Monday, October 26, 2015


Siding With the Villain

Clark Kerr famously quipped that university faculties are loose groupings of independent experts, united only by a shared grievance over parking.  As with the best hyperbole, it captured something real.  As anyone in academic administration can attest, there’s a constant, structural tension between the need for faculty autonomy and the need for institutional consistency.  

That tension, I think, is at the root of the story that emerged this week about the conflict over a math textbook at CSU-Fullerton.  The story itself combines several issues.  First, the department assigned a common textbook for every section of a class, regardless of who taught it.  Second, the required text was expensive.  Third, the required text was written by the department chair and vice chair.  Fourth, a professor who broke the rule about a common text assigned a combination that was much cheaper.

So the CSU case combines autonomy/consistency, cost, and conflict of interest.  Unfortunately, it’s easy to conflate the three, and therefore to sacrifice a larger point to a sticky case.  For example, what if the renegade professor had assigned a book that was more expensive?  Or that he had written himself?  I’d hate to base a general policy on the contingencies of a single case.

I’ll focus on the first issue, since it strikes me as the most common and the most important.  Should departments be allowed to determine, and enforce, common texts across every section of a class?


I say that not only because the AAUP agrees with it, although it does.  I say that because at the end of the day, credits are granted not by faculty, but by institutions.  Institutions determine the title, the goals, the credits, the day, the time, the location, and the duration of classes.  If a professor decides that she doesn’t like a given timeslot, she is not free to simply change it on her own.  Students plan on it, room conflicts are real, and for security reasons the college needs to know who is teaching where, and when.  

The interest in institutional coherence isn’t limited to logistics.  If a professor is hired to teach, say, digital circuits, and instead spends every class period discussing her family vacations, then she is not doing her job.  “Academic freedom” is not absolute, nor is it entirely individual.  Saying that “the faculty” controls curriculum is different from saying that each individual professor does.  Collapsing the first into the second actually destroys both.

I’ve seen conflicts like this at previous colleges.  A math sequence is built on the assumption that fractions are covered in the first class.  But a professor in the first class thinks that students struggle too much with fractions, so he decides unilaterally not to cover them, leaving it for the next class in the sequence.  The professor teaching the next class doesn’t know that, and proceeds on the assumption that students were taught something that they weren’t actually taught.  Students fall through the cracks, and a collectively-determined curriculum is jeopardized.  The first professor’s disregard of a collectively-determined objective created a problem for the second professor, as well as the students. This is not okay.

In the CSU case, my objection is not that the department chose to assign a common book.  It was within its rights to do so.  My objections are that it chose without apparent regard to cost or to a conflict of interest.  (I say “apparent” because I wasn’t there.)  It chose badly, which is fair game for criticism, but it was within its rights to choose.

Historically, one reason for assigning common texts was actually to reduce textbook costs.  When a common text is assigned, and held for multiple years, then a healthy market in used books can develop.  Students can save significantly with used books, as opposed to new.  When each professor assigns different books, the market for any given book becomes too small.  But when, say, a Psych department picks a common text for the Intro class, the volume is high enough to sustain buybacks and resales.

The IHE piece situates the issue within the larger questions of OER, which is exactly my concern.  For a college to be able to put together an entirely OER degree program, every class in the program has to use OER.  That means every professor teaching those classes.  If someone decides to go rogue and assign something commercial, the students will rightly complain about being misled.  And when financial aid awards are based on estimated costs, a professor who unilaterally decides to impose substantial new costs does significant harm.  Siding with the rogue in a sympathetic case would suggest a rule that would effectively forbid a much greater good.

Theoretically, of course, unanimity would smooth the tension between autonomy and consistency.  But unanimity is both rare and fleeting.  As a standard, it gives undue power to a single dissenter.   If you want a rule to hold up over time, you need to be willing and able to enforce it.  There’s a reason that democracies don’t rely on unanimity, or allow citizens to cherry-pick the laws they’ll choose to follow.  If they did, they’d essentially hollow out the point of majority rule.

I’m no fan of the choices the CSU department made,  I agree that there’s a serious ethical issue around the conflict of interest, and that textbook cost deserves more attention than it gets.  The department seems to have gone out of its way to make itself the villain.  But siding with a sympathetic rogue here would lose the larger cause.  

But there is obviously a point when the cost of the required textbook gets so high that a teacher has to rebel e.g. $1000? $10,0000? Even if it's just to say "if you can't afford this one then this one will be good enough".

I agree with the general point: academic freedom may inhere in the faculty as a whole, not an individual instructor, in the case of courses that have multiple sections. But in this case, the obvious conflicts of interest and the lack of any attention to students' budgets seem to have crossed the line from academic freedom to irresponsibility, if not cronyism, and I stand behind this instructor not because he chose different textbooks, but because he chose them on the reasonable grounds that his students' interests were not being served by the department policy—a policy that might well violate the CSU conflict of interest policy. (Maybe it doesn't, but best practices that I've seen for faculty who assign a textbook that they wrote is that they should offer it to students at the author's discount price, or use some other mechanism to ensure that they are not personally profiting by forcing students to buy a book they wrote, e.g. by donating any royalties they might receive from their own students to charity.)

Regardless of who wrote it, a textbook for differential equations and linear algebra should not cost $195. Brill, Ashgate, Brepols, and other academic publishers assure us that their monographs cost $120, $150, $200, and more because they publish scholarship that has a very limited audience, and they have fixed costs to cover, so the price has to be set high. I don't always think they're right, but they do have a point. They publish books that sell a few hundred copies at $150, and they're clearly profitable. Now take a linear algebra textbook. It will sell tens of thousands of copies, if not more. Mathematical typesetting is cheap these days. Textbooks are not made to last a long time. The obvious conclusion is that, if Brill can make a profit selling 500 copies of a $195 book, a company that sells 50,000 copies of a $195 book is price gouging, and that's easier for them if they have a captive market.

Books aren't that expensive to produce. I took linear algebra in 1985 and paid $25 for my textbook. Based on the CPI increase from 1985 to 2015, that would be $55 in today's dollars. Linear algebra at the college level has not changed significantly in the last 30 years. If publishers were interested in serving the interests of students, they wouldn't produce new editions every three years. Such "revisions" seem to be intended to make using earlier editions a hassle, and thus to undercut the used book market, rather than to reflect advances in the field.

So yes, I agree with your general principle, but in this case, the circumstances seem so abusive that I think this is a clear exception. If the CSU Fullerton faculty were being reasonable, they would evaluate the evidence and decide that their current policy was outdated and needed to be revised. However, that might be hard if it involved crossing their chair and vice chair. If CSU Fullerton had offered their students a royalty-free price for the textbook, on the grounds that it was the best book for the course but since their tuition was already paying the authors' salaries, it would be unseemly for the authors to also collect royalties, they would have a much better case.
In case anyone else was as confused as I was by the lingo, OER apparently means open educational resources in this context.
Like Matt, I also am of two minds on the CSU-Fullerton math textbook issue.

I fully recognize the need for uniformity in classes that have multiple sections. All of the sections should conform to the description in the course catalog, they should all have the same syllabus, they should all cover the same material, and they should all use the same textbook. This is especially important if the class is a prerequisite for more advanced courses in the curriculum, where mastery of the material covered in the introductory courses is important for success in the advanced courses. It is important that we don’t have an individual instructor in a multiple-section course going rogue, covering their own specially-favored material rather than the material called for in the catalogue description of the course. Even the AAUP recognizes the need for uniformity in multiple-section courses.

I remember in graduate school that a tenured professor had absolute freedom to do anything they pleased in teaching their class. They could teach anything they wanted to, even if it deviated significantly from the course description in the catalog. I remember hearing that one professor who was teaching an introductory course in quantum mechanics, saying on the first day: “I am not going to teach you quantum mechanics, I am going to teach you the quark model.” But there was only one section of the class. Nevertheless, students who enrolled in the class hoping to learn about quantum mechanics must have been sorely disappointed.

But the math textbook in the CSU-Fullerton case raises some serious conflict-of-interest issues, since it was written by the department chair and vice-chair and was super-expensive. Did the department as a whole have some say in what textbook was to be used, or was it strictly up to the department chair, who had a financial interest in the choice? Back when I was teaching at Research Intensive Technical Institute, as a lowly assistant professor I don’t remember having any say in what textbooks were to be used, the decision being made by the department chair and assistant chair, or by the tenured faculty members.

But in those days textbook prices were not super-expensive. Back when I was an undergraduate, a math textbook often costed less than $10. Nowadays, a textbook can cost $200 or more. There is often the possibility of going to the used-book market and getting a copy of the assigned textbook for a significantly lower price, but it seems that there is a new edition of the text put out just about every year, usually with only minimal changes such as a renumbering of the problems, a reshuffling of the chapters, or the addition of some extra graphical material. This is done primarily to beat the used-book market. The whole textbook game has become a huge racket, designed to benefit some of the more predatory textbook publishers.

Greater attention must be made at the departmental level to address this textbook cost issue. One possibility is to choose an open-access textbook, which can be obtained for a much lower cost or even for free. Another possibility is for the department members to write their own textbook and offer it to the students for only the cost of reproduction, or to make it available online for free.

I'm with the general consensus here that what CSU-Fullerton did institutionally was right on one level (common text and syllabus for a required introductory course) and wrong in the implementation. And I have a story.

In 1987 (or thereabouts) the institution at which I was teaching decided to move from anarchy in the selection of an intro econ textbook to a unified adoption. Two of us (as it happens the two junior, and untenured faculty) were asked to make a recommendation. The econ faculty agreed on a set of criteria to evaluate intro texts, we applied them, and made a recommendation. The senior faculty said no, we don't like that book, we're going to continue to use the books we've always used (which were, in my opinion, not very good, but that's a different issue). Our dean did not say to them, look, we agreed on a unified adoption. You don't like their choice, so you pick one. He shrugged his shoulders and the issue was dropped.

After all this time, it still rankles (can you tell)?

Later on, following a lot of turnover (so I was the only econ person remaining from that debacle), we were still in a state of anarchy. And a new faculty member selected his books. And what he did was terminally stupid. Intro econ books can come in "splits"--micro econ and macro econ. He chose books by different authors--but he assigned the combined micro/macro hardcover editions, not the paperback splits. I suggested to the dean that she (obviously a different dean) suggest to him that he should reconsider that. She did not.

So there are many ways to screw students over in the selection of textbooks...

Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?