Slate Star Codex

I.

Thanks to everyone who commented on the posts about conflict and mistake theory.

aciddc writes:

I’m a leftist (and I guess a Marxist in the same sense I guess I’m a Darwinist despite knowing evolutionary theory has passed him by) fan of this blog. I’ve thought about this “conflict theory vs. mistake theory” dichotomy a lot, though I’ve been thinking of it as what distinguishes “leftists” from “liberals.”

I went through the list of “conflict theory says X, mistake theory says Y” nodding my head and hoping that you and everybody else reading it had the same impression as me -that both theories are important and valuable frameworks through which to view the world. There are definitely common interests that everybody in America shares, and there are definitely some pretty significant conflicts of interests as well.

The reason that I do identify as a leftist and sometimes feel like an evangelist for conflict theory is that I get the impression that most people don’t even have conflict theory in their mental framework. Leftists all understand what the “we’re all in this together” liberal viewpoint is, while even incredibly smart and on it liberals like yourself can go for a long time without even thinking about the “politics is the clash of interest groups with conflicts of interest” leftist viewpoint.

I’m starting with this one, before I get to all the criticism and objections, to insist that there is some core worth saving here. To everyone who says this was obvious, I can only plead that you listen to all the people saying it wasn’t obvious to them at all. There are probably a lot of things wrong with how I described it, and the remaining comments will go over a lot of them, but I can only make the excuse that I’m groping towards something useful without any of the people who claim to already understand it perfectly helping me at all.

HeelBearCub writes:

First off, I think Scott’s post is a good post that seems to make good points and takes steps on the path to true knowledge. But, as I read it I kept thinking about Indian wise men describing the elephant. Or rather, I pictured two amateur carpenters arguing about how wood is joined together, one being a “nail” theorist and the other a “screw” theorist. Now obviously, you build using wood or nails, but structures out in the real world are not built exclusively using one or the other. This seems to me to be a recurring failure of thinking on Scott’s part, honestly. The tendency to naturally think in binary terms (even while knowing this is incorrect).

The goal of those posts wasn’t to present a perfect understanding and tell people that was it, but to gesture at a concept that needed refining.

By analogy, should we ever talk about the left vs. right axis in politics? Isn’t that just a “dichotomy”? After all, almost nobody is 100% leftist on everything or 100% rightist on everything.

What about introverts and extraverts? There’s no 100% introvert or 100% extravert. Everyone is a combination of introverted in some situations and extraverted in others. That doesn’t mean we should never use these terms, or that psychologists are being too binary and dichotomizing. It means we establish the terms so that everyone knows what we’re talking about, and then discuss where everybody is on the axis in between them.

I tried to exaggerate conflict and mistake theory equally, making it obvious I was presenting caricature versions to be filled in later. I think I was at least equally unfair to mistake theorists – presenting them as believing there’s no such thing as selfishness, as thinking that surely tobacco companies just deny their products cause cancer because their CEOs have an genuinely different interpretation of health statistics. The point isn’t “mistake theory is good and normal and conflict theory is bad”. The point is that both look ridiculous at the extremes and everyone combines them in some (different) proportion in reality.

II.

Tumblr user unknought writes:

Principal-agent problems, rent-seeking, and aligning incentives are things that socialists do talk about. Like, a lot. But even if they weren’t, it’s totally bizarre to represent these as mistake theory concepts. All three of these are concepts which are used to describe ways in which we don’t all want the same things, and how agents in positions of power whose goals don’t align with the common good can fuck things up for the rest of us. If conflict theory means anything, it means that.

On my hospital analogy for why mistake theorists like free speech:

So like, if you learned that your doctor’s recommendation was influenced by Pfizer owning the hospital and restricting doctors’ choices and suppressing information about their medication…wouldn’t that be an excellent reason not to be a mistake theorist in this case? Like, this is literally an example of where you can’t trust expert opinion because things are being controlled by a powerful entity whose interests do not align with the common good. This is about as clear-cut an example of conflict theory getting it right as can be imagined.

They quote another Tumblr user who I’m not sure I have permission link directly, responding to my analogy of the shill scientist with a PowerPoint:

In this scenario, even the strawman conflict theorist acknowledges that the people who believe the powerpoint are mistaken, and could be won over through reason and debate, thus undercutting the narrative scott is presenting in which strawmen ‘conflict theorists’ are generally uninterested in debate…the whole tendency to caricature opponents as anti-rational also plays in to his assumption that conflict theorists would be anti-intellectual, which ignores that in the context of a conflict, having intelligent people working on your side to win the conflict is obviously desirable…

even if i do think scott amounts to the hypothetical “Elite shill” with a powerpoint saying yellowstone will erupt, proving his claims wrong on a technical level is still my first and most important priority if i want him to not be a successful shill. simply accusing him of being a shill, and pointing out the $1,627,000 which MIRI (which scott is affiliated with) has received from Peter Thiel would be useless if i couldn’t also demonstrate to people that he was incorrect. the insinuation that viewing a particular disagreement as a conflict of interest presupposes any possibility of engaging on a factual level is ludicrous. even if i think the “conflict theorist”/”mistake theorist” is just a cheap rhetorical trick to sell people on anti-democratic ideology, i still need to demonstrate that it’s not reflective of material reality.

(one correction: I am not affiliated with MIRI. I spent a few weeks doing a minor writing job for them five years ago; I have had no formal relationship since then besides donating money)

The three comments above all seem right to me, and seem like the best examples of how my intuitive concepts of “conflict theorist” and “mistake theorist” fail to be captured by the ways I described them.

How fatal is this objection? A lot of things aren’t accurately captured by the words we use to describe them. Consider again the analogy of the political spectrum. Someone might say “You say Republicans are conservative, but they don’t want to conserve the traditional public school system.” Or “You say libertarians freedom, but what about freedom from powerful corporations?” Yeah, okay, you got me there, sometimes words are not perfect 100% handles for complex concepts. There’s got to be a balance between having simple words that mean exactly what they say, and accurately tracking the weird subtleties of how people really associate.

A lot of this discussion is conflating “conflict vs. mistake” with “protester vs. wonk” and “Marxist vs. neoliberal”. Sometimes they’re being conflated by me, where I say something is conflict theorist when I mean Marxist, or vice versa. Other times they’re being conflated by the commenters, who say “You say X is conflict theorist, but Marxists don’t actually believe X!” Well, maybe that’s because Marxists aren’t 100% stereotypical conflict theorist. The word “libertarian” can shed light on Ron Paul, even though he supports government intervention in some areas.

Obviously this idea isn’t useful if it totally fails to correspond to existing politics. But I think there are more subtle (read: dangerously susceptible to just being overfitting) ways of understanding this that make sense of these contradictions. For example, conflict theorists can certainly engage in intellectual debate if it helps them win. That doesn’t mean there’s not a difference between people who choose their side based on intellectual debate, and people who engage in intellectual debate if it helps their side. Also, it’s unclear whether a conflict theorist thinks intellectual debate is better than equally persuasive propaganda, whereas a good mistake theorist definitely should.

Likewise, any mistake theorist who didn’t acknowledge that there are lots of self-interested parties would be an extraordinary idiot. The mistake theorist calls these “special interests”, which I suppose is a pretty-loaded term suggesting they’re a few weird exceptions to the rule of supporting the general interest. The question can’t be whether special interests exist – obviously they do. It can’t even be whether they’ve seized control of the government – I think this is something of a consensus position right now. Maybe it’s something like “Is there anything other than special interests?”

In a book review a while ago, I talked about figure-ground inversions:

An example of what I mean, taken from politics: some people think of government as another name for the things we do together, like providing food to the hungry, or ensuring that old people have the health care they need. These people know that some politicians are corrupt, and sometimes the money actually goes to whoever’s best at demanding pork, and the regulations sometimes favor whichever giant corporation has the best lobbyists. But this is viewed as a weird disease of the body politic, something that can be abstracted away as noise in the system.

And then there are other people who think of government as a giant pork-distribution system, where obviously representatives and bureaucrats, incentivized in every way to support the forces that provide them with campaign funding and personal prestige, will take those incentives. Obviously they’ll use the government to crush their enemies. Sometimes this system also involves the hungry getting food and the elderly getting medical care, as an epiphenomenon of its pork-distribution role, but this isn’t particularly important and can be abstracted away as noise.

I think I can go back and forth between these two models when I need to, but it’s a weird switch of perspective, where the parts you view as noise in one model resolve into the essence of the other and vice versa.

I think any difference between mistake and conflict theorists on this issue has to be of the figure-ground inversion type. This doesn’t have to be about the number of people who are vs. aren’t special interests. It can be about the amount of each person’s soul devoted to their special interest as rich/poor/white/black/rural/urban, vs. the amount of each person’s soul devoted to loving truth and doing good. Framed this way, it sounds like we’re pretty much screwed, except that free rider problems might be working in our favor here.

A related difference: because mistake theorists think there’s some stable ground other than conflict, they picture themselves as potentially neutral referees (even better: Geneva Convention delegates) looking for ways to circumvent the conflict. Conflict theorists just want to win it.

Some examples of circumventing conflicts: free religion, free speech, federalism, reliance on scientific consensus. Free religion because instead of pushing for any one religion to win, you just create a system that defuses the threat of religious violence. Free speech for the same reason. Federalism because instead of saying that any one side wins, you just tell Side A to do it their way over there, and Side B to do it their way over there (even better: polycentric law). Reliance on scientific consensus, because instead of arguing over whether to have school vouchers or not, everyone just agrees to have unbiased scientists do a study and trust their findings.

I notice Jacobite’s describes itself as “the post-political magazine”. Politics is about having conflict. Mistake-theorists would love to become post-political, in the sense of circumventing all conflicts. Conflicts actually happening as conflicts is a failure, deadweight loss. This wouldn’t mean that nobody has different interests. It would mean that those different interests play out in some formalized way that doesn’t look conflict-y. Think of the Patchwork / charter city / seasteading dream, where there are lots of different polities and people vote with their feet for which one they prefer. Protests, bring-out-the-vote efforts, arguments – all have become obsolete. These ideas don’t deny the existence of conflict – they just represent a desire to avoid it rather than win it. Conflict theorists could theoretically want to avoid conflict, especially if they think they’d lose. But most of the ones I’ve met think that avoiding conflict is a better deal for the enemy than for them, and so would rather just have it and see what happens.

I think this is why I think of public choice theory and its relatives as basically more mistake-y. It’s not just that they sometimes say government doesn’t work, get seized on by libertarians, and so get a bad reputation among Marxists. It’s that they take this god’s eye perspective of trying to micromanage the rules of political conflicts instead of winning them.

III.

Some other good comments:

Fluffy Buffalo comments:

What struck me in your post was that the examples you gave for conflict theories all came from the Marxist perspective. While (cultural) Marxists may be the most obvious, unabashed conflict theorists these days, the behavior of the American right wing looks like they have their fair share of conflict theorists, and Republican tax and health care policy often smells more of an undeclared class warfare than of careful consideration of the pros and cons.

It’s definitely true that there’s a lot of right wing conflict theory and that my post mostly ignored this. They continue:

This looks not like a fundamental question of what the world is really like, and more like a multi-player game theory problem, in particular a multi-player prisoner’s dilemma. It’s all fine and dandy – in fact, it’s probably the most constructive, helpful thing to do – to play “mistake theory” if everyone else is playing the same game, but if you have a sufficiently strong faction playing “conflict theory” (refusing to compromise, because everyone else is the devil), they have more success than they should. “Conflict theory” is like a bad Nash equilibrium, a self-fulfilling prophecy – if everyone behaves according to the diagnosis “It’s power-hungry, uncompromising people on the other side who cause the problem”, there will be no lack of power-hungry, uncompromising people on all sides, causing all sorts of problems.

David Friedman makes the really bizarre topsy-turvy argument that conflict theorists should be people who want minor tweaks to the existing system, and mistake theorists people who want to get rid of it entirely:

A lot of whether you believe in mistake theory or conflict theory depends on how many mistakes you think there are, which comes down to how badly you think existing systems work. If you believe that government decisions are very far from optimal, that FDA regulation has little positive effect on how good drugs are, a huge negative effect on what drugs are developed and how expensive they are, and believe it not only from observation (Peltzman’s old article, for instance) but as what you would expect from the incentive structure, then you see the main problem as persuading the 99.9% of people who are worse off as a result. If you think government decisions are close to optimal, on the other hand, then all that is left to change is how they weight payoffs to different people, which is a conflict issue.

I guess this reinforces the point from above: taking some of these terms literally, or trying to reason with them, gets very different results from looking at how they operate on political coalitions.

From Jan Samohýl:

I really enjoyed the post, the two worldviews are interesting. But as many people have noted, things are often not so clear-cut. I would like to point out some interesting connections.

I think of myself being a more conflict-theorist, probably being on the left. But in many cases, I look at the world as mistake-theorist, having quite liberal views. To me, it’s somehow related to the idea “expect the worst (conflict), hope for the best (mistake)”. I am a big fan of direct democracy (which I see as a way to peacefully resolve innate conflict), and less of a fan of resolving politics through debate (which implicitly assumes the debaters are honest). To me, direct democracy is related to trust, as it is understood in computer security, for example. In computer security, trust is not given, it is earned and can be revoked at any time. You can only trust somebody who you don’t have to trust. In computer security, we assume worst intentions (conflict), not the best ones (mistake). And so I believe direct democracy is needed as a baseline, because the politicians are not to be trusted (principal agent problem).

This is another weird topsy-turvy thing. I was thinking in terms where mistake theorists are afraid of direct democracy because any random demagogue can sabotage it, and trust more in “post-political” systems like the ones mentioned in Part I above.

From Nootropic Cormorant:

A Marxist reader of your blog reporting in! I feel that this way of analyzing things is harmful (a charitable mistake) because ironically it conflict-theorizes the debate so that suspected conflict-theorists are automatically seen as beyond object-level discussion leaving you with no options but to act like one yourself. I would question whether this distinction is even relevant to people and ideologies rather than to situations and debates. Many of the commentariat have identified themselves as mistake-theorists, but I have to wonder whether this is because they abstain from the sin of conflict-reasoning or because it bothers them when the outgroup does it. This is some conflict-theorizing on my part. Also the willingness to discuss policy issues, aka the reformism vs. revolutionism axis, should not be conflated with this as one can go along it in a purely mistake-theorizing framework.

From Ben Wave:

I count myself as a Marxist, as someone who adores your blog, and as someone who primarily uses what you’ve deemed mistake theory as my lens on the world. It’s naive to expect that everyone else would or should have the same terminal values as I do, or that everyone should agree on the relative role of government in bringing those values about. So I don’t. I don’t share the view of some of my comrades who celebrate death and misfortune upon the rich. I do share their desire to take power from the powerful, and create a world in which the weak have more power. I desire this independently of a desire to increase living standards/lifetimes/happiness over all, and I desire this for three reasons – the first is that I see high levels of inequality as an existential threat to society. The second is that I would not consider the current distribution fair, if I was to be incarnated into a human chosen at random. The third is that the more unequal society is, the less the ethical premise of capitalism (that it is right to reward those who fill market needs because they are equivalent to human needs and desires) is valid. Which I feel is important seeing as it is the dominant method by which we decide what gets done by the ensemble of human endeavour. As regards the conflict view, yes many comrades use that. I see it as largely unhelpful, but I’m sympathetic to their situation. Marginalised people are often harmed or killed by inequality (medical, prison and law-enforcement, unsafe conditions of work or home life), so conflict is the reality that some of them face. I find it more productive to try and cooperate with my opposites who like cooperation than to fight against my opposites who do not. Hopefully the coalition of those willing to negotiate can improve conditions globally. Gather with those who share your terminal values. Gain power through that unity, and use that power in negotiation. Stay open to new information to best pursue your goals. I guess that’s the dream.

I find this interesting except that I can’t figure out why marginalized people being harmed by inequality, bad medical care, bad working conditions, etc makes conflict “the reality” for them. Surely it means bad things are the reality, and we’re back to the original question of whether bad things are mostly due to conflict or mistake?

From bosun_of_industry:

You position these in opposition or at least orthogonal to each other but I think that’s missing both how often they are layered and/or modal. Moreover, the Marxist portion of this essay is red herring carried forward from the Jacobite article and does it a disservice because it lends it a uniformity to any given actor’s approach. Instead, a common split is people approaching policy issues as mistake theory when within-group and as conflict out-group. There’s distinct strategic reasons for this sort of behavior, especially within a winner takes all political system. Similarly there’s a strong history of making a public show of conflict theorizing while acting as a mistake theorist in private amongst politicians and activists alike and for similar reasons.

From Alkatyn:

To go up a meta level, maybe conflict theorists are mistake theorists who have applied the same methodology to “how do we get our ideas implemented” and decided based on the empirical evidence that emotional appeals and group action are more effective than policy analysis.

The effective altruists have a lot of discussion on what the most effective way to push their preferred policies are, and usually they settle on some combination of appealing to academics and founding think-tank style organizations. I have noticed this working very well so far. This piece on how neoliberalism caught on is a really interesting example of doing this right and effectively. I suppose the conflict theory answer would be that this only works for policies that elites are already predisposed to like or at least not resist.

From John Schilling:

One thing that seems to be getting lost in a number of places here, so I’ll just address it this once. Mistake Theory doesn’t require denying that Rich Plutocrats are genuinely trying to further enrich themselves at the expense of the poor and that the Poor are trying to tear down the rich out of spite righteously take back what is theirs. More broadly, it does not require denying that people will selfishly pursue their selfish interests to the detriment of others or of the whole of humanity, does not simplify to “if we are all rationalist altruists the right answer would be…” Mistake Theory would hold that in almost all real conflicts, the best outcome for everyone is a negotiated solution and that the relevant facts (including the balance of power between competing interest groups) makes the range of plausibly negotiated agreement reasonably narrow. So failing to sit down and quietly negotiate that agreement, instead escalating to pointless conflict, is usually a Mistake and often an Easy Mistake. Figuring out what to do about people who persist in making that Easy Mistake, is the sort of problem that often leads to Hard Mistakes and sometimes to solutions that look like Easy Conflict.

Butlerian writes:

I think that the conflict/mistake distinction is being vastly overblown here. Scott is proposing some sort of fundamentally different personality type / rhetorical diachotomy between agents. I think these differences in behaviour can be explained in a much more proximate way, instead simply being answers of “Yes” or “No” to the question: “Do you believe that shills are amongst us RIGHT NOW”? If I am in a discussion where my answer to that question is “No”, I will discuss in a mistake-theorist manner: belief that other participants are arguing in good faith, and personaly arguing in good faith, and sticking to truth-searching, and at least trying to be open to the possibility of changing my mind. If I am in a discussion where my answer to that question is “Yes”, I will discuss in a conflict-theorist manner: having no genuine interest in the merits of suspect participant’s arguments, reading them only with an eye to finding errors / weak-man-able faultlines, and refusing to allow myself to change my mind even in the face of apparently convincing evidence because I strongly auspect that the evidence is the contaminated product of Yudkowsky’s Clever Arguer. I would think that this is obvious. When Marxists / feminists / Nazis are in their safe-space forums, and are happy that everyone in the discussion is a like-minded Marxist / feminist / Nazi who genuinely shares their endgoal of a well-functioning communist / non-cis-hetero-patriachal/ white ethnostate society, they will have amongat themselves truth-seeking mistake-theoretical diacussions. It is only when they go out into enemy territory and find themselves surrounded by perfidious capitalist / misogynist / Judeobolshevik agents that they switch to conflict-theoretical mode.

This reduction to “belief in shills” as the most important difference between the theories is interesting.

Aside from the people who wear it on their sleeves, like PR people and lobbyists, I find myself practically never believing anyone is a “shill” in the strong sense – ie they don’t believe the arguments they’re making, but they make them anyway for some personal advantage. Even when people do very shill-ish stuff – like put out a biased paper for a think tank that promotes its agenda – I assume the think tank probably just recruited people who already believed in their agenda, and let their personal bias do the rest.

In fact, the idea of “shill” is really complicated here. People accuse me of being biased or selfish all the time if I make arguments that seem to benefit rich people, or Jews, or white people, or any other category I’m a member of. But by far the most important thing that affects my finances and social status right now is any threat to the guild-based nature of the medical profession. And I’ve consistently argued in favor of things like letting psychologists prescribe, even though this will cost me money and status way more directly than anything that happens regarding class or race. If people were interpreting shilling in its most literal sense, they would get surprised that I’m not writing 100% about how great medical guilds are, finally assume I was insane or some vanishingly rare moral saint, and everything about race or class would be so irrelevant to this calculation that it would never come up.

The fact that nobody thinks this way has to be tied into some sort of answer to the class warfare free rider problem. Nobody is a first-level shill literally pushing their own self-interest, but weird class-consciousness/ideology style forces give people biases (mostly based on race or class) that they push at least somewhat honestly. I think this is at least sort of in accord with some forms of orthodox Marxism.

But this sounds a lot like…mistake theory. If people push their policies because they’re biased into thinking that’s the morally correct thing to do, then surely solving their biases and convincing them otherwise could change their policy preferences. Is there anyone who doesn’t believe this model? If so, what exactly are we talking about?

And from christhenottopher:

All this talk about how Marxists don’t frequent this blog and you go and make a post arguing for a Hegelian dialectic. Yes, yes of course this blog has always been on the side of Thesis. And certainly we must always beware the great enemy Antithesis. But let’s end the essay by arguing that what we really need is Synthesis! The professor in college who taught me what dialectical reasoning was warned the class that once we understood this, we’d see dialectics everywhere. And once again he was proven right.

This had BETTER NOT BE TRUE. If all of this time all this incomprehensible stuff about dialectics was just basic “start understanding a concept by giving binary examples of opposite sides, then correct it and make it more sophisticated later”, I am going to be SO ANGRY.