UCU NEC Election Position: UCU for (a Very Specific Issue in) Academic Freedom

UCU for Academic Freedom (which, as I will briefly explain, is a terribly misleading name) is a group which offers endorsements based on a pledge to support something like a 2019 motion about academic freedom specificallywith respect to “Gender Critical” feminism which (in general) seems more accurately termed trans-exclusionary (e.g., trans women should be separated from cis women in various ways) or perhaps trans-skeptical (e.g., trans women aren’t women).

As an NEC member, I definitely would not support a motion like HE32 in spite of being a pretty die hard and feel uncomfortable supporter of academic freedom. So I don’t have their endorsement 🙂 My fellow UCU Commoners re-upted a very good statement in response and I was happy to sign and endorse it and esp glad to see folks from other slates and groupings join it as well.

Sometimes, defending academic freedom requires defending reprehensible people or folks holding ridiculous or nasty views. I am committed to that! But this motion isn’t focused on a specific case of, say, an academic being fired for their personal political views (however abhorrent). Indeed, it seems designed to weigh in on a political dispute on, what I firmly believe to be, the wrong side while pretending to be about procedural issues.

This is not a group concerned with the massive attack on academic freedom we’ve seen from the government for quite some time culminating in a new governmental post designed to aggressively control academic research and teaching to governmental ends (actually quite close to the assault on woman’s studies in Hungary). They aren’t talking about redundancies notionally focused on “rebalancing” which is increasingly being normalised in the UK (and fundamentally undermines academic freedom even very narrowly construed).

I’ll work with all sorts of people and prefer to be charitable to all union members, but that doesn’t mean I’ll support things transparently in bad faith. Building good things on rotten foundations is very hard as witnessed by some attempts to amend the motion, e.g.,

HE32A.1  Higher education committee

Add at end:

reaffirm that the rights of trans people and women are complementary

This implies that trans people and women are not overlapping groups (i.e., that trans women are not women). This is not a neutral formulation!

So bad substance and bad form. I’m glad not to be endorsed by them. I’m glad on academic freedom grounds since I think they are a net detriment to it.

Move Toward Fairer Lotteries

The twin demons of meritocracy and illusion of control conspire to make tons of nonsense work for us all and to produce crappy outcomes. We have to allocate scarce resources all the time with attention being one of the scarcer. (Scarcity, or perception thereof, is one of the more valuable.) To pick a few with daily salience for me, we have university admission slots, academic jobs, paper acceptances, citations, paper reads, grades, and funding. Our current way of allocating these require enormous and escalating effort from everyone. The challenges of combating fraud and gaming alone are enormous and demoralizing. Combating gaming is generally not a productive activity. It does produce a net positive. Meritocracy tells us that rewards (a slice of the scarce resource) should be allocated in some positive proportion to merit. Illusion of control tells us we can assess that merit with high levels of accuracy and precision if we put sufficient effort into it. One key problem is the latter is wrong in an enormous number of ways. “Merit” is often incomparable, indiscernible, or easily confused with other things. As we metricise our judgements of merit, things get worse as people try hard to optimise for those metrics. In many, if not most, cases, optimising for the metric doesn’t improve merit so all that effort is dead loss. (Indeed, since optimising for the metrics tends to be unpleasant, the negatives grow high.) Furthermore, the metrics we devise tend to be hard to apply and to be correlated with problematic properties. People of high socioeconomic status have the resources to optimise for the metrics so overperform on them. If we’re selecting for high SES, let’s just do that! Lying about it tends to make SES based injustices worse. One key justification for academic freedom is that we don’t know, with high confidence, what research lines will work out so we need to support a fairly high degree of risk taking. Yet we routinely see disciplines get “stuck” because of bonkers resource allocation policies. Consider this article on the “Alzheimer’s ‘cabal’“:

In more than two dozen interviews, scientists whose ideas fell outside the dogma recounted how, for decades, believers in the dominant hypothesis suppressed research on alternative ideas: They influenced what studies got published in top journals, which scientists got funded, who got tenure, and who got speaking slots at reputation-buffing scientific conferences.

We need to be careful here. Some sort of gatekeeping is necessary. We are talking about scarce, valuable resources and there are a ton of scammers who’d love access to them. But consider:

Nearly half a dozen journals rejected Itzhaki’s paper before it was accepted by the Journal of Medical Virology, not a bad journal but not a leading one. A frequent reason top journals declined to publish her papers, as they did those of other amyloid skeptics, was previous rejections. As one peer reviewer wrote about a funding proposal Itzhaki submitted in 2010, “very few [of your] papers have appeared in the most highly regarded journals.”

Argh argh argh! This is an obvious abdication of responsibility by that reviewer! Of course, relatively unpopular ideas have trouble getting published in the best venues! And those who don’t get published in the best journals don’t get as much funding. Which makes it harder for them to produce work that gets published in the best journals. I do appreciate that that grant reviewer needs some sort of time savers to manage the review load. But this is saving the wrong effort. It gets worse:

Harder to understand is why drug companies embraced the dogma even after the repeated failures of experimental drugs based on it, which has cost them billions of dollars. A longtime pharma scientist who recently joined a biotech startup offered one explanation: If company executives greenlight the development of an amyloid drug and it fails, they don’t lose their jobs because “the smartest guys in the room, meaning academia, said this was the way to go,” he said. “But if you greenlighted a different kind of Alzheimer’s therapy, and it failed, good luck with your career.”

Why should ones career in funding and guiding research be secure only if you find the right sort of failure? We’re not talking about bonkers alternatives here presented by kooky people! So we have embedded, interlocking biases, multiple sorts of gatekeeping (wherein the worst are foundational), and a lot of randomness which hardens. We need to explore better mechanisms. My favourite is to incorporate straight up lotteries as much as possible. It’s often cheap and easy(ish) to filter out really bad stuff. At least, allocating more effort there is reasonable. I still do a better job filtering out spam than my spam filters, after all. Once you pass a threshold, just allocate things by random chance. It’s super cheap to implement. It pushes the influence of various biases to one place (the initial filtering) which is easier to analyze and it produces more churn. It lowers the overvaluing of success in acquiring the resources (“Oh, you went to Harvard? You got lucky then”) and it’s hard to see that it’ll produce worse outcomes in most cases. If that makes you uncomfortable, add some weighting. Rank applicants (or papers) on a simple scale (3 or 4 points) and give higher ranked applicants extra chances for selection. In so many cases, we get down to a pool of candidates who are functionally indiscernible. And yet the twin demons force us to pretend we can discern. Let’s be more honest and flip a coin. (And good god let’s separate feedback from acceptance!)

Steven Salaita’s (Academic) Freedom

It seems that the current state of Salaita’s academic freedom is freedom from academia:

Even as I complicate honest work, I’m aware of how indebted I am to the notion.  It guided my exit from academe and my rejection of the pundit economy.  I’ve always overvalued recalcitrance, a compulsion, as I understand it, to vigorously avoid situations that require ass-kissing, usually resulting in significant reputational harm.  Since elementary school, I’ve searched for a space where I could conform to my surroundings without feeling unmoored from an inner sense of decency.  That space, it turns out, is equivalent to the volume of a school bus.

I pitched honest living to my parents when I told them about the new job.  Despite being aware of academe’s ruthless memory, they hoped that I’d one day be a professor again.  They probably still do.  In a better world, my redemption would happen in the United States.  I wanted to quell that expectation.  “Even if Harvard offered me a job I’d say no,” I proclaimed with earnest hyperbole.

They feigned support but didn’t believe me.  I understand why.  It’s hard to imagine coming of age in reverse.  Hollywood doesn’t make inspirational movies about struggling to overcome material comfort.  We don’t aspire to the working class.  Personal fulfillment occurs through economic uplift.  We go from the outdoors to the office, from the ghetto to the high-rise, from the bar rail to the capital.  That’s the dream, to become a celebrity or a tycoon or, in humbler fantasies, a bureaucrat.  But forward progress as material comfort is cultivated through the ubiquitous lie that upward mobility equals righteousness.  Honest living is a nice story we tell ourselves to rationalize privation, but in the real world money procures all the honesty we need.

For immigrants, these myths can be acute.  I could see my parents struggling between a filial instinct to nurture and an abrupt recognition of their failure.  My mom, a retired high school teacher, seemed interested in the logistics of transporting students, but my dad, the original professor, clenched his hands and stared across the table.  It’s the only time I’ve seen him avoid eye contact.

I’ve encounter, recently, students who are are studying computer science not because they like it, or chose it, but because their parents picked it for them. These students, even the good ones, are very unhappy.

I’ve been trying to communicate to PhD students that a PhD program is just one form of life and not one that everyone (or most people!) find worth while. It’s hard to accept that a PhD doesn’t make you better and that failing to get one doesn’t make you worse. It’s not clear that the best way to become a good researcher is to “finish your PhD”. For many fields, PhD projects are not very much like how you go on to do research.

Salaita remains an interesting case. His academic career was definitely and perniciously destroyed on false and illegitimate grounds. He was not a marginal figure but a minor “star” professor. His case was an egregious violation of both the spirit and letter of academic freedom. I should think that the standard advice is to make sure any new contract was fully enacted before resigning one’s old position, just in case.

So he was wronged but seems content with where he ended up. This demands respect. But it’s also a bit tricky to get right.

Of course, less dramatic versions of his story are playing out all the time. The adjunct crises is ongoing. Unemployable people with huge student debt are all over. We have to think hard about how higher education is going to work in the coming decade and to make sure we aren’t causing unnecessary harm.

Some Results On Student Evaluations

Erik posted about student evaluations last week and I took the opportunity to bring up the  famous Air Force Academy study. (the paper; an accessible blog post about it).

The setup is extraordinary:

Prior to the start of the freshman academic year, students take course placement exams in mathematics, chemistry, and select foreign languages. Scores on these exams are used to place students into the appropriate starting core courses (i.e., remedial math, Calculus I, Calculus II, etc.). Conditional on course placement, the USAFA registrar employs a stratified random assignment algorithm to place students into sections within each course/semester. The algorithm first assigns all female students evenly throughout all offered sections, then places male-recruited athletes, and then assigns all remaining students. Within each group (i.e., female, male athlete, and all remaining males), assignments are random with respect to academic ability and professor.12 Thus, students throughout their 4 years of study have no ability to choose their professors in required core courses. Faculty members teaching the same course use an identical syllabus and give the same exams during a common testing period. These institutional characteristics assure that there is no self-selection of students into (or out of) courses or toward certain professors.

They focused on math classes because the grades are highly normalised:

The integrity of our results depends on the percentage of points earned in core courses being a consistent measure of relative achievement across students. The manner in which student scores are determined at USAFAparticularly in the Math Department, allows us to rule out potential mechanisms for our results. Math professors grade only a small proportion of their own students’ exams, vastly reducing the ability of “easy” or “hard” grading professors to affect their students’ scores. All math exams are jointly graded by all professors teaching the course during that semester in “grading parties,” where Professor A grades question 1 and Professor B grades question 2 for all students taking the course. These aspects of grading allow us to rule out the possibility that professors have varying grading standards for equal student performance. Hence, our results are likely driven by the manner in which the course is taught by each professor.

In some core courses at USAFA, 5–10 percent of the overall course grade is earned by professor/section-specific quizzes and/or class participation. However, for the period of our study, the introductory calculus course at USAFA did not allow for any professor-specific assignments or quizzes. Thus, potential “bleeding heart” professors had no discretion to boost grades or to keep their students from failing their courses. For this reason, we present results in this study for the introductory calculus course and follow-on courses that require introductory calculus as a prerequisite.

This is really amazing! It’s pretty close to a highly controlled experiment!

They found a pretty strong effect from instructor quality:

The USAFA’s comprehensive core curriculum provides a unique opportunity to test how introductory course professors affect follow-on course achievement free from selection bias. The estimate of 2 Var (lj 1) is shown in row 2, column 2 of table 4 and indicates that introductory course professors significantly affect follow-on course achievement.19 The variance in follow-on course value-added is estimated to be 0.0025 (SD p 0.050). The magnitude of this effect is roughly equivalent to that estimated in the contemporaneous course and indicates that a one standard-deviation change in introductory professor quality results in a 0.05-standard-deviation change in follow-on course achievement.

The striking bit (for me) is their examination of student evaluation and grades:

Next, we examine the relationship between student evaluations of professors and student academic achievement as in Weinberg, Hashimoto, and Fleisher (2009). This analysis gives us a unique opportunity to compare the relationship between value-added models (currently used to measure primary and secondary teacher quality) and student evaluations (currently used to measure postsecondary teacher quality).

…In column 1, results for contemporaneous value-added are positive and statistically significant at the.05 level for scores on all six student evaluation questions. In contrast, results in column 2 for follow-on course value-added show that all six coefficients are negative, with three significant at the .05 level and three significant at the .10 level .Since proposals for teacher merit pay are often based on contemporaneous teacher value-added, we examine rank orders between our professor value-added estimates and student evaluation scores. We compute rank orders of career average student evaluation data for the question, “The instructor’s effectiveness in facilitating my learning in the course was,” by professor… As an illustration, the calculus professor in our sample 1 who ranks dead last in deep learning ranks sixth and seventh best in student evaluations and contemporaneous value-added, respectively.

Our findings show that introductory calculus professors significantly affect student achievement in both the contemporaneous course being taught and the follow-on related curriculum. However, these methodologies yield very different conclusions regarding which professors are measured as high quality, depending on the outcome of interest used. We find that less experienced and less qualified professors produce students who perform significantly better in the contemporaneous course being taught, whereas more experienced and highly qualified professors produce students who perform better in the follow-on related curriculum. Owing to the complexities of the education production function, where both students and faculty engage in optimizing behavior, we can only speculate as to the mechanism by which these effects may operate. Similar to elementary and secondary school teachers, who often have advance knowledge of assessment content in high-stakes testing systems, all professors teaching a given course at USAFA have an advance copy of the exam before it is given. Hence, educators in both settings must choose how much time to allocate to tasks that have great value for raising current scores but may have little value for lasting knowledge.

And the key bit:

Regardless of how these effects may operate, our results show that student evaluations reward professors who increase achievement in the contemporaneous course being taught, not those who increase deep learning. Using our various measures of teacher quality to rank-order teachers leads to profoundly different results. Since many U.S. colleges and universities use student evaluations as a measurement of teaching quality for academic promotion and tenure decisions, this finding draws into question the value and accuracy of this practice

Now, in the comment thread, people pointed out that this result, while strong, was narrow (restricted to fairly low level math and “adjacent” courses). These facts are related, of course. By focusing on the data with the fewest confounders, they narrowed the scope but increased the strength. However, there have been subsequent students. For example, “Evaluating students’ evaluations of professors“:

The empirical analysis is based on data for one enrollment cohort of undergraduate students at Bocconi University, an Italian private institution of tertiary education offering degree programs in economics, management, public policy and law. We select the cohort of the 1998/1999 freshmen because it is the only one available where students were randomly allocated to teaching classes for each of their compulsory courses.
The students entering Bocconi in the 1998/1999 academic year were offered 7 different degree programs but only three of them attracted enough students to require the splitting of lectures into more than one class: Management, Economics and Law&Management. Students in these programs were required to take a fixed sequence of compulsory courses that span over the first two years, a good part of their third year and, in a few cases, also their last year.

The exam questions were also the same for all students (within degree program), regardless of their classes. Specifically, one of the teachers in each course (normally a senior faculty member) acted as a coordinator, making sure that all classes progressed similarly during the term and addressing problems that might have arisen. The coordinator also prepared the exam paper, which was administered to all classes. Grading was delegated to the individualteachers, each ofthem marking the papers ofthe students in his/her own class. The coordinator would check that the distributions were similar across classes but grades were not curved, neither across nor within classes.

They also looked at evaluation/grade correlations:

In this section we investigate the relationship between our measures of teaching effectiveness and the evaluations teachers receive from their students. We concentrate on two core items from the evaluation questionnaires, namely overall teaching quality and the overall clarity of the lectures.

The key hit:

Our benchmark class effects are negatively associated with all the items that we consider, suggesting that teachers who are more effective in promoting future performance receive worse evaluations from their students. This relationship is statistically significant for all items (but logistics), and is of sizable magnitude. For example, a one-standard deviation increase in teacher effectiveness reduces the students’ evaluations of overall teaching quality by about 50% of a standard deviation. Such an effect could move a teacher who would otherwise receive a median evaluation down to the 31st percentile of the distribution. Effects of slightly smaller magnitude can be computed for lecturing clarity.

Finally, I did peek at a 2017 meta-analysis which overrides earlier meta-analyses. The abstract:

Student evaluation of teaching (SET) ratings are used to evaluate faculty’s teaching effectiveness based on a widespread belief that students learn more from highly rated professors. The key evidence cited in support of this belief are meta-analyses of multisection studies showing small-to-moderate correlations between SET ratings and student achievement (e.g., Cohen, 1980, 1981; Feldman, 1989). We re-analyzed previously published meta-analyses of the multisection studies and found that their findings were an artifact of small sample sized studies and publication bias. Whereas the small sample sized studies showed large and moderate correlation, the large sample sized studies showed no or only minimal correlation between SET ratings and learning. Our up-to-date meta-analysis of all multisection studies revealed no significant correlations between the SET ratings and learning. These findings suggest that institutions focused on student learning and career success may want to abandon SET ratings as a measure of faculty’s teaching effectiveness.

I mean, it’s brutal:

In combination, our new up-to-date meta-analyses based on nearly 100 multisection studies, as well as our re-analyses of the previous meta-analyses make it clear that the previous reports of “moderate” and “substantial” SET/learning correlations were artifacts of small size study effects. The best evidence − the meta-analyses of SET/learning correlations when prior learning/ability are taken into account − indicates that the SET/learning correlation is zero. Contrary to a multitude of reviews, reports, as well as self-help books aimed at new professors (a few of them quoted above), the simple scatterplots as well as more sophisticated meta-analyses methods indicate that students do not learn more from professors who receive higher SET ratings.

And student evaluatons are a substantial part of the Teaching Evaluation Framework.

So that’s really bad!

When we throw in gender (and other) biases, it seems clear we have a huge problem.

Unionising Graduate Students

Erik Loomis at LGM has an awesome series called “This Day In Labor History”. For Feb 17th, he wrote about the Yale graduate student strike of 1992. It’s definitely worth reading (and please check out this amazing comment).

Of course, an interesting feature of graduate student organisation is opposition by academic staff:

Graduate student unionization has long been controversial on college campuses. Are graduate students primarily students or apprentices? The answer should be obvious that all graduate students getting paid for work are workers, but you would be surprised how many liberal faculty members simply cannot accept this idea.

Cf the NYU graduate student dispute where Bohgossian and Velleman cover themselves with shame. Here’s a good summary about the (very bad) position and a rebuttal:

The second line of NYU administration defense (or, as Velleman seems to suggest below, a different way of making out the first line of defense) is that grad student unionization institutionalizes a relationship that shouldn’t be institutionalized.  As Boghossian says: “Our basic impulse is not to want to lock into place a relationship to our graduate students, whom we treat as developing colleagues, that considers them to be laborers, and we don’t want to institutionalize that relationship”.

A difficulty with Boghossian’s presentation here was pointed out by Manuel Cabrera in the comment thread on Jason’s post:

Assume that it’s true that unionization will alter the relationship between the TAs and those who employ them. Assume that the relationship changes in the following way: tension increases between TAs and their employers. The thing is: the TAs aren’t employed by the professors in their departments, but by the university. Their professors are fellow employees, not their employers. The TAs aren’t appealing to their advisors and mentors to unionize, they’re appealing to the entity that employs both groups: the university. The mentor-mentee relationship between professors and students that is, indeed, so important to graduate education need not shift fundamentally even if the employer-employee relationship does, for they are two different relationships between two different pairs of groups.


and in a later comment, Velleman says:

The dispute over whether graduate assistants are “employees” is not about whether they are performing services for compensation. Of course they are […] The question is whether their relations with the university are best regulated by categories and procedures that are appropriate to relations between labor and management in, say, the auto industry. I have already given clear illustration of this question in my previous comment, showing how the concept of seniority can be disruptive to academic decisionmaking.

When a graduate student is not given a teaching assignment for the coming semester — say, because of failing to make adequate progress toward his or her degree — is that best conceived as a “layoff”? Is the student’s return to the classroom best regulated by the sort of rules governing the return of laid-off auto workers to the assembly line? Should the student’s appeal of the decision be referred to a labor mediator? The answers to these questions are clearly and unambiguously no. That’s the sense in which graduate student instructors are not “employees”.

These are the questions that arise about the graduate students’ relations with their departments.

The Academic’s Obligation to Support Graduate Student Unionisation

I believe academics — especially if they supervise graduate students — have a strong obligation to support graduate student unionisation. I don’t think this is a hard question! Indeed, it think it’s strongly overdetermined on various moral, prudential, and pedagogic grounds. (These intertwine, obviously.) Some considerations:

  1. Graduate students have common interests, have little power in their work life (either formal or informal), and are formally and informally compensated by their university for a wide range of work.
  2. Existing grievance procedures are often (nearly always?!) developed and administrated without graduate student substantive input. Furthermore, the arbitrator of any grievance is exactly the organisation that bears ultimate responsibility for a grievance.
  3. We are supposed to be training graduate students for their future career. It is commonly understood that such training includes professional development (e.g., how to negotiate a job offer,  how to interview and mock interviews, how to write grants, etc.).

1 and 2 strongly point to the need and justification for graduate students banding together to negotiate with and counterbalance the university. Graduate students form a distinct group with many common interests. They should be allowed, indeed, encouraged to work together to articulate and defend those interests. Any argument against this has to show why graduate students (who are extremely vulnerable) do not deserve the basic move to collective bargaining. We encourage graduate students to support each other in all sorts of other ways (e.g., mentoring). It requires a strong argument that one which gives them actual power is counterproductive for them.

3 is a clincher if you think that academics need unions. (I’m happy to substitute “professional organisations” in there for union, but I’m strongly pro-union.) If you aren’t prepared to make an argument against unions altogether, you are stuck I think. Even if you think 1 and 2 are overridden by…something…the point is that we want to train grad students to be good members of our profession. And that means giving them the opportunity to work with a union.

Academic Freedom, Shared Governance, and HR Policies

A core feature of effective academic freedom is devolution: The most academic control should be at the most local level. Thus, individual instructors and researchers are free to chart their own teaching and research. Evaluation is primarily done by those most expert and experienced in an area (i.e., peers), thus promotion or censure should be managed at a very local level. Similarly, organisational decision (e.g., who chairs the department/is head of school) should be a bottom up decision in the normal order. Similarly for hires. Similarly for program structure.

This isn’t to say that there are no reasonable external constraints or default constraints. It’s perfectly reasonable for the university to set the academic calendar, general exam requirements, etc. There should be substantial input from lower levels, but there is value in uniformity in some areas. These areas of uniformity should be very carefully delineated and managed. They are a regrettable necessity, not something to be sought after.

In particularly, we must be very careful about seeking “efficiency”. Efficiency is laudable in a lot of cases. But efficiency from a central perspective can impose costs on and inefficiencies at a local level and the local level is key.

For example, Manchester Computer Science used to have its own Research Office funded and staffed by us. It was relatively large as we generally have a lot of research funding. Everyone liked it because 1) it was very good with great people and 2) it in our building in a central place so you could just pop in if you had a question or issue.

Several years back the next unit up (“Faculty”) decided to reorganise all research offices and share them between Schools. So roughly every two schools gets one research office. We share ours with Math (which is not a horrible fit intellectually) but they got the office in their building. Which isn’t far, but I’ve never been to their office.

So, are there efficiencies? Maybe? I don’t see the bottom line so I don’t know. They added funding so they weren’t trying to save money, but they claimed they wanted better coverage, etc. But I don’t feel that things at that level have gotten better. We got a wonderful local research support officer, but she’s the only one I have any contact with. Things that used to be handled in a minute now can take days and I won’t know why. I’m more alienated from the process in general. Most people I talk with miss the old arrangement (though we’d like to keep the support officer).

This was a decision that was thrust upon us, against our expressed wishes. That alone is a cost. The distancing and alienation is another cost. The efficiencies of propinquity are also lost.

Autonomy and independence are things that require nurturing and a suitable environment to thrive in. Stripping people of decision making power tends to blunt that. People get less used to making decisions or withdraw from various collective activities or both. This degrades a university.

Of course, control, autonomy, and freedom can be abused in all sorts of ways major and minor. I don’t think a school should be free to discriminate on the bases of sex, race, etc. Students need rights that are enforceable against local units. Cronyism is the flip side of collegiality. The tyranny of PhD supervisors and research lab leaders should not be unchecked.

But a university functions best as a confederation not a corporation. Just as we local units should recognise the dangers of unchecked local control, so too the centre should be careful not to degrade and destroy local autonomy unnecessarily. The head of a school should be primarily answerable to the school. To do that, they must be selected by the school. The centre can ratify, but that ratification should be very light touch.

Academic Freedom from first principles

Recently there have been two, somewhat prominent, cases where a professor received a job offer from an academic department had that offer rescinded by the administration (typically without explanation). The cases have interesting parallels and raise interesting questions about academic freedom. I’ve encountered a fair bit of confusion about how academic freedom plays into all this. In this essay, I attempt to construct a model of academic freedom and its limits (acceptable and otherwise) derived, more or less, from first principles. So I don’t rely on e.g., AAUP statements except as cross checks that I’m in the right concept space. I’ll try to apply this derived notion to the cases (Salaita and McGinn) to test it out.

Some caveats:

  1. While the S&M cases are prominent and important, they are by no means the most significant impositions on academic freedom today. For example, the treatment of adjuncts (and the reduction of tenure lines) is far more significant and destructive. I’m sort of wimping out by starting with S&M, but that’s where my head has been, so that’s where I’m starting. Sorry.
  2. I don’t particularly care for Salaita or McGinn.
  3. I’m more personally sympathetic toward Salaita than McGinn (by like a ton).
  4. My judgement is that Salaita has a very strong case for his having treated unjustly. McGinn much less so.

The Academic Mission

The standard primary (at least nominal) goal of a university is the pursuit and dissemination of knowledge. Universities can have other goals, even other primary goals (e.g., social justice or, more problematically, profit), but if knowledge creation and sharing aren’t in there it’s hard to see how they are a university per se. A university can choose to emphasise one over the other (“teaching colleges” vs. “research universities”), but without the combination you just are something else. [Note: We need to construe knowledge broadly so that the creation of art in all its forms is covered.]

It is important not to prima facie constrain the form or sort of knowledge. It happens that scholarly knowledge is the primary form of knowledge generated by universities, but it’s not the only form, and part of the teaching aspect is to produce self-knowledge in students.

Given this mission, we immediately face two questions:

  1. What are the necessary means to achieving this mission?
  2. What are the most effective means to achieving this mission?

The banal answer is that we need a certain amount of resources. People have to eat while they teach or do research. Labs and classrooms cost money.

Another answer is that we need rigour, quality, skill, and so on. Sometimes this is called expertise, but there are some issues there (e.g., when “being a credentialed expert” substitutes for “has a specific capability”) . So, let’s just call it skill. We need skill in research and in teaching.

The distinctive answer answer to both these questions is free inquiry.1

Free Inquiry

Why does inquiry need to be free? Even in areas which might seem to be highly constrained as to method and topic (say, physics or biology), the space of possible hypotheses, techniques, and methods is too large and complex for exhaustive search.2 We need to winnow down our explorations to likely to be fruitful efforts. And we have a plethora of mechanisms to do this: teaching, funding, hiring criteria (and actual hiring), promotion criteria (and actual promotions), peer review, peer acclaim and scorn, public acclaim, prizes, and so on. The social mechanisms in place to direct research both in form (e.g., thou shalt publish in high ranking journals; thy thesis shall cite appropriate literature) and in content (e.g., see the history of cold fusion3).

Furthermore, one of the jobs of a PhD supervisor/advisor is to steer their students to projects that are likely to result in a successful PhD defence.4 A key job of peer reviewer is to ensure that certain quality, relevance, and significance standards are met.

All this suggests that inquiry is de facto highly constrained and such heavy constraints are embraced by the academic community.

However, history is littered with surprising scientific ideas and techniques. To grossly oversimplify, science works by a generate and test methodology: We generate theories and test them against reality. The problem is that we are often wrong and wrong headed. It’s not just that we make errors, but that we make errors at every level, including what’s worth investigating and what’s the right way to do it. Given the enormous forces pushing for conformity, we need something to counterbalance it. We need to make space for stabs into other parts of the investigatory space.5 If there is a lot of momentum in a certain direction, then other moves are both inherently risky (after all, we generally have good reasons for the current direction) and institutionally risky (the rest of the community is trying to keep every one on “fruitful” paths).

We need to mitigate the risk of groupthink and communal misguidedness. The way we do this is by providing some mitigation of the individual (or sub community) risk and some encouragement to go off the common paths. Some example mechanisms:

  1. Tenure and other job protections.
  2. Conscious efforts to support diverse views.
  3. An ethos of critical reflection, exploration, and devil’s advocacy.
  4. Less constrained fellowships such as MacArthur awards, etc.
  5. Opener fora such as workshops where we relax e.g., rigour standards with an eye to exploring new ideas
  6. Strong socialisation into a culture of academic freedom

It’s instructive to note that the formal mechanisms are relatively few. In the US, tenure is an important one, but it is very minimal. Tenure is just a contractual provision that you cannot be fired except for cause where cause is restricted to fairly minimal levels of competency. Classically, esp. in the humanities, if you can teach a couple of courses, you can keep your job. Thus, pursuing a failed or even bonkers line of research does not mean you’ll be fired.

This is a ridiculously minimal protection! It covers less than a third of academic staff in the US, some portion of those are merely “tenure track”. The pre-tenure track is brutal with a common bit of advice being not to rock the boat in anyway before you get tenure (!). In other words, there are all sorts of personal risk that tenure doesn’t protect against. You can get some more protection by being otherwise successful (and there, it probably should be lots of grants, at least in the sciences).

If we think more broadly than the sciences, the forms of knowledge become more diverse and the idea that we might be able to restrict the “freedom” to something like one’s academic speciality becomes untenable. We don’t know what will work at all or work for a given person.

My favourite description of this comes from Robertson Davies’ novel, Fifth Business:

Yet often, usually at about three o’clock in the afternoon when the air grew heavy, and scholars at nearby desks were dozing over their notes, I would think: Dunstan Ramsay, what on earth are you doing here, and where do you think this is leading? You are now thirty-four, without wife or child, and no better plan than your own whim; you teach boys who, very properly, regard you as a signpost on the road they are to follow, and like a signpost they pass you by without a thought; your one human responsibility is a madwoman about whom you cherish a maggoty-headed delusion; and here you are, puzzling over records of lives as strange as fairy tales, written by people with no sense of history, and yet you cannot rid yourself of the notion that you are well occupied. Why don’t you go to Harvard and get yourself a Ph.D., and try for a job in a university, and be intellectually respectable? Wake up, man! You are dreaming your life away!…

Despite these afternoon misgivings and self-reproaches I clung to my notion, ill defined though it was, that a serious study of any important body of human knowledge, or theory, or belief, if undertaken with a critical but not a cruel mind, would in the end yield some secret, some valuable permanent insight, into the nature of life and the true end of man….The only thing for me to do was to keep on keeping on, to have faith in my whim, and remember that for me, as for the saints, illumination when it came would probably come from some unexpected source.

Taking such risk needn’t (and really isn’t!) be consequence free: If you pursue unfashionable lines, you will have trouble with your career. You might lack for colleagues, endure scorn and mockery, and end up with little to show for your efforts. You might be profoundly wrong or merely unrecognised. But being wrong is the natural state of someone seriously engaged in inquiry. We are nothing if not fallible.

People are strange. We can be brilliant in one moment or area and bonkers everyone else. Linus Pauling was undoubtably one of the most significant and productive scientists in history, but he was a big ole quack about vitamin C. So, we should be careful about excluding people from community of inquirers because of, let us neutrally say, quirks, whether they are quirks in their field but especially if they are from outside their field. (“Field” is, itself, not a crisp, static notion.)

Politics is both something we study and something that affects what and how we study. It is also something that is easy to get wrong in a variety of ways.

Obviously, for all the propaganda, no university is organised around a extreme principle of academic freedom. We put lots of barriers against (what we think are) unfruitful or kooky ideas. And a good deal of this is appropriate! Remember, the goal isn’t freedom per se, but success in the creation and dissemination of knowledge. So, while people are amazingly fallible, we are also the only court available. There are conflicts between people’s academic freedom (e.g., funding is often zero sum, so if we give it to one person we necessarily deny it to others; instructor’s freedom to innovate pedagogically is restricted by the fact that students need certain conditions for effective learning; medical researchers are restricted in their experiments on both humans and animals even when less restricted mechanisms might be more effective). The name “academic freedom”, itself, suggests that we aren’t protecting people at large (though universities should make it easy to enter and benefit from support for free inquiry).

Given that we don’t have radically free inquiry, we need to be extremely cautious about the restrictions we do have and any new restrictions we introduce. A good deal of academic freedom (in the sense of intellectual risk taking and support for risk taking) comes from socialisation into a professional ethic (or, at least, the show of it). This socialisation is extremely fragile, is steadily being attacked, and in many ways is diminished in, at least, the US and the UK.6 Thus, some obvious moves:

  1. It’s not enough to give lip service to academic freedom, we must enact and defend it even when it hurts.
  2. A restriction of academic freedom should be primarily justified by an overall gain in academic freedom with the burden of proof being against an active restriction.

The second condition supports the “whole picture” standard of evaluation of hires and tenure decisions. One way we can expand freedom is allowing failures to be balanced out by other successes. Thus, if someone pursues a faltering line of research, they should be able to be promoted at least partially on the basis of their excellent teaching. Similarly, while we might determine that certain constraints on political expression in some classrooms are appropriate (intro classes are potentially more restrictive than advanced graduate seminars), we should be extremely wary about making decisions based on external political considerations or unpopular political expression.


I trust it’s obvious at this point that I can derive my favoured position: The Salaita unhiding is a gross and unjustified violation of both his and the hiring department’s academic freedom. The tweets were far from the classroom and his peers determined that his whole picture was highly favourable.7 The nominal rationale (that some students might be made to feel uncomfortable in his classes because of their political views) is strongly undermined by his prior record. The actual motivation (placating donors) is precisely the sort of thing we should resist, even though it hurts. The power of money is already tremendous in universities and not letting funders control every appointment is an extremely modest restriction on it.

What’s more, the chilling effect is obvious and powerful: Don’t express political opinions that might offend a donor unless you are 1) happy in your current position for the rest of your career and 2) extremely careful about any contractual stuff. Since among the things that offend wealthy donations is freaking climate science this pretty much exposes us all.


McGinn’s case is structurally similar: he was given an offer by a department but an administrator over-ruled it. As much as it pains me to say it, we have to start with a presumption that this is not a good thing. We have fewer details on the rationale, but deference to local academics is a powerful restraint on administrative violations of academic freedom.

That being said, there’s considerable room to critique the East Carolina hiring committee (the CHE article is behind a paywall so I requote from here):

Michael Veber, an associate professor of philosophy who led the search committee at East Carolina that chose Mr. McGinn, says he didn’t put much stock into what went on at Miami. “After reviewing the evidence, Miami never even accused him of harassment,” says Mr. Veber. “So I don’t see how anyone could justify denying him a position because of any of that.”

Nicholas Georgalis, a distinguished research professor of philosophy at East Carolina, says his department wanted to hire Mr. McGinn because of his eminence as a scholar. “His publication record is amazing, he is a rare public intellectual, and his letters of recommendation from top people in the field were extraordinary,” says Mr. Georgalis, who has been at East Carolina since 1973 and says he can’t remember administrators ever turning down a philosophy-department hire before this. “We felt this was a great opportunity.”

As for the reaction to the sex-harassment allegations Mr. McGinn faced at Miami, Mr. Georgalis says “a lot of it was hysteria, reactions based on rumor.” He adds, “I evaluated him based on his academic credentials, which are stellar.”

Here’s a case where the “whole picture” is relevant. I’ll leave it open whether a whole consideration of the McGinn record would justify his appointment (I tend to think not), but McGinn resigned his prior position for things where a teacher’s freedom is weakest: intimacy (whether benign or malign) with students. What’s more, the public record McGinn himself put forth is gives a troubling picture of his judgement. For one, he was actually accused of unprofessional conduct:

Pursuant to Section B4.9 (d)(ii) of the Faculty Manual, the University believes that Professor McGinn’s conduct is unprofessional due to the amorous relationship that developed between a senior faculty member and his student [my italics]…. The “relationship” violated the Faculty Manual’s policy on Consensual Amorous, Romantic or Sexual Relationships. Pursuant to that policy, ‘a faculty member who engages in amorous relationships with a person over whom he or she has evaluative authority without taking steps necessary to resolve the conflict, including reporting such a relationship at the earliest opportunity, may be subject to disciplinary action under the policies embodied in the Faculty Manual governing charges of unprofessional conduct.’… In this case, there is no dispute that…prior to and during the duration of the Independent Study Course and the research assistantship, Professor McGinn and Ms. X had a romantic relationship…

McGinn admits he had such an unreported amorous relationship, even though he thinks his conduct was professional. (Let us note that if everything were hunky dory about the relationship, the student wouldn’t have filed a complaint. It’s still theoretically possible, given what is public, that it was a wrongfully motivated accusation, but esp. given McGinn’s display, it doesn’t seem likely.)

At the very least, the whole Genius Project (wisely hidden on the main site) is a perfect example of where we should not take academic freedom in spite of McGinn’s protestations:

The student (hereafter NN) and I were engaged on what we called “the Genius Project”. The purpose of the genius project was to make NN into a truly original and outstanding young philosopher (one who could expect to find an attractive job later). Part of this project involved techniques for encouraging unconventional thinking, and the concept of “taboo-busting” was deemed helpful towards this end. We had developed a very open and candid style of communication, and were in contact extremely frequently. (She came to my house several times to play tennis and go paddle boarding, as well as talk about the work we were doing.)…

In this spirit I sent NN two short email messages, spaced over three months, which contained some (mild) sexual content, which was related to the seminar of mine NN had attended and which was relevant to work we were doing together. This content pertained to the hand in relation to human evolution and human life (including sexual life), and referred back to material discussed in the seminar I gave and which NN enthusiastically attended. These emails were received in the spirit in which they were intended (certainly no complaint was voiced about them), and they gave rise to some mild amusement between us over the months. They were quite banal. Many other personal and professional matters were discussed in our email correspondence, as well as our very frequent meetings; these two were exceptional and designed to achieve a specific pedagogical end. In no sense were they intended (or interpreted) as attempts to elicit anything from the recipient (except perhaps a chuckle).

To impose disciplinary action on the writer of these two emails would be a clear infringement of academic freedom and freedom of speech. I judged that these emails were in line with the project on which NN and I were (consensually) engaged, and I think they served their purpose (taboo-busting—though they described rather mundane matters). Most of the genius project took a more conventional form, but it is within this context that they should be interpreted. They were not just gratuitous snippets of risque prose, sent out of the blue. I believe that had the genius project continued it would have borne significant fruit; and indeed a colleague has remarked to me that NN’s philosophical abilities went from “good” to “superb” following the several months during which I was attempting to make her into a “genius”.

McGinn wants to have it every which way: On the one hand, he’s in a consensual, amorous relationship, but it’s part of his formal instruction! And he can’t, just can’t report it since the student didn’t want him to and he knows where to draw the line!

As asinine as his description of “the Genius project” was, he might have been able to make room for it by, for example, discussing it with his chair. Given that this is precisely an area where things can go spectacularly wrong in pretty standard ways, it is ever more important to be above board about it. Even though the university didn’t charge him with sexual harassment, from his own description and subsequent events, it looks like he at least came pretty damn close.

None of this is indicative of a careful teacher with good judgement. Indeed, it shows a person with, at best, a recklessness fed by his ego. (The ego is well documented.)

Just as a plagiarist might also produce good (and honest) work, so too might McGinn now be a reasonable instructor (contrary to the evidence of his own making). But both warrant heightened scrutiny. As far as I can tell, the East Carolina department just brushed all that away. That is something where it is possible that administrative override would be reasonable and necessary.

However, we should not give either side the benefit of the doubt here. Administrative interference should always be met with high suspicion. There are other measures they could take (like having him undergo sensitivity training; making sure he has limited student contact; requiring the committee to undergo similar training). Similarly, the committee has done a terrible job at least in their public pronouncements in a way that is likely to feed into the ongoingly hostile climate for women in philosophy.

S vs. M

The brief summary comparison:

  1. In Salaita’s case, we have an effective fire (which leaves him much worse off than before), at a point where his academic freedom should be very strongly protected, for professed reasons that are far from any normal expectation and farcically weak (with no whole picture analysis), with revealed reasons which are those which require strongest resistance and with secondary effects that are extremely chilling across the US as well as at Illinois.
  2. In McGinn’s case, we have a similar event (I’m not sure how far along it go) which doesn’t damage McGinn’s prior position directly (he having no prior job to lose), where we don’t know anything about the administration’s rationales, but for which there is at least a consideration about his fitness where his academic freedom is fairly weak and his public explanations are transparently poor and where the possible chilling effect is small or beneficial (chilling harassment is a positive good!) and there’s a possible climate benefit.

I think the cases are distinguishable. In McGinn’s case, I have problems with both the administration and the hiring committee, but, on balance, I think the hiring committee fares poorest. With Salaita, it’s not even close — the administration is deeply in the wrong.

  1. Wordpress ate a bunch of work at this point! Fooey! So this was slower in coming than I hoped. 
  2. Not even using robots! “Adam [the robot scientist] is capable of creating up to 1000 individual experiments in a day, with a typical experiment running for 4 days.” 
  3. “””Steven E. Koonin of Caltech called the Utah report a result of “the incompetence and delusion of Pons and Fleischmann,” which was met with a standing ovation. Douglas R. O. Morrison, a physicist representing CERN, was the first to call the episode an example of pathological science….A 1991 review by a cold fusion proponent had calculated “about 600 scientists” were still conducting research. After 1991, cold fusion research only continued in relative obscurity, conducted by groups that had increasing difficulty securing public funding and keeping programs open. These small but committed groups of cold fusion researchers have continued to conduct experiments using Fleischmann and Pons electrolysis set-ups in spite of the rejection by the mainstream community. The Boston Globe estimated in 2004 that there were only 100 to 200 researchers working in the field, most suffering damage to their reputation and career.””” 
  4. A department can face significant problems if they have low completion rates. For example, BBSRC has for certain studentships a (section 2.9) that a department have a 70% 4-year submission rate in order to be eligible. Manchester’s computer science has felt this and related pinches. 
  5. For those of you who like mathematical metaphors, a research community can get stuck in a local optimum
  6. In many ways, we’ve advanced! We’re more cognisant of cognitive blindnesses and things with negatively affect people like micro aggressions, stereotype threat, implicit bias, etc. 
  7. My only prior exposure to his work didn’t leave me particularly impressed. But this was only a tiny bit.