the present system that provides options for evaluating faculty opinion has served us well, and will continue to do so in the future."

Senate President Tierney has asked each faculty council to respond to the Provost s rejection letter. We now do so as we consider the Provost s arguments.

1. The Letter does not consider the "Extraordinary Circumstances" Provision in the Handbook

The Provost s letter casts the resolution as requiring that "an election will always be the best way to evaluate the results of that [faculty] consideration" (emphasis added). The letter s last sentence confirms this reading of the resolution when it states that the present system "provides options" (emphasis added) and thus the proposed system in the resolution does not.

But this reading casts the quantifier "usually" as "always". For the subsequent language of the Handbook on chair selection lets the administration override the voting process in "extraordinary circumstances": "Appointments would normally be in conformity with department members judgments. If, in extraordinary circumstances, agreement is not reached after such consultation and consideration, the dean shall consult with the appropriate vice-president(s), who shall have the full and separate reports and recommendations of the departmental faculty, prior to decision" (emphasis added).

The Handbook does not explain how to tell such extraordinary circumstances from non-extraordinary ones. So this gives the administration full discretion in exercising this option. Again each senator had a copy of the complete language from the Handbook when the Senate heard and passed the resolution in May. And the senators from the School of Engineering made a point of calling out the broad discretion of the "extraordinary circumstance" clause. Changing "and/or" to "and" did not alter the other language in the Handbook. Hence the resolution does not require elections come what may.

The resolution requires that faculty will usually but not always elect their chairs. This measured approach is not trivial because it restricts the exceptions to voting to just those cases that rise to the high standard of an "extraordinary circumstance." We presume that a few controversial cases would in time come in practice to define such circumstances or at least illustrate them.

We also note that President Tierney s 1 September 1999 letter to all council chairs asking for responses to the Provost s letter also casts the resolution in terms of "always" rather than "usually" per the "extraordinary circumstances" clause: "As you will see, the resolution that passed the Senate stipulated that all department chairs must be elected." Again this is simply and crucially not so.

The Provost s letter also suggests in its "options" sentence that the resolution would allow voting only. Changing "or" to "and" does not eliminate the current Handbook option of "thorough consultation with the full-time faculty of the department and of related departments." Indeed it requires this second option. The point is that the resolution did not replace the inclusive "or" of the informal "and/or" connective with the exclusive "or" of "either/or" (in the sense of either option A or B but not both or neither). That would have been a very different resolution as would have been one that required elections only.

The logic of the resolution also allows for null votes: Departments can vote not to vote for a chair (perhaps by voting for that option or voting for "none of the above") and still maintain the legitimacy of the democratic process. The null-vote option further weakens the alleged extreme "always" status of the resolution.

The change from "and/or" to "and" in one sentence of the Handbook retains the full force of the Handbook language in the very next sentence, it maintains and requires the option of faculty consultation rather than eliminates that option, and it allows for null votes.

2. The Small-Department Argument is Not Persuasive

The Provost s letter asserts that elections are not the best way to select chairs when "departments are so small that elections make no sense." But just when is that? If a department is so small that elections make no sense then does the department itself make sense? And why would appointing the chair make any more sense?

Sample size does affect the structure of a democratic vote. A voter has increasingly more say in the election outcome as the number of voters falls. But this means that the election process only empowers voters as the number of voters falls in a department. Why would voting not make sense for such small departments? And if there are some objective cases where this holds then why not suggest the appropriate clause in the Handbook that would restrict the resolution from holding in just those cases?

The integer nature of voters lets us check cases. Suppose first that a department has only one person in it. Then either that person has no chair or that person is the chair. Voting is pointless in either case but not nonsensical per se. Suppose next that a department has just two persons in it. Voting can make sense even in this highly unlikely case. The two faculty members may both vote to elect one or the other as chair. Or they may both vote for themselves or for each other. Then the "extraordinary circumstance" clause could apply and the administration could choose for them. Or they could vote to have no vote and so again they could invite the administration to choose for them. Or they could vote on some fair (equally likely) selection method such as a flipping a coin.

The case of just three faculty clearly allows for meaningful votes. Faculty panels and committees with three members vote this way all the time. They do so with and without secret votes and despite potential "hard feelings" among the voters. Indeed both state and federal courts of appeal use three-judge panels to vote whether to grant or deny appeal requests (except in rare cases where many judges sit en banc) and thus to set often crucial legal precedents.

The cases of four, six, eight, or other even-numbered departments do allow for split votes but again here either the "extraordinary circumstances" clause can apply or the department can vote on some other tie-breaking method. Simple majority votes will always be meaningful in odd-numbered departments with three or more faculty. Again faculty routinely vote among themselves in such numbers when they assemble as subgroups in departments or in research institutes or in other academic units.

The alleged senselessness of voting nowhere emerges in this enumeration of cases. And in any case the "exceptional circumstances" clause could cover this undefined problematic case.

3. The Polarization Argument is not Persuasive

The Provost s letter further says that an election is not the best way to select a chair when "the climate might be such that an election is likely to lead to a destructive polarization."

This polarization claim is no more obvious than the prior claim about the voting in small departments. And the "extraordinary circumstances" clause could also cover it. This also suggests that appointing a chair amid such volatile politics would somehow avoid such "destructive polarization." But why should it? An appointment that overturns or bypasses an election in such a charged political atmosphere might just as well exacerbate the destructive polarization as lessen it. An election might well be the lesser of the two "destructive" outcomes.

The point is that the key issue is not the popularity of the outcome. The key issue is the legitimacy of the selection process. But any such appointment process can never have the same moral legitimacy as participatory democracy. Only the imperfect process of (secret) voting has ever succeeded at expressing the unforced consent of the governed.

And once again the "extraordinary circumstances" clause in the Handbook allows the administration to pick chairs despite or even without an election in such allegedly problematic cases.

The polarization argument also does not generalize. Why not ban voting at any level of society if the results would produce "destructive polarization"? Can we apply this reasoning to all divisive or contentious majority votes that pass 51% to 49%? Why not be consistent and apply such limits to how the faculty vote for their leaders in the faculty councils or in the Senate? Would USC s own Board of Trustees not vote for directors or other leaders if the outcome might unduly polarize members? Why should university faculty be an exception?

4. The Letter Does Not Address the Rationale of the Resolution

The Provost s letter does not address the "whereas" arguments in the language of the resolution. We stress that the first two "whereas" arguments come from the Provost s and Academic Senate s Joint Sub-Committee on the Review of Deans and Department Chairs:

"Fundamental to the success of a Chair is a capacity to coalesce interest and commitment to this objective with his or her faculty."

"The core ingredient is that whatever steps are taken are believed by those who take time to be serving their interest, and that their stake in a common venture is enhanced by taking them."

This gets to the heart of the issue: Do faculty have the right to pick their own leaders? The practice of such democracy is increasingly common at other universities and throughout the world.

There is no question that such participatory democracy strengthens faculty morale, improves faculty retention and recruitment, and sets an honorable example for the students we teach. But should the right of the faculty to vote for its faculty leaders at the department level have to rest on pragmatic arguments? Should not democracy simply be the moral default?

III. CONCLUSION: LET THE FACULTY VOTE ON THE RESOLUTION

We have shown that the arguments in the Provost s letter are not persuasive and that they do not overcome the crucial "extraordinary circumstances" clause of the Faculty Handbook that would provide for the two cases the letter puts forth to oppose the election process. We fully understand that USC is a private institution and that the Senate and faculty councils serve only an advisory role to the administration. The Provost can legally reject any resolution that the Senate approves. But such actions can easily demoralize the Senate and the faculty at large. Why should any faculty member spend the many hours of service required in any form of faculty governance if the process is wholly or even partly an ineffective one? The opportunity costs are simply too high. The collegial process requires compromise on both sides and compromise based on reasoned argument.

This lone instance of the Provost rejecting the current Senate resolution by no means proves that the Senate s advisory process is itself unreasonable. Provost vetoes can provide a healthy check on potential Senate excesses. But the current action does raise an important and more general point: What recourse does the Senate have if the administration rejects its duly processed resolutions?

The origin of the current resolution suggests a general answer for at least those vetoed resolutions that clearly affect faculty governance: Poll the faculty.

1. The Senate Should Conduct a Faculty-Wide Referendum on the Resolution

The Engineering Faculty Council put its initial resolution on electing chairs to a yes-no vote of the engineering faculty. This simple but potent action had many effects. It secured a direct measurement of the will of the faculty on a central issue of faculty governance. It let all faculty directly participate in the EFC s decision-making process. And it gave the council a mandate to share not only with its dean but with the Senate and the administration.

We suggest that the Academic Senate do the same: Conduct a yes-no faculty-wide vote on the resolution. Such a focused referendum on an issue that affects every faculty member would give the Senate a direct and statistically significant measure of the will of the entire faculty. The proportion voting or abstaining would be meaningful. The vote result itself would be meaningful and might strengthen or weaken the Provost s claim about voting in small or polarized departments. This information would in turn help the Senate decide whether to pursue or drop further negotiations on the matter with the administration.

The crucial point is that no Senate committee should have the power to decide this issue or any other issue that arises from a duly processed Senate resolution. The very gravity of a veto on such an important issue of faculty governance warrants a direct and immediate response from the entire faculty. We should not only inform our faculty peers. We should consult them.