January 21, 2019 (1600 words)
::
What brought me to the left was finding a cognitive 'click' when I discovered big-t Theory.
Tags: the-left, personal, david-foster-wallace
This post is day 21 of a personal challenge to write every day in 2019. See the other fragments, or sign up for my weekly newsletter.
In computer science, there’s a concept called “regular expressions”. It’s basically a way to search for a sequence of characters matching a certain pre-defined pattern. Every programming language or program that supports regular expressions has a slightly different syntax for it, but there is an underlying theory that is fairly consistent across the different implementations.
For example, if I wanted to search a document for all instances of the word “theorise” or “theorize” (i.e., including both US and UK spelling), I would issue a search for theori[sz]e.
That’s one very basic example. The cool thing about regular expressions is that you can search for fairly complicated constructions. I could search for anything that resembles a URL, or an email, or date in MM-DD-YYYY format (or any other format). They’re an incredibly powerful tool, and they’re important building blocks in lots of the software you use today.
I first learned about regular expressions in a very prosaic, pragmatic context, as a web developer trying to build websites using PHP. I didn’t have any formal training at first - I learned from a combination of reading other people’s code, asking questions on discussion forums, and frantic googling - and I learned how to use regular expressions merely because they were the most efficient way to accomplish what I needed to get done. I didn’t have any clue how they worked, under the hood; I just looked up the proper syntax any time I needed to use them (for example, to check if the user-inputted username is actually a valid username). In my mind, they were a useful tool, but their inner workings felt like a black box. I didn’t really care how they worked behind the scenes - I just needed them to work.
I first started taking computer science courses in college. Some of it wasn’t very useful - rehashing the basics of languages I already understood, or teaching the details of programs no one used anymore. But some classes were genuinely mindblowing. One class in particular sticks with me to this day: COMP 330, Theory of Computation. (The link is to the lectures notes I wrote when I took the course in 2013.) We learned about the concept of “finite automata”, which we could also think of as “finite state machines” (FSMs). If you know what Turing machines are, you can think of FSMs as being similar, but simpler and less powerful. Basically, they’re a mathematical concept, or thought experiment, involving machines that enact certain operations depending on the input they receive.
It turns out that these finite automata (and specifically, a subset of them that are “deterministic”, i.e., involve one-to-one mappings) are the theoretical underpinnings for regular expressions. Any regular expression can be represented as a deterministic finite automaton (at least, this is true for the “classical” conception of regular expressions, which don’t include backrefs).
Basically I learned how regular expressions actually worked, and also what made them work. This was the underlying mathematical concept that both proved their correctness and explained how they could be implemented. Every time I wrote code to validate that a sequence of characters was actually a URL or email or date, these imaginary mathematical metaphors were being invoked, turning my regex symbols into a sequence of instructions for carrying out my wishes.
I had been using regular expressions for so long without knowing how they worked. That wasn’t necessarily a problem - I could still use them, after all - but once I learned how they worked, there was this click, and it was like a whole new world opened up. It just all made sense. They became more familiar to me, and easier to use, and the satisfaction of successfully solving a problem was complemented by the absolute joy of harnessing the power of these mathematical devices. There was a strange beauty embedded in the relation between theory and reality. A little piece of the universe had revealed its inner workings to me, and for the next few days I was insufferable, telling anyone who would listen about how cool DFAs were.
That wasn’t the first time I came across the “click”, nor would it be the last. As a computer science and math major, my entire college experience was permeated with “clicks” of various intensities and contexts. Learning about functional programming opened some heavy intellectual doors for me. A class I took on mathematical logic (MATH 318) felt like it was designed to cram the maximum number of “clicks” into a semester, teaching me the mathematical underpinnings of logical reasoning, and how you could re-create the ordinal numbers with set theory. A class on algebra (MATH 235) unfurled more slowly, because it started with very basic building blocks, but eventually it put everything together to explain the the logic behind public-key cryptography and it all just fucking clicked. It felt like having x-ray vision, allowing me I could peer inside this mathematical black box in order to see the bare reality underneath. There was something majestic about it, and maybe even a little terrifying.
People are often asking me to explain my “conversion” to the left over the last few years. I’ve written about it before, in previous fragments and in my Silicon Inquiry piece for Notes From Below, and have talked about it on various podcasts, but the truth is, I’m still trying to make sense of it myself. It was, I believe, partly happenstance. I had lost faith in my startup (and startups in general), and was starting to slip into existential malaise for want of a better theoretical framework with which to live my life. At the same time, I was starting to read more: literary fiction, literary criticism about that fiction, philosophy to help me fill in the blanks of that literary criticism, economics and political theory to explain the increasingly crumbling world order around me.
I wasn’t following any prescribed syllabus, at first. I would even wander around the library and pick books off the shelf at random. I didn’t know what I was looking for, or at least, I couldn’t have explained it to you if you had asked. Unconsciously, though, I was looking for things that clicked. I was looking for some theoretical underpinnings that would eludicate the world I saw around me, and as much as I loved mathematics & computer science, I was started to realise that these self-enclosed domains had a very limited purview. If I wanted explanations, I’d have to look elsewhere.
And so I followed the click. I read lots of books and watched lots of YouTube videos that ended up being dead ends. I even went through a brief Jordan Peterson phase, which didn’t last very long, thankfully (there wasn’t enough click). The stuff that did “click” for me, I would pull at the threads, grasping for more of that same feeling.
It wasn’t solely about understanding how things worked, you see. It was more than that. You learn how things work all the time, and it’s usually a mundane and forgettable experience, fading to become an archived memory. But there are times when that understanding is illuminated with a special kind of beauty, and you just have to stop and marvel at it for a while, because it explains the world to you in a way that changes the world for you.
David Foster Wallace was a big “click” for me. As a result, I made it a mission to read everything he’d ever written, and all the interviews he’d ever done (in which he has mentioned his own “click”), and then all the literary criticism that was written about him. The latter led me to big-t Theory, and all its attendant critiques of capitalism.
The rest, as they say.
I don’t know if following the “click” is a tenable personal philosophy for trying to understand the world. Maybe it’s misguided; maybe there are more dispassioned ways of reasoning about things. But even if the pursuit of analytical beauty is a choice, not a necessity, then it’s a path that I have no qualms about taking.
None of this is to say that Marxism, or any of its descendants, is as analytically rigorous as the link between regular expressions and deterministic finite automata. It’s not, obviously. It’s animated by scientific rigour to a point; the rest is a combination of moral critique and the kind of theoretical beauty that only comes from over-simplification of complex phenomena. You can’t theorise about the actually existing world the way you would about the artificial world of computers, because the latter is a hermetic construct whose terms can be fixed in advance, and the former is a variegated spectrum of infinite complexity.
Like that Borges story about a fully exact map, the only 100% accurate model to describe the world is the world itself. Of course, that’s hardly useful. But if you’re looking for actionable theories to explain how the world works, you’ll need to settle for a patchwork approach, and you’ll have to accept that some things will always be outside the ambit of any theory. There will always be bits left unexplained, lurking uneasily in the shadows. But as long as it shines a bright enough light somewhere else, then maybe that’s okay; maybe your theory doesn’t have to explain everything.
On its own, theory can’t fix the world. It can give you a reason to, though, along with a roadmap, however smudged and incomplete. That’s all you can really ask for.