Modern politics: Where social science ends and fiction begins
I think I may be coming upon a structure for In Tenure Veritas. Topical posts on Saturdays, and more thematic stuff on Sundays. It was what I did last week, and I'm doing it again this week. And in fact, I want to pick up on what, specifically, I did last week. I wrote about the applicability of a science fiction novel to modern politics (Neal Stephenson's Fall; or, Dodge in Hell).
I do this in my classes too. I will assign science fiction novels as a way to understand politics, economics and concepts in a broad array of social sciences. More pointedly for discussions here, though, do you ever feel like things have gotten... a little strange?
The basic theme of yesterday's post on the Democratic nomination contest was that presidential nomination contests are difficult for political scientists because one contest is rarely all that similar to previous contests, and this one is particularly idiosyncratic. So, we have no historical guidance about what's going to happen. In the absence of historical guidance, or put another way, with minimally informative data, on what basis do we make inferences? How do we make predictions? With great difficulty, at best.
Of course, the Democratic nomination contest is about the least abnormal thing about politics right now. Spending this Sunday morning attempting to recount everything that is now weird and unprecedented about American politics, much less world politics, doesn't sound like fun, and it would never end. So instead, I'll muse on epistemology. Deductive reasoning versus inductive reasoning.
The sun rose yesterday, and every previous day. It rose today. I infer, by inductive reasoning, that it will rise tomorrow. Patterns repeat. That isn't a formal proof, but it works pretty well. Most of the time.
Then, there is deductive reasoning. You begin with a set of assumptions, which you take to be true even if you cannot prove. You then formally prove everything that follows from those assumptions. The process of proving what follows from your assumptions is the deductive process. It is, actually, "proof," and I am very picky about the word, "proof."
And getting picky, there is a mathematical principle called "induction." Suppose there is a set of observations, from 1 through N. If a statement is true for observation 1, and every time it is true for observation n, it is also true for n + 1, then it follows that it is also true for every observation in the set. That's the principle of induction. You can prove it, deductively. How? Watch. This is cool. Suppose X is true for observation 1, and any time X is true for n, X is also true for n + 1. Let S be the set of observations for which X is false.
Assume that S is not an empty set. If S is not an empty set, then it must have a first element. Let that first element be k. That means k - 1 is not an element of S, so X is true for element k - 1. But, any time X is true for n, it is true for n + 1, so if X is true for k - 1, it must also be true for k - 1 + 1 = k, so X is true for k. But... we just said k was the first element of set S, where all elements were elements such that X was false.
Contradiction.
What does that mean? It means that S must be an empty set. If you assume it is non-empty, that leads to a contradiction, so S must be empty. (Yes, I'm a political scientist. Some of us do math.) That means the principle of induction must be true. We have deductively proven the formal principle of induction.
Now, that doesn't mean the process of inductive reasoning has been proven. That's just terminology, and me messin' around with math, because what do you do on a Sunday morning?
Anywho, inductive versus deductive reasoning. Inductive reasoning is the process by which we infer that the past will repeat itself because it always has.
What do we do when it... stops? In Cixin Liu's The Three Body Problem, there is a computer game built around, as it turns out, an alien civilization that developed in a star system with three stars. Physics problem. Stable orbits? Not gonna happen. The planet will periodically fall into a stable orbit around one star, and experience a stable era, during which a civilization will be able to rise, but when it falls out of that orbit because of the three body problem, there is a "chaotic era," in which the game participants don't know when the sun will rise or set or why, until they figure out that there are actually three stars (which really takes them too long given who the players are). At that point, at least they know why. They can't really do much, because you need to be able to predict when the sun will rise and set to plan, but... at least you know. And periodically, the instability of a three-star system leads to civilization-ending cataclysms that wipe out everything. Knowledge and predictability... pesky stuff, that.
All social science, and all understanding of the social world is based on the premise that patterns will repeat themselves. I can't, as much as I'd like, deduce everything about what you'll do. In my professional writing, I'm a game theorist. That is, technically, deductive. I begin with a set of assumptions, and deduce what I can from those assumptions, but any game is nothing but a model, with George E.P. Box's aphorism applying: "All models are wrong. Some are useful." And if you read The Three Body Problem, you'll see some creative explanations for chaotic eras, and marvel at how some supposedly very smart people who understand physics can't put 2 and 2 together. Mostly, though, empirical social science is about looking for past patterns and expecting repetition.
You know who does something closer to deduction? Even more than game theorists?
Authors. Good ones, anyway. A bad author doesn't think logically, but Neal Stephenson? Good one. Cixin Liu? Mostly good, although don't bother with the third book in the Three Body Problem trilogy.
One of the more disturbing aspects of modern politics and society is the post-truth nature of dialog. This is something that we can try to discuss with scholarship. We can examine how people come to believe conspiracy theories, what happens when you expose people to facts, how ideological siloing takes place, and so forth. We are reacting piecemeal, though. What Stephenson was able to convey was a deductively structured world, thinking through the bigger picture in a way that scholars are not generally trained to do. (And that was only part of the novel.)
What authors do is create a world. By definition. Even if the world runs analogously to our own, it is still a world. In Charles Stross's Merchant Princes series, the entire concept is about parallel universes, and the universe in which Miriam initially lives looks like it is intended to be ours. Spoiler alert: it ain't. There's a big event a few books in that makes that timeline diverge dramatically from our own, and the point is to trace out effects. Deductively.
We, scholars, are usually working inductively, even when we pretend otherwise. Yeah, game theory, whatever. I'm still basing my mathematical models on empirical observations and posing the notion that the past will repeat itself. That's induction, and a demonstration of where the lines get blurry. Authors, though, are the ones who are working deductively, to construct a world, and address the bigger questions, allowing them to see what happens if the world diverges in bigger ways.
Look around. Does the world seem like it is getting weirder? Are we... diverging? Diverging from past patterns? Breaking from historical norms?
The extent to which that happens is the limitation of social science. We think small. We must because we are constrained by methodology. Larger changes require bigger thinking, but we aren't good at that.
Good authors think bigger than social scientists.
I do this in my classes too. I will assign science fiction novels as a way to understand politics, economics and concepts in a broad array of social sciences. More pointedly for discussions here, though, do you ever feel like things have gotten... a little strange?
The basic theme of yesterday's post on the Democratic nomination contest was that presidential nomination contests are difficult for political scientists because one contest is rarely all that similar to previous contests, and this one is particularly idiosyncratic. So, we have no historical guidance about what's going to happen. In the absence of historical guidance, or put another way, with minimally informative data, on what basis do we make inferences? How do we make predictions? With great difficulty, at best.
Of course, the Democratic nomination contest is about the least abnormal thing about politics right now. Spending this Sunday morning attempting to recount everything that is now weird and unprecedented about American politics, much less world politics, doesn't sound like fun, and it would never end. So instead, I'll muse on epistemology. Deductive reasoning versus inductive reasoning.
The sun rose yesterday, and every previous day. It rose today. I infer, by inductive reasoning, that it will rise tomorrow. Patterns repeat. That isn't a formal proof, but it works pretty well. Most of the time.
Then, there is deductive reasoning. You begin with a set of assumptions, which you take to be true even if you cannot prove. You then formally prove everything that follows from those assumptions. The process of proving what follows from your assumptions is the deductive process. It is, actually, "proof," and I am very picky about the word, "proof."
And getting picky, there is a mathematical principle called "induction." Suppose there is a set of observations, from 1 through N. If a statement is true for observation 1, and every time it is true for observation n, it is also true for n + 1, then it follows that it is also true for every observation in the set. That's the principle of induction. You can prove it, deductively. How? Watch. This is cool. Suppose X is true for observation 1, and any time X is true for n, X is also true for n + 1. Let S be the set of observations for which X is false.
Assume that S is not an empty set. If S is not an empty set, then it must have a first element. Let that first element be k. That means k - 1 is not an element of S, so X is true for element k - 1. But, any time X is true for n, it is true for n + 1, so if X is true for k - 1, it must also be true for k - 1 + 1 = k, so X is true for k. But... we just said k was the first element of set S, where all elements were elements such that X was false.
Contradiction.
What does that mean? It means that S must be an empty set. If you assume it is non-empty, that leads to a contradiction, so S must be empty. (Yes, I'm a political scientist. Some of us do math.) That means the principle of induction must be true. We have deductively proven the formal principle of induction.
Now, that doesn't mean the process of inductive reasoning has been proven. That's just terminology, and me messin' around with math, because what do you do on a Sunday morning?
Anywho, inductive versus deductive reasoning. Inductive reasoning is the process by which we infer that the past will repeat itself because it always has.
What do we do when it... stops? In Cixin Liu's The Three Body Problem, there is a computer game built around, as it turns out, an alien civilization that developed in a star system with three stars. Physics problem. Stable orbits? Not gonna happen. The planet will periodically fall into a stable orbit around one star, and experience a stable era, during which a civilization will be able to rise, but when it falls out of that orbit because of the three body problem, there is a "chaotic era," in which the game participants don't know when the sun will rise or set or why, until they figure out that there are actually three stars (which really takes them too long given who the players are). At that point, at least they know why. They can't really do much, because you need to be able to predict when the sun will rise and set to plan, but... at least you know. And periodically, the instability of a three-star system leads to civilization-ending cataclysms that wipe out everything. Knowledge and predictability... pesky stuff, that.
All social science, and all understanding of the social world is based on the premise that patterns will repeat themselves. I can't, as much as I'd like, deduce everything about what you'll do. In my professional writing, I'm a game theorist. That is, technically, deductive. I begin with a set of assumptions, and deduce what I can from those assumptions, but any game is nothing but a model, with George E.P. Box's aphorism applying: "All models are wrong. Some are useful." And if you read The Three Body Problem, you'll see some creative explanations for chaotic eras, and marvel at how some supposedly very smart people who understand physics can't put 2 and 2 together. Mostly, though, empirical social science is about looking for past patterns and expecting repetition.
You know who does something closer to deduction? Even more than game theorists?
Authors. Good ones, anyway. A bad author doesn't think logically, but Neal Stephenson? Good one. Cixin Liu? Mostly good, although don't bother with the third book in the Three Body Problem trilogy.
One of the more disturbing aspects of modern politics and society is the post-truth nature of dialog. This is something that we can try to discuss with scholarship. We can examine how people come to believe conspiracy theories, what happens when you expose people to facts, how ideological siloing takes place, and so forth. We are reacting piecemeal, though. What Stephenson was able to convey was a deductively structured world, thinking through the bigger picture in a way that scholars are not generally trained to do. (And that was only part of the novel.)
What authors do is create a world. By definition. Even if the world runs analogously to our own, it is still a world. In Charles Stross's Merchant Princes series, the entire concept is about parallel universes, and the universe in which Miriam initially lives looks like it is intended to be ours. Spoiler alert: it ain't. There's a big event a few books in that makes that timeline diverge dramatically from our own, and the point is to trace out effects. Deductively.
We, scholars, are usually working inductively, even when we pretend otherwise. Yeah, game theory, whatever. I'm still basing my mathematical models on empirical observations and posing the notion that the past will repeat itself. That's induction, and a demonstration of where the lines get blurry. Authors, though, are the ones who are working deductively, to construct a world, and address the bigger questions, allowing them to see what happens if the world diverges in bigger ways.
Look around. Does the world seem like it is getting weirder? Are we... diverging? Diverging from past patterns? Breaking from historical norms?
The extent to which that happens is the limitation of social science. We think small. We must because we are constrained by methodology. Larger changes require bigger thinking, but we aren't good at that.
Good authors think bigger than social scientists.
Comments
Post a Comment