Three Kinds of Contrarian
Abstract beliefs have two big causes: vibes and analysis. We vibe beliefs mostly via intuitively feeling out their associations with people, other beliefs, and our personal status. (Music, art, eloquence, and status often influence this a lot.) We analyze such beliefs by more consciously and explicitly comparing our beliefs logically to concrete analysis of relevant data and established theory.
Most people form and change most of their abstract beliefs via vibing. (In far mode.) But there are experts in the world who have learned about specific relevant data and theory, and have learned how to apply those to estimate nearby abstract beliefs. Those calculated abstract beliefs are often at odds with the most popular vibed versions.
Most public talk on abstract beliefs is vibes. You can tell this by comparing how fast, fluid, and vague is such talk, relative to how slowly, carefully, and precisely experts must proceed to figure out the logical implications or data and theory, and to communicate that to other experts.
Most experts allow their expert knowledge to change some of what would otherwise be their vibed beliefs. However, they usually try to minimize the impact of their expert knowledge, which usually only applies to narrow areas, on their network of vibed beliefs, which covers a much wider range of topics. Often they emphasize the limits of their expert tools, and invoke piecemeal “common sense judgments” to protect their vibed beliefs from being changed by expert knowledge.
Most experts only have a few related areas of expertise. But some “polymaths” (like me) work to acquire expertise across a much wider range of topics. They then have more chances for expert knowledge to overturn vibed beliefs. For them there is a conflict not only between their network of vibed beliefs and expert knowledge, but also between expert knowledge in different areas.
Experts in different areas often claim that their expertise is relevant to the same vibed belief topics, with conflicting implications for those topics. Polymaths must then decide on what basis to choose. Do they go with what their vibes say, with the expert area to which they feel most allegiance, or do they try to use discipline-neutral principles of evaluation to decide which area’s data and theory speak most strongly on the topic?
The above framing allows me to explain and argue for the virtue of two key ways that I am contrarian. First, I prefer to accept the results of the usual specific expert data and theory when that conflicts with vibes, even when most adjacent experts go with vibes. For example, most folks with adjacent expertise estimate the health value of medicine to be high, agreeing with the usual vibes, while it seems to me that our best specific theory and evidence says otherwise. A similar case happens re the value of democracy.
Second, I usually presume that most areas of expertise have substantial insight into topics when thousands have studied an area for decades. So I try to accept as many as possible of the usual specific data and theory driven conclusions of all areas of expertise. When they conflict, I try to use discipline-neutral principles of evaluation to decide which area’s data and theory speak most strongly on the topic.
For example, though I’m an econ professor, I don’t just assume that economists are always right. This is “contrarian” in the sense that few experts are polymaths, and most of those stay loyal to a single home area of expertize.
The third way to be a contrarian is to just embrace contrary vibes. “That’s just your opinion, man. You guys are so full of yourselves. You can’t tell me what to think.” Sometimes that leads to wisdom, but more often to error.
Of course you could just not be any sort of contrarian, and embrace the most common prestigious vibes. That’s the easiest approach, and it beats contrarian vibes on accuracy. But I think that on average embracing specific expert data and theory is the most accurate approach. Even when that’s not what most experts themselves do. Yes, they plausibly achieve social ends from that behavior which I forgo.
Then goes back to vibes again, do I trust this polymath or that, maybe gpt is the best polymath, this feels like some form of appeal to authority.
“Maybe this is all straw-manning, all of this is taken out of context, and the only place that Robin says his true opinion is in his book. But in that case I feel like this is a pretty extreme failure of communication that’s not entirely my fault. Also, other people seem to interpret it the same way I do:“
Based on the link above to Robin’s “debate” with Scott Alexander, I went back and reviewed the whole thing again, possibly in more detail than I did the first time.
I think I am mostly on Team Robin.
But the one place I agree with Scott is on the quoted bit above.
If Robin stuck to and repeated the “we spend too much taxpayer money on healthcare, including how we subsidize it in the tax code” and “we should reduce by half medical spending done on behalf of others” given the highly questionable returns on marginal spending, and “consider whether you would do xyz if you had to pay for it all yourself”, he would be on MUCH firmer ground.
What I found most surprising reviewing the debate was that no one brought up the alternative uses of the money, nor that marginal value differs enormously for each person - in no small part based on their income and assets! [Apologies to Robin if he did bring it up in one small place and I missed it.]
The fact that upper middle class Scott would value and pay for so many of the examples is kinda irrelevant. The better marginal point is whether - even assuming a generally accurate understanding of the risks and rewards - any given beneficiary of government or mandated insurance company largesse for a medical intervention would prefer the cash to the intervention.
Paternalistic leftists, of course, would say people don’t know what’s good for them - or can’t know, in the case of medicine. Re: the first part, I just roll my eyes; re: the 2nd part, this is where Robin’s points about ultimate outcome differences are entirely relevant.
Re: public policy the enormous total spending by taxpayers for limited / questionable benefits is the key point.
For individuals making decisions, their marginal utility of the medical intervention versus the cash spent alternative is the key point.
I put this on Robin more as a) the economist and b) the one pushing the more “controversial” view.
But it’s also on polymath Scott who should know better about these two basic points of public policy and economics, despite Robin’s failings at communicating this clearly.