scientists depend on what rationality researcher Keith Stanovich calls “cognitive decoupling.” Decoupling separates an idea from context and personal experience and considers it in the abstract. It is the approach used in the scientific method, when performing thought experiments, and when generalizing principles from individual examples. . .
The contrary mode of thinking sees every argument embedded in a particular context. The context of an idea includes its associations, implications, and the motivations and identities of those who advance it.
Pointer from a commenter. I recommend the entire article.
My thoughts:
1. The “contrary mode of thinking” strikes me as post-modernism.
2. It seems to me that “decouplers” and post-modernists must unavoidably talk past one another. The decoupler thinks that decoupling is necessary for rationality. The post-modernist thinks that decoupling is impossible.
3. This language can be used to re-cast my 2003 essay criticizing Paul Krugman. We might now say that my claim was that economists are trained to decouple, and instead he was doing the opposite.
4. The term “outgroups”in the article is illustrated by a chart in which coastal progressives are the primary outgroup for the Intellectual Dark Web. Of course, since coastal progressives wield the post-modernist cudgel.
5. In fact, I had come to think of the left as always refusing to decouple. But there are those on the left who are ardent decouplers, and they are in the IDW. I suppose that by the same token there must be people who, even though they are on the right, routinely deploy post-modernist arguments.
I still have not finished processing the Falcovich piece, and again I thank the commenter for the pointer. This will not be my last word on the topic.
Several people have articulated this same dichotomy. Scott Alexander talked about it as conflict versus mistake. I think this twitter thread is particularly illuminating.
I believe it’s an underappreciated dimension of personality difference. Perhaps someone could create a test to assess it. I think the “conflict” or Machiavellian approach is the default, natural human mode of thinking, while the “mistake” or “decoupling” approach is much less natural and is exemplified by scientific thinking.
While decoupling is of course less relevant if your research topic is, say, molecules or galaxies rather than humans or social institutions, I wonder if differences in this dimension among researchers explain, to some extent, the relative lack of progress in soft versus hard sciences, i.e. the natural sciences attract more people who are better at decoupling.
“In fact, I had come to think of the left as always refusing to decouple”
Once you start thinking of left or right poles, you are by definition refusing to decouple.
No one is allowed to stay strictly in the decoupled realm of the rational. If anyone wants to try it, go ahead. They will quickly find out they will have no political traction. Almost any successful person has to get a mixture of both channels right.
Trump is an interesting case because he is the first politician in the American system to completely drop any pretense of a rational channel, and he has been shockingly successful doing so.
Wow, Tom- so there is literally no chance that Trump is successful by doing what you claim successful people usually do?
Can you cite a single instance where Trump made a “decoupled” public argument? Where he laid out a case rationally, separating an idea from context and personal experience and considering it in the abstract? Where he addressed the primary counter arguments without attacking motives of those that disagree?
Expression, then deliberation. In Herodotus, writing about the Persians deciding to go to war- A night of drunkenness followed the next evening by sober deliberations. In courtrooms, discovery then judgment. In Congress, fact-finding missions and hearings followed by politicking. In brain-storming done right- get all the possible ideas out there no matter how nutty, then hack them to bits and re-assemble. For some economists life gets lived in reverse here- establish prestige in deliberative academics then get pulled into policy and politics where the emphasis is on consequences and media circuses. Post-modernism is an attempt to improve the behaviors of the past by heavily subjecting deliberative processes to expressive politics (misguided I think.)
“The post-modernist thinks that decoupling is impossible.”
When people push a standard that is impossibly impractical or silly, it’s suggestive evidence that they’re going to use it as an inconsistent double standard so they don’t need to believe it, just trumpet it. With some impossible standards, like witch hunt standards of evidence, this tends to play out immediately. With others it takes longer, but we still see it play out over longer periods over and over again: George Orwell and Ayn Rand were not in exact agreement on every point, but their catchy “some animals are more equal than others” and “aristocracy of pull” resp. seem to be making the exact same point about this.
Non-evasive, non-vacuous versions of the cod po-mo “meaning is impossible” theme are both impractical and silly. They are particularly amusingly silly on the web, because the web and the underlying Internet protocols work because IETF RFCs have sufficiently clear meaning that people can use them to make equipment that interoperates reliably in a very complicated way.
Witch hunt standards are not exactly a fringe problem, nor is escalating privilege and corruption in the name of equality of outcome is not either. Cod po-mo isn’t either, though you need to look a little harder, to what people say to justify legal decisions. E.g., I “like” the Sotomayor “wise latina” speech https://www.nytimes.com/2009/05/15/us/politics/15judge.text.html for this. I read the speech because I thought it was suspicious that I had seen the passage out of context too many times. Sadly, the context doesn’t help, just contains more toxic crap
. In particular, Sotomayor approvingly invokes an impossibly high word-has-no-meaning standard for a claim she wants to discredit
There is fundamentally no meaning for her opponent’s “wise”, but in the very next sentence — setting a high standard for playing out immediately — Sotomayor herself freely says “better” and expect to be understood without having to do something sophisticated to meet the impossibly high standard of meaning she has just applied to her opponent’s term, because Sotomayor only applies that standard selectively.
So I suggest “the post-modernist claims that meaning is impossible” tends to be closer to the truth than “believes”.
It does often happen these days that capable people think seriously about the fundamentals of meaning being difficult (or totally dependent on context, or other such po-mo-sounding limitations), but that is not called po-mo but “information theory” and “coding theory”, working with numbers and yielding practical results that are important to making things work, e.g., making your cell phone work. Credentialed specialists in dogmatically innumerate fields playing games with what is essentially the almost-completely-false dichotomy between 100.0% confidence and 99.999999% confidence (and very seldom making it qualitatively clear that one can asymptotically approach knowledge, and essentially never bringing the numbers) are generally unserious and, based on their actions, likely to be insincere as well, whether they are “scholars” in tax-funded back-scratching grant arrangements or oathbreaking judges.
Martha Minow is a really interesting example.
She doesn’t call herself a creationist. But there are millions of NPR creationists and arugula creationists who don’t see themselves as creationists. They’re wedded to the idea that creationists are other people.
If Martha Minow had called herself a creationist she would never have become the Dean of Harvard Law. Equally if she had defended the reality that our species is a product of evolution and not some exception to it, then in that case too she couldn’t have been Dean. Larry Summers couldn’t be President of Harvard without bowing and scraping before the creationists at Harvard. Social justice creationists are scary.
She says she believes in both “intellectual pursuit and social justice” but she doesn’t. Something’s gotta give, and she squares the circle by simply asserting that reality is false. She’s not interested in intellectual pursuit when it conflicts with social justice, but she won’t admit that to herself, and so she solves her problem by pretending it doesn’t exist, and that the correct intellectual position is to not pursue the data beyond the point at which it conflicts with people’s strongly held feelings about right and wrong. If she was a scientist she’d have decoupled the questions of right and wrong from the questions of true and false.
But even if she was a scientist, she’d still have to worry about her career. If you want a career, there’s a lot of lying involved, and it’s best if you can trick yourself into believing the official fiction.
By this reckoning, I would be a postmodern contextual thinker. For some, the context is infinite and thus claims are necessarily indeterminate. I call that approach a “flat earth” view. I reckon the actual context to be infinite, but the PRACTICAL context to always be limited, making determinate claims possible. Given that whatever we are viewing/talking about is from a particular context, that context has horizons of relevance, depending on what we’re trying to do. Scientific thinking doesn’t (at least at its best) deny context, but narrowly defines the horizons for sake of the practical aim of the moment. Ceteris paribus clauses point at this.
If you take this far enough, you reach meta-rationality – the art of choosing the best context for approaching an issue rationally, depending on what you want to understand about the issue.
https://meaningness.com/metablog/bongard-meta-rationality
In my mind, the decouplers are more “productive”, but the contextualists (I prefer this to post-modernism, as post-modernism carries a lot of baggage) are more “correct”. I don’t think applying it to left/right is quite right.
To stay out of politics for a moment, consider computer science and the concept of leaky abstractions. Decouplers build abstractions on top of abstractions, and accomplish great things. But the abstractions do leak, and never quite work perfectly. They can often be exploited.
Hackers can be thought of as contextualists (or post-modern programmers). They find points where the abstractions leak, and exploit those points. The recent raft of Intel processor exploits are good examples. Steal information from high level abstractions by exploiting decisions much, much lower in the abstraction stack.
In science, decouplers are the theoretical scientists, and contextualists are the applied scientists and engineers. Both sides are very valuable.
In politics, perhaps the right-wing decouplers are the libertarians, and the right-wing contextualists are the social conservatives. Left-wing decouplers are the IDW, and left-wing contextualists are the social justice movement.
First, theoretical scientists, applied scientists, and engineers all strike me as decouplers. I also would replace social conservatives with the Alt-Right as the right-wing contextualist group.
More importantly, you claim that both decouplers and contextualists are “valuable”. However, if politics consisted solely of libertarians and the IDW, what problems do you think would be left unaddressable without adding the Alt-Right and social justice warriors to the mix? Conversely, which elements of current political dysfunction would you attribute to libertarians and the IDW rather than squabbling between the Alt-Right and SJWs?
Contextualists themselves wouldn’t necessarily argue that both types of thinking are valuable. Their argument is that no one can truly be a decoupler, and that decoupling is just a pretense for gaining an upperhand in an argument between contextualists. Saying both are valuable is like saying both honesty and corruption are valuable. Even corrupt people don’t think corruption is (socially) valuable. At most, they just doubt that anyone is truly honest.
IDW/libertarians have a bias for wanting he world to be one in which their philosophy is viable. However, empirical results keep coming up again this.
Why does the IDW keep losing to progressives? Why are there no libertarian states? They can’t answer the question. The alt right can provide answers. It’s reality and result oriented. The IDW/libertarians are like the marxists of old. They have a theory of humanity that would be beutifal if it were remotely true. But it isn’t.
Thanks for posting the link to your PK column. I thought it was really good.
Perhaps I am over-fitting, but it it strikes me that the “decoupling vs. contextualizing” dichotomy matches closely to the “systemizing vs. empathizing” distinction noted by Dr. Simon Baron-Cohen [1]. If you are trying to understand a system, you need to employ a high degree of decoupling, both in white-box analysis (i.e. looking inside the system at the various components that compose it) and in black-box analysis (i.e. looking at a system merely in terms of its inputs and outputs, and varying the inputs only in one dimension at a time to see how that affects the outputs).
Empathizing, or ” the drive to identify another person’s emotions and thoughts, and to respond to them with an appropriate emotion”, requires contextualizing because you need to process all of your observations of another person in their totality in order to provide the correct emotional response.
I wouldn’t be surprised if there was some correlation between the empathizing vs. systemizing axis and the left vs. right political axis, but I doubt that correlation is very strong (though I do think it is worth exploring why that is). From what I understand, it is the libertarian political alignment (independent of left vs. right) that most closely correlates with the systemizing personality. This line’s up with RohanV’s comment about libertarians/IDWers favoring decoupling.
What I don’t fully understand is that it seems that high-IQ (particularly high verbal IQ) empathizers seem to congregate on the left, and are over-represented among the ideological and political elite of the left. Granted, I imagine people of this profile tend towards the ideological and political elite of whatever faction they belong to. Part of my question is: why is there not (or does not appear to be) an even distribution of people with this profile across the different political factions?
[1] http://www.people.fas.harvard.edu/~ssiegel/papers/baron-cohen.html
Coupling seems like a weak axis.
Ezra sells a news site, he has a particular audience and couples his articles to audience as a matter of course.
Politics is about trying to motivate people and build coalitions; there’s no way to approach it that doesn’t involve appealing to emotions and group loyalties.
On the other hand, I was trained as a physical scientist. We certainly have biases and preconceptions. We also have a bunch of tools like controlled experiments and double-blinding that are specifically designed to make it as difficult as possible for our biases to influence our results. Good scientists are always asking themselves how they could be wrong and what data they should seek to dispel (or confirm) that doubt. But in politics this sort of detailed, careful, deliberately humble analysis seems unconfident; it doesn’t swing people’s ideas.
“…make it as difficult as possible for our biases to influence our results”
As I understand it, trying to minimize or remove biases is a decoupler concept. Contextualists claim that it’s impossible to remove biases so what’s important is to choose the *right* biases. In the IQ debate, Klein’s argument is that one *should* be biased in favor of finding no genetic component to IQ. That bias is better in terms of associations, implications, motivations, and identities. Since Klein is skeptical that one can remove bias in trying to answer the IQ question, he thinks that anyone claiming to try to do so should arouse suspicion that they might secretly favor the opposite bias, i.e., be biased in favor of finding a genetic component.
Yeah, “politics vs. science” is the same thing as “coupling vs. decoupling”. My point was that neither approach works well for the other sort of problem.
Where the “coupler” approach is to acknowledge your biases and give your favored answer the benefit of the doubt, the “decoupler” approach is to collect better data, taking extreme care to minimize the effects of possible biases, and (ideally) have the experiments replicated by multiple groups who are unlikely to share the same biases. If you’re doing it right, you should be able to converge on an answer.
On the other hand, the decoupler approach assumes that society wants to know the answer. When it comes to the genetic basis for IQ, this is manifestly false. 23andme could provide a robust answer in six months, tops, but that answer would likely upset the political apple cart.
Also, don’t think this is limited to liberal issues. All ideologies have truths they don’t want to deal with; libertarianism has all sorts of trouble coming up with a reasonable answer to global warming, and nobody has a good idea of what to do about the Federal debt.
Any ideology has to be simple enough for average people to understand it and emotionally appealing to humans. Reality has no such constraints. Attempts to impose any ideology on reality (even libertarianism) leave a remainder of troubles that can’t be dealt with with but can be ignored for a while. Eventually the remainder grows to the point where the ideology becomes unstable, it collapses, and somebody has to think up a new ideology to deal with what’s left.
For a more detailed discussion of the principle, see http://archdruidmirror.blogspot.com/2017/06/the-fifth-side-of-triangle.html
This reminds me of the idea of Conflict Theory vs Mistake Theory which was described over at Slate Star Codex a few months back…
Came across this debate about Originalism that strikes me as another example of decoupling vs. contextualism: [http://reason.com/reasontv/2018/06/22/randy-barnett-originalism-constitution].
Barnett’s originalism strikes me as decoupling: examine original public meaning of Constitutional text as an abstract exercise, decoupled from implications, motivations, etc. in current context. Living Constitutionalist Dorf claims that such clinical analysis is impossible, or at least is not done consistently by self-proclaimed originalists, so we might as well give up that pretense. I think he even says something like he has no problem with Barnett’s version of originalism but, if we accept it, then some will use that acceptance to give “old-style originalism”, which is different and which Dorf rejects, a “patina of legitimacy”. That’s pure contextualism: we can’t decouple the legitimacy of Barnett’s arguments from how others might misuse them illegitimately. That directly parallels Klein’s arguments about IQ: we can’t decouple the factual questions of IQ from the context of how others might “use” those facts.
Journalists have also been debating decoupling vs. contextualism over the last few decades. In the past, journalists strove for objectivity: report facts decoupled from what readers might do with those facts, i.e., report facts and “let readers decide”. More recently, contextualists claim that “objectivity” is impossible so journalists should drop that pretense. Instead, journalists should shape their reporting to provide readers with the “right” narrative so that the “other side” won’t “manipulate” fact-based reporting to its own advantage. That’s the contextualist approach of Klein and Dorf. Klein, of course, is part of the younger, post-objectivity generation of journalists.
What is the point of debating IQ separate from “using it”?
The reason IQ is so sensitive is because the justification for progressive programs fall apart if you accept IQ research. Ezra knows this, that’s why he goes nuclear.
It’s not just the justification for specific programs, it’s the entire dream of a racially-equal society. The racial divide in American politics is about who’s to blame for the poor performance of (many) blacks. Identitarians believe the failures are due to white racism, and their proposed remedy is comprehensive reform of white behavior and institutions. The Murry via Harris research favors the conclusion that blacks mostly fail from lack of talent, and that white people have little control over the matter.
I lean toward the heuristic that if three generations have tried and failed to fix a problem, the burden of proof rests on anyone proposing a new fix (if the fix is costly or troublesome).