These are my course design notes. This also doubles as my manifesto for teaching.

I. The Idea

The idea, reduced: on the courses that I design and run, you will be learning by building learning resources for other people to use.

The idea in longer form: On courses that I design and run, you will build. You will build resources for each other to use, utilising resources from generations of students before you, in order to provide resources to the students who come after you. Like every other social contract in history: you never signed up to it, but you have to live with it.

This focus upon building might strike you as odd, since I don’t teach STEM subjects. But you don’t need to be an engineer to build things. So what is meant by “building” in this context?

II. Building vs Make-Work

For the purposes of argument, let us contrast building stuff with make-work. A good example of make-work is the ten-minute powerpoint presentation in a seminar. Consider that it takes an individual 2-4 weeks to do the research required to prepare a decent powerpoint presentation, and that this primarily benefits the individual by giving them a reason to do 2-4 weeks worth of research. This work benefits the cohort to the extent that individuals learn more in discussion than they could have done outside it.

I didn’t do powerpoint presentations at university. We did presentations. Both, in my view, are largely redundant exercises that are better replaced with other activities that are better for students as individuals, and as cohorts. They are examples of make-work that litter education, and these activities should be minimised where possible, and replaced by exercises that are productive at a cohort level.

The reason: presentations are worthwhile when introducing an unprepared audience to a topic. They are also useful when trying to sell something, or persuade people to do something. At graduate level, an audience should not be unprepared, and it shouldn’t be about marketing.

Now consider that this presentation will go no further than the seminar room. It is transient in nature. At best, it will linger on as a saved file on your hard drive (or cloud storage). Maybe someday you will not recognise the filename and re-open it and have a flash of memory about the subject, or the seminar. But that’s about it. Speaking from experience, I have a number of powerpoint presentations on my current laptop that I can neither remember delivering, nor remember who I delivered it to. That’s the nature of presentations - they are fleeting entities.

So by make-work, I mean two things: one, a task that does not have a constructive long term purpose, and two, a task that is transient in nature. Exams are make-work. Outside of educational institutions and professional accreditation, you don’t get exams, you get success or failure at your task or job.

By extension, when I mean “build stuff” I’m talking about working to create resources that other people can use. People you’ve never met, and likely never will meet. And if you want a reason for that, it’s because a fractional contribution to a shared resource that actually gets used will push the world forward more than any seminar powerpoint presentation that has ever existed, or ever will exist.

III: The Flywheel

The projects that I design for you to do on my courses are there to do three main things. First, by doing them you’re going to be learning a little slice of the course in more depth. Second, by doing them in a structured way, you’re going to be able to benefit from the work of everyone else on the course. Third, they’re a means of transferring tacit knowledge from me to you. So, in a sense, this is all about improving your education over the course of a module. But a designed byproduct of this is to generate resources for other people, regardless of who they are.

The group projects that I’m going to ask you to do are an efficiency measure at cohort level. Take the book digest task, for example. There are core books on every course - the required reading that one simply needs to do in order to master the core course. I could ask everyone to do a book digest of the same book, or of a book of their choice. On one hand, there would be incredible redundancy of work (like the powerpoint example above, we would quickly identify the limits of orginality in abstracting a book’s argument), on the other, it would be quite likely that some students pick irrelevant books, or spend an inordinate amount of time trying to figure out which books are relevant. In social science and the humanities, there are always more relevant books to any given course than a single individual can read. Moreover, there is an information processing bottleneck - to understand what you need to read or don’t need to read you would need to spend a lot of (precious) time reading. So if, instead, someone who knows the topic in depth (me), is able to pre-select a list of fifty books to be processed by the course, then you end up with a short guide to the fifty most relevant books on the course (which I’ll edit to correct any glaring errors). Yes, some serendipity is lost in the process, but you can save that for your own research projects.

I liken this process to a flywheel. It takes a lot of effort to spin up a course (like, plan it out, write lectures, plan group projects, etc). It also takes a fair amount of effort to keep the thing going (delivering lectures, feedback, marking, etc), which means many of the kinds of things that I see as key to pushing the envelope (exhaustive literature searches, resources of 50 book digests, etc) are beyond individual instructors. So to get things moving faster, it means that I instead put effort into designing group research projects, and then turn your efforts into a shared resource.

Underlying all of this is a search for efficiency. A lot of underlying elements of academic study are quite inefficient. For example, if it takes a minute to locate a footnote to copy and paste (or type out), and you happen to need to locate about 450 references over the course of a year (sounds like a lot, but they add up), then that amounts to seven and a half hours of your life that you are not going to get back. One of the things I’ve built for my modules is a bibtex file containing everything in the extended reading list. Add that to a reference manager such as Zotero, and you’ll have a couple of hours of your life back with which you can do something more productive, like read a book. If everyone on a course uses it, then a single resource can save tens of hours of time which can be better used by students to learn stuff. It doesn’t sound like much, but these kinds of efficiencies can add up. More importantly, they can add up beyond the timespan of a single module.

IV. From Groundhog Day to Accretive Knowledge

Modern universities treat education as a product. It is part of a service that students buy, often with government-backed loans of some form or another. There are long-standing objections to this product model. Some traditionalists like to think of education as something more than a service interaction, some radicals see the purpose of teaching as generating networks of informed resistance to the hegemony of capitalism. Either way, the bureaucracy that pays our salaries tends towards the product/service model of education. I’ll sidestep the endless debates about the neoliberalisation of the academy by pointing to something more fundamental: regardless of what people think teaching is for, it is something that gets repeated each year to the beat of the academic calendar. Whether the lecturer is in the business of service delivery, producing an informed citizenry, or radicalising the youth against the structural inequalities of the world, “It’s week three, so today we’ll be discussing causal inference.”

Depending upon your attitude towards repeated activities, you could call this the Groundhog Day model of teaching, or the Sisyphean model, or, if you’re really not into it, the Promethean model. Under this model, improvements to teaching are generally the result of the lecturer or professor learning from mistakes and innovations, and applying that learning to the delivery of next year’s course. Fundamentally, very little is generated as a by-product of teaching, even if each student leaves the course having made leaps and bounds in their understanding of the world.

Let’s return to the powerpoint presentation. As discussed above, one issue with the powerpoint presentation is that it is fundamentally limited to helping a single seminar room learn. Of course, that’s not a bad thing, but it is also incredibly suboptimal in 2020. Under the Groundhog Day model of education, a seminar presentation might inspire the lecturer to re-consider something, but the possibilities of it doing so tail off after enough yearly repetitions. That’s not a failing of students, that’s implicit in the design of the activity. After a few annual cycles of powerpoint presentations, there is not much new to be said on a topic that can fit in a 10 minute presentation that needs to inform complete newcomers to the subject.

So let’s recap the problem: A cohort of 20 students doing 2 presentations per person will, at a conservative estimate, perform 80 weeks worth of research that goes no further than the seminar room. Scale this up to a student body of 500 students, and we can estimate that, conservatively, 2000 weeks worth of research will be done in a given year that goes no further than the seminar room. Humans work something like 50 years, maybe 48 weeks a year (give or take, I know this is unevenly distributed), that’s 2400 weeks worth of work. So another way of looking at this is that each student body of 500 students will probably do enough make-work in a year to cover most of an entire human lifespan. That’s an awful lot of effort that goes no further than the seminar room.

What if you could transform 1% of that effort into something useful for future students, or other researchers? What about 10%? 50%? 90%? Conversely, how much more could students learn over the course of a degree if 1% of the effort of the students who came before them had been dedicated to building stuff to help those who followed?

Teaching, in the groundhog day form, lives and dies with the module. For optional modules, this often tracks the lecturer’s moods and research agenda. A module is born, is taught, and then ceases to exist except in the memories of students, a module outline, and perhaps a book or article if the lecturer wrote them up. That’s fine, but we can do better, particularly since the internet has pushed the marginal costs of digital publishing to near zero in a variety of media.

Thankfully, this doesn’t require anything revolutionary. It already exists in stuff like Wikipedia or Stack Overflow. By structuring knowledge and turning Q&As into resources, these sites enable newcomers to learn quickly.

In my ideal world, everything that I ask you to do would be convertible into something that other people might be able to use, re-purpose, or build upon. In practice, this is not possible. Some of the effort that you make needs to be standardised so that it can be tested in a fair manner, for the purposes of awarding marks. But beyond that, your time will be much better served working to build stuff than working to deliver a ten minute powerpoint presentation.

So ask yourself, would you rather spend your time on a powerpoint presentation that ends in the seminar room, or would you rather make something for the future?