Since the dawn of time, mankind has been inclined to hate the player. But what if I told you it’s not always the player’s fault, and sometimes you should instead hate the game?
This is the basic idea behind systemic critique, as contrasted with the moral critique of individuals.
The target of a moral critique is a “bad apple” who has made naughty choices—the goal is to reform that individual, or if that’s not possible, to keep them out of power. The target of a systemic critique is a broken or unjust institution—the goal is to change the rules of the game, not necessarily the hearts of the players.
Moral critiques come very naturally to us human beings, because we like to see people’s behavior as an expression of their personal values: if someone’s acting like a jerk, we assume they must be a jerk. For example, if you look at the American Congress, you see a lot of nutty behavior. It’s almost irresistible to think, these people are nuts! Or you might read about a case of police brutality—this guy is a bastard! Or when you see that Mario Kart is $80—Nintendo’s greedy!
To be clear: you might be right.
But there might also be a deeper, systemic reason for the behavior you’re seeing. For example, there might be a filter mechanism that only allows certain types of people to make it to the halls of Congress—only the top-tier shmoozers, say. There might be broken feedback mechanisms that prevent bad cops from being removed or disciplined—as in American courts, where over 99.5% of cops found liable in civil rights cases don’t have to pay a penny (per Joanna Schwartz’s analysis). Or there might be role-based incentives for people running a company—they get rewarded for big profits, not happy customers per se.
Whenever there’s a systemic explanation like this, it would be a mistake to limit yourself to a moral critique, on pain of missing the forest for the trees. For even if you do manage to “throw the bastards out” of Congress, that won’t fix anything if the next batch still has to fit through the same busted filter. In general, moral fixes won’t solve systemic problems.
This case for systemic critique, which I associate with Alan Garfinkel’s cult classic Forms of Explanation, is one of the most mind-blowing ideas in all of social theory. I don’t know how you can hear it and not think, Damn, that’s cool.
But like any form of discourse, systemic critiques can be done badly. In fact, I’d wager that most of them you see in the wild are worse than useless. The problem is that system-blaming is irresistible for people who want to sound deep but are incapable of having deep ideas. It’s also a cheap way to place yourself beyond moral criticism.
Today, if only to reduce the number of dumb slogans on Etsy products, I’ll lay out three common problems with lazy system-blaming.
Problem 1: Actually Bad Apples
The first problem with always blaming the system is that bad apples really do exist. This fact, as noted earlier, is consistent with there being deeper systemic reasons for the badness of this or that apple. But sometimes the bad apples deserve all the blame. You might have a system whose rules are optimal, but where the people in charge are just absolute monsters. Then reworking the system will only make things worse—the political equivalent of bloodletting. You’re changing the system to solve a problem that’s happening despite the system, like draining blood from a patient who’s suffering despite having a healthy amount of blood.
For example, even the biggest fans of democracy have to admit that it’s logically possible for the voting public to be overwhelmingly composed of vicious, narrow-minded fools. In that scenario, even an optimally designed election would result in a terrible leader. Or suppose you have the opposite philosophy, and you believe the system works best with a benevolent strongman in charge. Very well—suppose he goes nuts?
The problem is perfectly general. Any system for regulating human behavior requires some degree of knowledge, prudence, and good faith on the part of the humans being regulated. There is no social structure that can alchemize pure vicious stupidity into 24-carat prosperity and justice. Indeed, the entire point of having social structures is to economize on the precious specks of knowledge, prudence, and good faith that people do possess. You can blame a system for squandering human goodness, but not for failing to perform miracles. (Come to think of it, even Jesus needed water to make wine.)
That’s problem #1: whatever the right system design, you have to admit that there are some cases where it’s wrong to blame the system rather than the individuals therein. (Unless you’re assuming a society of “angels,” in which case, as James Madison counsels in Federalist 51—and here I’m paraphrasing—“Get a life, nerd.”)
Problem 2: Trans-Systemic Flaws
This one’s deeper.1
Let’s say you’ve put your finger on a flaw in the system, something that just can’t be blamed on individual foibles. Even so, it might be wrong to blame that system for the flaw, since the flaw might equally afflict every alternative system.
For example, let’s say you’re on the far left or far right, and you hate your country’s democracy because it fails to deliver electoral results that satisfy your preferences. You complain: the system fails to give non-centrists what they want. This could be true! It may well suck. But there is, very obviously, no system for running elections that can give both extremes what they want simultaneously. A system that caters to the far left will necessarily be thwarting the far right—and vice versa.
Here’s another example: health insurance and end-of-life care. Back in the day, many conservatives objected to Obamacare on the grounds that it would empower government “death panels” to decide who gets to live or die. Sarah Palin pressed the point in a 2009 Facebook post:
The America I know and love is not one in which my parents or my baby with Down Syndrome will have to stand in front of Obama’s “death panel” so his bureaucrats can decide, based on a subjective judgment of their “level of productivity in society,” whether they are worthy of health care. Such a system is downright evil.
Now, some of this critique is meant to be specific to Obamacare—the idea that the government is making choices about how to ration care, and that it will be making these choices on dubious grounds. These specific allegations turned out to be false.
But the essence of the critique, the part that resonated, was that Obamacare would involve any sort of death panel at all. It just seems bad to have some group of people that gets to decide how to ration out care, especially when the patients are elderly or disabled.
This critique isn’t false, exactly—just shallow. Yes, it sucks that we have to ration healthcare. But what’s the alternative? Infinite free healthcare for everyone? At some point, we are going to run out of emergency rooms, geriatric specialists, and medicine. Then we face the question of who gets the last room, the last appointment, the last dose. There has to be some mechanism for deciding this. And unless we want to allocate healthcare based on Rock, Paper, Scissors, we’re going to need intelligent agents making choices about who gets what. You can call these “death panels” if you want, but then you have to admit that there were already “death panels” pre-Obamacare. They’re called insurance companies.
(Yes, I know, they’re constrained by government regulations, market forces, etc.—but you get the point.)2
Problem #2, then, is that some systemic flaws transcend the system. If you blame a particular system, you risk missing the forest for the trees, much like the moralist who blames the bad apple without noticing the background process that makes good apples impossible to grow.
This may remind you of Ken Arrow’s legendary “impossibility theorem,” which shows that certain principles of democracy can’t be jointly satisfied by any possible voting system. (See below if you want an intro.)
Strictly speaking, impossibility theorems don’t indicate that every system has some particular flaw. Instead, they show that certain flaws can only be avoided at the cost of creating new ones. For example, Arrow’s theorem shows that you can’t have a transitive voting rule unless you’re willing to pay some cost—say, having a dictator. This is Problem #2 in its full generality: some systemic problems are the result of an inevitable tradeoff, rather than a mere blunder in design. Switching systems will just mean swapping hardships.
Problem 3: Switching Costs
Finally, even if you’ve got a valid systemic critique, and your system really does have some flaw that makes it worse than the alternatives, that still doesn’t mean that switching systems is necessarily a good idea.
For one thing, it’s costly to switch. As Daniel Greco reminds us: coordination is hard! There’s a reason why Americans haven’t even been able to move over to the metric system.
Also important is that switching can be risky. Conflicts tend to break out, not just between the revolutionaries and the status quo defenders, but between different factions of the revolution who seek to end up on top—if we’re switching to a monarchy, expect a fight over the throne. Related to this is the problem of instability. Even if you’ve managed to come out on top after the revolution, now you have to convince a bunch of revolutionaries to stop with all the revolutionary business and start playing by the rules. (Your mileage may vary.)
Finally, you might find you were overrating the merits of your new system. “The grass is always greener” is timeless wisdom for a reason.
Conclusion
Hopefully this is all so obvious that you didn’t need to hear it.
But with the amount of braindead, overconfident political content sloshing around these days, maybe it’s worth reminding ourselves that system-change is always a gamble. We don’t want to blithely accept the status quo—but that doesn’t mean we should go all in for costly bloodletting, either.
David Hume once called his own preferred reforms “dangerous novelties.” I like that. If we never try out novel systems, we’ll never improve. But if we’re blind to their dangers, our experiments could easily fail.
Because here’s the thing no one likes to hear: some existing systems are, in some notable respects, actually pretty good. They’re not ideal. They’re mostly not even Pareto optimal. But in this era of relentlessly negative sentiment, it is worth emphasizing that things used to be (and could easily get) a lot worse. We are lucky not to be living in the midst of a catastrophic world war. Global poverty has been cut in half, then in half again, then in half again. It used to be that 50% of everybody died before age 15. Now that figure is 4%. Hundreds of millions of people worldwide enjoy amenities (like refrigeration and electricity) and freedoms (like the vote) that were simply unthinkable for the commoners of 500 years ago.
So by all means, hate the game. But please remember not to play yourself.
I associate this problem with Joe Heath’s book Cooperation and Social Justice. For my thoughts on Joe, see the start of this:
By the way, the reason why I said “intelligent beings” rather than “humans” is that Dr. Oz and the Trump Administation have begun rolling out a pilot that lets AI assist in decisions about whether to give “prior authorization” to Medicare patients seeking treatment.
This post is such a banger!
I feel like by far the most common problem with systemic critiques is that people are bad at identifying the problems in a system.
Eg it's very unlikely that the current level of theory vs empiricism in academic econ is optimal. So the academic econ system is likely "systemically flawed"
However it's very difficult to identify the optimal level of theory in an academic field, and mildly difficult to identify the current level of theory in an academic field.
Many critics of economics completely fail to do even do the second, much easier, problem[1]. Nor do they even attempt to articulate a precise definition of the first. So their critiques (of perfect competition or rationality or whatever) are useless but somehow manage to sound wise to other unthinking critics.
This problem is unusually sharp in academic econ imo because it's an unusually good example of the critique being inverted from reality: the field is arguably being too empirical these days, yet "everybody" complains about it being too theoretical. But you also see eg people blaming AI companies for water usage, capitalism for housing shortages, academic intersectionality for anti-white racism, FPTP for political polarization, etc. In these cases these are systems that have genuinely major problems, but the most popular critiques are entirely off-mark, for reasons that shouldn't be too difficult to understand, and yet people somehow fail to do so.
[1] eg when I confront popular critics of academic econ, most of them haven't even do the single most obvious search possible relevant to this question: https://www.google.com/search?q=graph+of+theory+vs+empirical+studies+in+academic+econ+papers