BLAKE: Hello. My name is Blake Stacey. I am a physicist at the University of Massachusetts Boston, my pronouns are he/him (they/them is also fine), and today I will be talking about the unjustifiable cost of physics textbooks.

Physics is a subject which has a remarkably high barrier to entry, for multiple reasons. One of which is the separation between the reasons why people get interested in the subject and what an introductory education in it actually involes. So many homework exercises in introductory physics courses are really the analog of practicing chords on the guitar when you want to be a rockstar. And we as a profession have failed to find a way to connect the introductory exercises with the promise of the subject as it is more widely advertised.

Another difficulty is purely economic! Making the commitment to learn the subject all the way through is a fantastically high barrier to entry --- it is a pocketbook-draining subject to learn to any significant extent. And just to put some actual numbers to this, I looked at something I have experience with, the course requirements for getting an undergraduate degree in physics from the Massachusetts Institute of Technology.

These are just the physics and math courses that one is required to go through to get the plain-vanilla physics degree from MIT. And more or less in order, starting with first-year introduction to Pulleys and Inclined Planes (Mechanics 1), and taken simultaneously with that generally is a calculus course. Then the following semester one does electromagnetism the first time around, coupled with another semester of calculus. And then in the next year, one does a course that's kind of a junk drawer on waves and vibrations. Simultaneously with that, roughly, you're sent over to the math department to learn differential equations. Somewhere in there you take a course on relativity (special mostly, with a dose of general). And then you get into fairly standard run of some semesters of quantum mechanics, some statistical physics and thermodynamics; there's a year-long lab course (which is a whole thing and of itself), and then you start having more choice later in the degree, but generally you'll be taking most all of a second and possibly third round of classical mechanics, more advanced electromagnetism, more advanced statistical physics. And somewhere in there you have to go over to the math department for a higher-level elective course.

And this, plus passing the swim test and getting some gym points, and various other degree requirements, are what is required to get a physics degree as an undergraduate from MIT.

Now I went through the course catalog and looked up what books you're expected to buy for all of these. There's some flexibility: some things are recommended, some are required, some are "recommended" but very strongly recommended (I know from experience). Some courses didn't have specific books listed because the particular professor in charge of teaching this course the next time around didn't submit a syllabus in time. There's some fuzz in this number. But if you add everything up for all the books that you're strongly expected to buy so that you can have on your own shelf everything that you are expected to read from and do exercise out of... It's in excess of 2,400 dollars.

Now, why?

No, really, really, why? Why does it cost 2,400 in books --- not tuition, not room and board --- why does it cost 2,400 dollars just to buy physics in book form? It's completely unjustifiable. OK? This is the Year of Our Nonbinary Lord 2022, and nothing in this curriculum is recent. There are plenty of books on a lot of these subjects that are perfectly fine with minimal supplementation that were written 40 years ago. Why do we have to buy the pricey new editions that come out every 2 years for all of these books? There's something completely unreasonable happening here. You think, OK, maybe there are resources out there where just by going online and hunting around for open course ware, open textbooks... Surely physicists write so much --- we crank out words to such a great extent --- surely by making better choices of the resources that are already available, we can bring that number down.

So I looked at OpenStax, which is a several years old now publisher of generally pretty good-looking, free to read online, purchasable in printed form at low cost textbooks. Not a fly by night operation, science books written by actual science people. So what happens if we turn to the open textbook community and see what they have to offer --- how does that bring the number down?

Well, OpenStax would give you viable substitutes for the recommended and required reading in four out of those courses! The first-year courses in mechanics, electromagnetism and calculus. Usually these things may be called something like "University Physics" or "University Physics 1 and 2", "University Calculus 1, 2, and 3". Different universities might divide the material up somewhat differently, but basically, that's what's covered by the most well-known and established official publishers of open educational materials.

Look a little more broadly: the American Mathematical Society has a pretty extensive collection of online open-access reading materials, so if you look at OpenStax and the American Mathematical Society, you could also get by... Or if you were designing a curriculum, you could pick a less expensive textbook option... for differential equations and the typical choices of math electives, like higher-level linear algebra. Great. The courses in bold are what we have free or much-lower-cost textbook options for. Adding up those numbers, that's still over 2,000 dollars a student will be expected over the course of getting a physics degree.

That's not a great savings! We've barely managed to save perhaps a fifth of the expense on textbooks. That's a pretty poor performance for a problem that has been recognized for about thirty years. You look back at people writing about the possibilities of how computers and the interconnection of computers will change publishing, that started off in the very early days of the world wide web and just prior. So, the early 1990s, people starting to realize that things could be done differently with the publication of physics journals at least. Why have we done so badly and so slowly on the teaching of our subject? We're willing to devote our lives to this subject matter. And we definitely believe it when we talk about the importance of understanding the origin of rainbows and lightning and galaxies and the dread of the atomic bomb. But we've raised what by now seems like it should be a completely artificial barrier to entry. And we've failed to lower that barrier to entry by more than about a fifth of what it would be without all the advances of the past thirty years in open access, open education, open publishing.

So, what's gone wrong? What's gone wrong?

Well, one indication of a cultural factor that we should find concerning is that there are some physics books which are known by their author's name alone. If I say "Griffiths", that narrows it down to two books. If I say "Griffiths quantum" or "Griffiths electromagnetism", every physics student will recognize what I'm talking about. If I just say the one word "Jackson", physicists the world over will know exactly what I'm talking about, and will probably shudder in suppressed fear, just a little bit, based on their prior experiences with "Jackson".

So we have these culture-wide standards, that we treat as embodying what these standard subjects are about, but none of these ubiquitously known, famous or infamous standard texts, known by their author's name alone, are ever free. There's no open-access physics text that has reached the level of fame or infamy of Goldstein on classical mechanics or Jackson on electromagnetism or Griffiths on quantum mechanics. I find that a straight-up peculiar phenomenon. How is it that physicsts who write on a daily basis --- that's one of our major outputs is the written word in the form of journal articles --- and we live online, we were responsible for the invention of the world wide web in the first place --- why have we not coalesced about an open standard for any of these subjects that becomes as ubiquitously known as a Griffiths or a Jackson?

I should caveat that statement a little. Over on the math side, Gilbert Strang's calculus is probably up there in widespread recognition, and that is available to read at no cost online. So that's a promising step in the right direction. And there are non-textbook resources which could be highly recommended reading, for example the Feynman Lectures on Physics can all be read thanks to Caltech's website, but for multiple reasons we could get into in considerable depth, the Feynman Lectures were not, are not and won't be textbooks usable in classes for day-to-day coursework. So, while there are other resources and there are some promising indications that as far as physics books go this could change, the status hasn't changed yet.

Why and wherefore?

First point, capitalism.

When there are about five textbook publishers who control upwards of 80 percent or more of the textbook market, we shouldn't even expect market dynamics to work. Even if you think that there are cases in life where a market can be efficient, you won't think that this is one of them. Because the competition among sellers is minimal. And there isn't anything like efficient distribution of information in this economy, because the people making the choices of which textbooks are required for which courses aren't the ones forking over money to pay for them. Recent studies on this were kind of hard to find, but from several years ago, maybe only half, roughly speaking, of the faculty assigning textbooks were aware of the prices. So, that's miserable. That's bad. That can't be a good thing. Prices are free to float higher and higher because there's nothing in the market stopping them.

And giant established publishers looked at the Internet and evaluated what it did for their business model, and they've come up with a way for the Internet to make things worse. Textbook rental, digital textbook rental, is a thing that exists in our timeline. And what it has done is kill the used textbook market. It used to be that you buy secondhand, perhaps a slightly older edition of a book, or a perfectly usable copy of a textbook, at a much lower cost than getting one new. This was the case back in my day when I was a student. I could get significantly lower-cost versions of everything I was supposed to read, pretty much. But now that textbooks can be sold digitally, and then on a time-controlled basis access to the textbook removed, or the textbook only ever being rented, in this horrible "you don't own what you pay for" mode of life that we're stuck in, well now there are no physical copies to sit on the shelf of the just-off-campus used-book store.

I'm not sure whether this is as bad in the sciences as it is in the humanities, I've heard from colleagues on that side of the aisle. But it is a contributing factor to the ever-higher floating of prices throughout the industry, and that's kind of a factor that across the whole industry keeps prices inflated.

But the physics community is far from blameless. We can't put all the responsibility on evil multinational publishing conglomerates.

There's no career incentive within the physics community to work toward improving the situation. Zip. You are in a publish or perish environment; what you are expected to publish are journal articles in Physical Review, Science, Nature. You are not expected to produce educational material for the betterment of the next generation. That's the size of it! And if you find that an important thing to work on, you are just taking time away from what you are expected to be producing.

No one gets points for trying to promote the open resources that already do exist. So there's no social push in our scholarly community to make any of the existing open resources (the few that do exist) the new Griffiths, the new Goldstein, the new Jackson. And there's no community gathered together to create such resources for those higher-level courses, where you can really rack up the expenses, having to buy 2 or 3 quantum mechanics books, 2 or 3 statistical physics books, and so on. Well, there's no social push saying that those courses should also have open-access course materials. So no one has yet made the drive for a depth-first creation of open educational materials.

There's been some movement in the breadth-first direction, of, OK, now we'll have introductory textbooks on this, this, this and this --- for as many introductory courses as possible --- so for everything that is Physics for Nonmajors, or the physics courses that everyone in the university is expected to take, we have some things available for that, as we saw on the chart earlier. But there has not been a push to correct that situation any further, any later.

And I have to admit, I am as much the problem as anybody. I wrote a textbook --- a very specialized textbook, probably only suitable for graduate students. But it was an eye-opening experience, in that I, as an author, have basically no leverage to negotiate a lower price point or change the terms of online distribution. I can either take the publisher's terms, or leave them. It's great. Thanks. I hate it.

Having gone through the process, the best I can say is, well now I have a line on my CV. Having gone through that, I at least have the credential of now being an author and have that much more respectability, perhaps, so that I can now say, OK, the next book will be free. My hope is that the course material that I've been working on over the past couple years will become an open-access textbook on introductory quantum and statistical physics, filling in some of the empty spots in the next higher level beyond what is available so far. An incremental change, at least in the supply of open educational resources, even if I on my own can't create demand for them or the social impetus within the community to adopt such resources on a widespread scale.

OK, thanks! The first two images, the collage of glamorous physics topics and the introduction to the inclined plane were both from respective pages on Wikimedia Commons. And the tower of ramen noodles was photographed in my kitchen.

MODERATOR: OK, thank you. I would like to run the Q&A by asking people to raise their hands, using the Raise Hand function, and then I will send them a message asking them to ask their question. I think that will be the easiest way. I also was asked to ask a question that was previously typed into the chat. So I will read it out and that will be the first question --- not from me, but someone whose Internet connection is poor asked me to do it for them. And the question is, "In South America, back when I was an undergrad, most students had no scruples about Xeroxing expensive textbooks. Many local shops even helped us at it. Is it not a thing in NA unis, or is it not a thing any more? Music video piracy is rampant and it drove down prices on those, so I wonder why the same does not happen for textbooks." So that's our first question to our presenter.

So, students have no scruples. That's I think a worldwide truth. Physical Xeroxing, to the best of my knowledge, is not so much a thing. You'd have to find an unscrupulous copyshop to help with that, because at least the official university copytech services do raise eyebrows at the very least when large amounts of course materials are copied. So you may run into trouble doing that. With DRM digital copies, that has to be circumvented. And there are scans for basically all the standard books available online if one knows where knows where to look. But the problem with all that is, it just means that bootleg copies circulate, and it's difficult for professors to recommend theft on the course syllabus. Not everyone is aware of those channels of distribution as young as they should be. The "Do Crimes" part of the physics curriculum is an elective that is not listed on the books, and not everyone takes that elective at a sufficiently early age to make a difference. The circulation of scans, of PDFs and DJVU files, if you know where to find, can really bring down one's expenses. But it in some ways only furthers the problem, because if the expectation is the professor says, "I can assign Griffiths as the required textbooks", then everyone will either buy Griffiths or steal Griffiths online. Then that just keeps Griffiths being the standard. So there's no impetus to change to a standard which doesn't have to be stolen. It's kind of a short-term solution that doesn't get at underlying causes. And doesn't seem to have, just looking at textbook prices that keep going up and up and up --- you know, an increase of two thirds in the decade prior to the pandemic, for textbook prices overall? Whatever beneficial effects they may have had on prices in music, copyright scofflaws don't seem to have had a correspondly good effect on book prices.

I should say that because of the oddiities of screen sharing, I was unable to see the timer, for which I apologize, and I was unable to see comments in the chat while it was going. I am catching up now on people's comments in the little sidebar. "... spending hours in front of the photocopier. To a first approximation, all my textbooks were photocopies." Yeah. Wise move, wise move!

MODERATOR: Hi, first of all, thank you for the talk. Two very quick questions. One, have you ever gone, "You definitely should not go to Sci-Hub and download from this link here or I will be very cross!"? And two, what do you think the way forward to fixing this is? Do you start by trying to engender a culture of solidarity with other professors so they modify their textbooks assigned, or do you start by trying to encourage people to write the open-source textbooks? What do you think a good first step would be?

BLAKE: To answer your first question, that is something I do hear more and more commonly, the wink-wink, nudge-nudge. "People have been trying to shut down this website, it must be really dangerous, don't go to libgen dot eye ess, it looks really sketchy!" To provide at least whatever measure of short-term relief is available. As for the first step to take towards a solution, I think there are multiple first steps that can be taken simultaneously. One of which is to encourage the more widespread adoption of the introductory level open educational materials that already do exist, so that the pressure is there, so that the expectation can be fostered that this is what textbooks should be about. Not the cranking out of needless new editions every 2 years to make the previous 2-years-old edition unususable, but the preservation and propagation of useful knowledge. And if it becomes accepted that first-year mechanics and E&M and calculus are things that one should just be able to go to the American Physical Society website and click and download, the more people expect that, they more they will want that, the more they will be willing to work on building that knowledge base out, to the higher level and more specialized works. Simultaneously with that, we can encourage the creation of those higher level and more specialized works, and one thing to do is just to get a more systematic survey of the various lecture notes posted here and there for some of these higher level courses. Because there's a time barrier to entry. Let's say if I want to teach quantum mechanics, as I will be doing in the fall, I can either tell everyone to buy Griffiths (or wink wink nuddge nudge find your own copy), or I can block off a lot of time and look for, OK, this professor wrote quantum mechanics lecture notes and put them online, and so did this professor, so that seems to cover most of the introductory material of the subject, and maybe there are other resources I can use to fill in, and I can custom-curate a reading list... And that's a significantly greater time investment on the instructor's part. And of course intructors are never overworked and underpaid.

One thing we can do is lower that time and mental energy barrier by doing that curation work and making the results of that curation available in a more obvious place than they are to date. And one way in which we might be able to make progress is divide up the work of writing expository material and making exercises --- you know, homework and exam problems. Because I've noticed when course notes get posted online, one way in which they are unfinished from the standpoint of being textbooks, is --- very few exercises for the student. And I'm not entirely sure why this falls out. My guess is that it's just that much less familiar to the writing habits of your typical physicist. I mean, we're writing journal articles all the time, and writing lecture notes is not unlike writing a journal article. It's more expository, you have more room to breathe, the goals are different, but the mindset of, "this is a thing we need to explain, let's step through and explain it" is pretty much the same. A different skill set comes in when trying to invent homework problems. So if anything gets lost, if anything is the low-priority item that is too much work for the instructor to do, that's what's going to be omitted. So we have this problem where there may be great freely available to read expository material on some of these subjects... plenty of lecture notes online about quantum mechanics... you have to find them, it's not really quality controlled... "All right, I will go and search the web for everything that's been written in PDF form on a .edu website" ... gee great, that's a lot to sift through, and when you do find it, it's missing things you need if you're actually going to be using them for instructional purposes. So we need to organize what we already have to see where the gaps are, and we need to find some way of fostering the production of material that people have been less inclined to write so far.

MODERATOR: Would anyone else like to ask a question? There's no one else in the queue at the moment, and you have 11 minutes or so of time.

MODERATOR: So, yes, I think we can wrap up here, unless anyone else has anything they'd like to say, thank you for that, that was very interesting. And, well, infuriating, I guess in some ways! I don't know to what extent we have the same kind of problems in philosophy. We're less textbook-focused in my experience, but also often lots of reading, often hard-to-find or expensive-to-find if you can't illegally...

BLAKE: Yeah, I figured a lot of these problems replicate from one field to another, but for the sake of being specific, I wanted to focus on what I have the most personal experience with.

MODERATOR: Right.

BLAKE: I have heard many of the same things from friends and colleagues who teach, say, American history. "Oh, there just aren't used books any more because everything is digital rentals." The cost reduction possibilities of digital distribution are hardly realized, and the new method of distribution eliminates the reduced-cost method of acquiring books that we had before, so it's a step backwards in many ways.

MODERATOR: That's a reason that had never occurred to me before actually, the lack of being able to buy these things secondhand, which I definitely did when I was a student.